SPB Git
spboucher.ai

spb/wp44_uqo

Public

WP44: semantic scripts read-only (DuckDB lock)

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Simon-Pierre Boucher committed (Nov 4, 2026) parent 24c0632 Browse files

50 changed files +6,165 −425

50 changed files
Unified Split
modified bibliometrics/04_diversity_cagr.py +1 −1
@@ -12,7 +12,7 @@ import argparse, os, json, numpy as np, pandas as pd
12 def main(): 12 def main():
13 ap = argparse.ArgumentParser(); ap.add_argument("--root", required=True); ap.add_argument("--B", type=int, default=500); a = ap.parse_args(); R = a.root 13 ap = argparse.ArgumentParser(); ap.add_argument("--root", required=True); ap.add_argument("--B", type=int, default=500); a = ap.parse_args(); R = a.root
14 out = f"{R}/results/bibliometrics"; cc = pd.read_csv(f"{R}/results/graph/category_counts_yearly.csv"); cc = cc[cc.year <= 2025] 14 out = f"{R}/results/bibliometrics"; cc = pd.read_csv(f"{R}/results/graph/category_counts_yearly.csv"); cc = cc[cc.year <= 2025]
15− piv = cc.pivot_table(index="year", columns="category_id", values="n_primary", fill_value=0); cats = list(piv.columns) 15+ piv = cc.pivot_table(index="year", columns="category_id", values="n_primary", aggfunc="sum", fill_value=0); cats = list(piv.columns) # aggfunc="sum": the file has one row per (year, period); the default mean halved the 2018-2022 levels (fixed 2026-11-04, Part IX writer)
16 rows = [] 16 rows = []
17 for c in cats: 17 for c in cats:
18 b0, b1 = piv.loc[2018, c], piv.loc[2022, c]; p0, p1 = piv.loc[2023, c], piv.loc[2025, c] 18 b0, b1 = piv.loc[2018, c], piv.loc[2022, c]; p0, p1 = piv.loc[2023, c], piv.loc[2025, c]
modified embeddings/semantic_atlas.py +2 −3
@@ -24,9 +24,8 @@ def main():
24 ap = argparse.ArgumentParser(); ap.add_argument("--root", required=True); ap.add_argument("--model", default="allenai/specter2_base"); ap.add_argument("--n", type=int, default=200000) 24 ap = argparse.ArgumentParser(); ap.add_argument("--root", required=True); ap.add_argument("--model", default="allenai/specter2_base"); ap.add_argument("--n", type=int, default=200000)
25 ap.add_argument("--seed", type=int, default=44); a = ap.parse_args(); R = a.root; tag = a.model.replace("/", "__"); out = f"{R}/results/semantic"; os.makedirs(out, exist_ok=True) 25 ap.add_argument("--seed", type=int, default=44); a = ap.parse_args(); R = a.root; tag = a.model.replace("/", "__"); out = f"{R}/results/semantic"; os.makedirs(out, exist_ok=True)
26 E, ids = load_embeddings(R, tag); print("embeddings", E.shape) 26 E, ids = load_embeddings(R, tag); print("embeddings", E.shape)
27− con = duckdb.connect(f"{R}/data/processed/wp44.duckdb") 27+ con = duckdb.connect(f"{R}/data/processed/wp44.duckdb", read_only=True) # read-only: other processes may hold the database
28− con.execute(f"CREATE OR REPLACE VIEW work_primary AS SELECT * FROM read_parquet('{R}/data/processed/classification/primary/*.parquet')") 28+ meta = con.execute(f"SELECT c.id AS work_id, c.publication_year AS year, c.period, c.first_month, p.primary_category FROM corpus_works c LEFT JOIN read_parquet('{R}/data/processed/classification/primary/*.parquet') p ON p.work_id = c.id WHERE c.eligible").df().set_index("work_id")
29− meta = con.execute("SELECT c.id AS work_id, c.publication_year AS year, c.period, c.first_month, p.primary_category FROM corpus_works c LEFT JOIN work_primary p ON p.work_id = c.id WHERE c.eligible").df().set_index("work_id")
30 pos = {w: i for i, w in enumerate(ids)}; meta = meta[meta.index.isin(pos)]; meta["row"] = [pos[w] for w in meta.index] 29 pos = {w: i for i, w in enumerate(ids)}; meta = meta[meta.index.isin(pos)]; meta["row"] = [pos[w] for w in meta.index]
31 rng = np.random.default_rng(a.seed) 30 rng = np.random.default_rng(a.seed)
32 # (4) category centroids per year + drift + similarity 31 # (4) category centroids per year + drift + similarity
modified figures/fig_category_growth.pdf +0 −0

Binary file not shown.

modified figures/fig_category_growth.png +0 −0

Binary file not shown.

modified figures/make_figures.py +1 −1
@@ -225,7 +225,7 @@ def fig_category_growth(R, out):
225 if not os.path.exists(p): 225 if not os.path.exists(p):
226 return 226 return
227 names, areas = _cat_names(R); d = pd.read_csv(p); d = d[(d.year >= 2018) & (d.year <= 2025)] 227 names, areas = _cat_names(R); d = pd.read_csv(p); d = d[(d.year >= 2018) & (d.year <= 2025)]
228− piv = d.pivot_table(index="year", columns="category_id", values="n_primary", fill_value=0); base = piv.loc[2018] 228+ piv = d.pivot_table(index="year", columns="category_id", values="n_primary", aggfunc="sum", fill_value=0); base = piv.loc[2018] # aggfunc="sum" over the (year, period) rows (fixed 2026-11-04)
229 idx = piv / base.replace(0, np.nan) * 100; cats = sorted(idx.columns, key=lambda c: -idx.loc[2025, c] if 2025 in idx.index else 0) 229 idx = piv / base.replace(0, np.nan) * 100; cats = sorted(idx.columns, key=lambda c: -idx.loc[2025, c] if 2025 in idx.index else 0)
230 fig, axes = plt.subplots(1, 2, figsize=(7.4, 3.6), sharey=False) 230 fig, axes = plt.subplots(1, 2, figsize=(7.4, 3.6), sharey=False)
231 for ax, sel, title in [(axes[0], cats[:8], "Eight fastest-growing categories"), (axes[1], cats[-8:], "Eight slowest-growing categories")]: 231 for ax, sel, title in [(axes[0], cats[:8], "Eight fastest-growing categories"), (axes[1], cats[-8:], "Eight slowest-growing categories")]:
modified knowledge_graph/04_semantic_edges.py +3 −4
@@ -33,11 +33,10 @@ def main():
33 print(f"{s+len(qi)}/{len(q_idx)} queried, {len(rows)} edges, {time.time()-t0:.0f}s", flush=True) 33 print(f"{s+len(qi)}/{len(q_idx)} queried, {len(rows)} edges, {time.time()-t0:.0f}s", flush=True)
34 df = pd.DataFrame(rows, columns=["work_id", "neighbour_id", "cosine", "rank"]); df.to_parquet(f"{out}/semantic_edges_{tag}.parquet") 34 df = pd.DataFrame(rows, columns=["work_id", "neighbour_id", "cosine", "rank"]); df.to_parquet(f"{out}/semantic_edges_{tag}.parquet")
35 json.dump({"model": a.model, "k": a.k, "min_sim": a.min_sim, "n_vectors": int(n), "n_edges": int(len(df)), "index": "HNSW32 efC80 efS64 inner-product", "seconds": round(time.time() - t0)}, open(f"{out}/semantic_edges_{tag}.json", "w"), indent=1) 35 json.dump({"model": a.model, "k": a.k, "min_sim": a.min_sim, "n_vectors": int(n), "n_edges": int(len(df)), "index": "HNSW32 efC80 efS64 inner-product", "seconds": round(time.time() - t0)}, open(f"{out}/semantic_edges_{tag}.json", "w"), indent=1)
36− con = duckdb.connect(f"{R}/data/processed/wp44.duckdb") 36+ con = duckdb.connect(f"{R}/data/processed/wp44.duckdb", read_only=True)
37− con.execute(f"CREATE OR REPLACE VIEW work_primary AS SELECT * FROM read_parquet('{R}/data/processed/classification/primary/*.parquet')") 37+ P = f"read_parquet('{R}/data/processed/classification/primary/*.parquet')"
38− con.execute(f"CREATE OR REPLACE VIEW sem AS SELECT * FROM read_parquet('{out}/semantic_edges_{tag}.parquet')")
39 con.execute(f"""COPY (SELECT c.publication_year AS year, pa.primary_category, count(*) n_edges, round(avg((pa.primary_category <> pb.primary_category)::int),4) share_cross_category 38 con.execute(f"""COPY (SELECT c.publication_year AS year, pa.primary_category, count(*) n_edges, round(avg((pa.primary_category <> pb.primary_category)::int),4) share_cross_category
40− FROM sem s JOIN work_primary pa ON pa.work_id = s.work_id JOIN work_primary pb ON pb.work_id = s.neighbour_id JOIN corpus_works c ON c.id = s.work_id 39+ FROM read_parquet('{out}/semantic_edges_{tag}.parquet') s JOIN {P} pa ON pa.work_id = s.work_id JOIN {P} pb ON pb.work_id = s.neighbour_id JOIN corpus_works c ON c.id = s.work_id
41 GROUP BY 1,2 ORDER BY 1,2) TO '{R}/results/graph/semantic_crossing_yearly.csv' (HEADER)""") 40 GROUP BY 1,2 ORDER BY 1,2) TO '{R}/results/graph/semantic_crossing_yearly.csv' (HEADER)""")
42 print("done", len(df), "edges") 41 print("done", len(df), "edges")
43 42
added literature_synthesis/agent_E/contributions.csv +235 −0
@@ -0,0 +1,235 @@
1+openalex_id,title,year,type,primary_category,doi,arxiv_id,c36,ncs36,contribution_type,description,judged_from,data_quality_flag
2+7172491899,Toward effective Text-to-MongoDB query translation,2026,article,llm,10.26555/ijain.v12i2.2360,2005.11401,3026.0,3367.3,engineering,"Text-to-MongoDB query translation (GAT + RAG + T5); record carries arXiv 2005.11401 (RAG, Lewis et al.) and its citations",abstract,identifier_mismatch
3+7154011589,Modelos pre-treinados de LLM e questoes eticas face aos direitos autorais,2026,article,llm,10.55905/rdelosv19.n79-020,2005.14165,2847.0,3168.1,empirical,Portuguese-language essay on LLM pre-training data and copyright; record carries arXiv 2005.14165 (GPT-3) and its citations,abstract,identifier_mismatch
4+4362515116,A Survey of Large Language Models,2026,article,llm,10.1007/s11704-026-60308-3,2303.18223,1549.0,1723.7,survey,"Survey of LLM techniques (pre-training, adaptation, utilisation, evaluation); journal version of the 2023 arXiv survey",abstract,ok
5+4288253152,Multilingual Classifier for Automatic Dewey Decimal Classification trained on Open Access Linguistics Abstracts,2024,dissertation,llm,10.5281/zenodo.12935200,,364.0,3102.1,engineering,Dissertation: fine-tuned multilingual sentence-transformer for Dewey Decimal classification; NCS36 implausible for a dissertation,abstract,normalisation_artefact
6+4322718191,LLaMA: Open and Efficient Foundation Language Models,2023,preprint,llm,10.48550/arxiv.2302.13971,2302.13971,3629.0,1206.5,engineering,LLaMA: 7B-65B open-weight foundation language models trained on public data; the abstract reports LLaMA-13B outperforms GPT-3 on most benchmarks,abstract,ok
7+4323655724,ChatGPT for good? On opportunities and challenges of large language models for education,2023,article,llm,10.1016/j.lindif.2023.102274,,4227.0,990.7,survey,Position paper on opportunities and challenges of LLMs for education (student and teacher perspectives),abstract,ok
8+4367369313,"Prompt Engineering for ChatGPT: A Quick Guide to Techniques, Tips, and Best Practices",2026,article,llm,10.18260/jet.43.1.5,,261.0,290.4,survey,Practitioner guide to prompt engineering for ChatGPT,abstract,ok
9+4226278401,Discernment and Social Learning as a Companion Training Layer,2022,preprint,llm,10.48550/arxiv.2203.02155,2203.02155,3077.0,894.2,method,Proposal of a 'companion curriculum' for LM training; record carries arXiv 2203.02155 (InstructGPT) and its citations,abstract,identifier_mismatch
10+4396531244,The Fusion of Multilingual Semantic Search and Large Language Models: A New Paradigm for Enhanced Topic Exploration and Contextual Search,2024,dissertation,llm,10.22215/etd/2024-15938,,20.0,778.8,engineering,Dissertation: multilingual semantic search + LLMs for news retrieval; NCS36 implausible for c36=20,abstract,normalisation_artefact
11+4221143046,"BNAI, NO-TOKEN, and MIND-UNITY: Pillars of a Systemic Revolution in Artificial Intelligence",2022,preprint,llm,10.48550/arxiv.2201.11903,2201.11903,2465.0,716.4,theory,Essay on 'theatrical compliance' failure mode; record carries arXiv 2201.11903 (chain-of-thought) and its citations,abstract,identifier_mismatch
12+4327810158,GPT-4 Technical Report,2023,preprint,llm,10.4230/lipics.cosit.2024.11,2303.08774,1968.0,654.3,engineering,Title = GPT-4 Technical Report (arXiv 2303.08774) but the stored abstract describes an unrelated 'MFOUR Vibe Framework',abstract,abstract_mismatch
13+7135096569,Navigating the Jagged Technological Frontier: Field Experimental Evidence of the Effects of Artificial Intelligence on Knowledge Worker Prod,2026,article,llm,10.1287/orsc.2025.21838,,183.0,203.6,empirical,Preregistered field experiment (758 BCG consultants) on the uneven ('jagged') effect of GPT-4 assistance on knowledge-work tasks,abstract,ok
14+4414281281,DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning,2025,article,llm,10.1038/s41586-025-09422-z,,1099.0,374.4,method,DeepSeek-R1: reasoning incentivised by pure reinforcement learning without human-annotated reasoning traces (Nature version),abstract,ok
15+4389437528,Education in the Era of Generative Artificial Intelligence (AI): Understanding the Potential Benefits of ChatGPT in Promoting Teaching and L,2023,article,llm,10.61969/jai.1337500,,2361.0,553.4,survey,Review of potential benefits of ChatGPT for teaching and learning,abstract,ok
16+4292779060,"Aion Framework: Dimensional Emergence of AI Consciousness, Observer-Induced Collapse, and Cosmological Portal Dynamics",2023,article,llm,10.4230/lipics.giscience.2023.43,,13333.0,534.7,theory,"Speculative 'Aion Framework' on AI consciousness; c36=13,333 cannot belong to this record",abstract,identifier_mismatch
17+4360836968,Sparks of Artificial General Intelligence: Early experiments with GPT-4,2023,preprint,llm,10.48550/arxiv.2303.12712,2303.12712,1420.0,472.1,empirical,'Sparks of AGI': qualitative experiments on an early GPT-4; contested interpretation; preprint,abstract,ok
18+4406800520,The rise and potential of large language model based agents: a survey,2025,article,llm,10.1007/s11432-024-4222-0,2309.07864,914.0,311.4,survey,Survey of LLM-based agents (brain/perception/action framework); journal version of arXiv 2309.07864,metadata_only,no_abstract
19+4385223296,"Open AI in Education, the Responsible and Ethical Use of ChatGPT Towards Lifelong Learning",2023,book-chapter,llm,10.1007/978-3-031-37776-1_17,,612.0,421.3,survey,Book chapter on responsible use of ChatGPT in education,metadata_only,no_abstract
20+4287689466,From Prediction to Precision: Leveraging LLMs for Equitable and Data-Driven Writing Placement in Developmental Education,2025,conference-paper,llm,10.4230/oasics.slate.2025.1,,722.0,548.6,empirical,"LLM essay classification for writing placement (450 essays, four models); c36=722 implausible for an OASIcs paper",abstract,identifier_mismatch
21+4366420437,What Is the Impact of ChatGPT on Education? A Rapid Review of the Literature,2023,article,llm,10.3390/educsci13040410,,1541.0,361.2,survey,Rapid review of 50 early articles on ChatGPT in education (Dec 2022-Feb 2023),abstract,ok
22+4317910584,ChatGPT: Bullshit spewer or the end of traditional assessments in higher education?,2023,article,llm,10.37074/jalt.2023.6.1.9,,1530.0,358.6,survey,Commentary on ChatGPT and assessment in higher education,abstract,ok
23+4313564799,ChatGPT User Experience: Implications for Education,2023,preprint,llm,10.2139/ssrn.4312418,,811.0,335.9,empirical,SSRN preprint on ChatGPT user experience in education,metadata_only,no_abstract
24+4382618722,Chatlaw: A Multi-Agent Legal Assistant based on a Role-Aligned Mixture-of-Experts Architecture,2026,article,llm,10.1016/j.fmre.2026.03.026,2306.16092,94.0,98.9,engineering,Chatlaw: multi-agent legal assistant with role-aligned mixture-of-experts (Chinese law),abstract,ok
25+4384464487,"Students voices on generative AI: perceptions, benefits, and challenges in higher education",2023,article,llm,10.1186/s41239-023-00411-8,2305.00290,1398.0,327.7,empirical,Survey of 399 Hong Kong students on generative AI perceptions,abstract,ok
26+4392773210,"Embracing the future of Artificial Intelligence in the classroom: the relevance of AI literacy, prompt engineering, and critical thinking in",2024,article,llm,10.1186/s41239-024-00448-3,,1085.0,326.7,survey,"Discussion of AI literacy, prompt engineering and critical thinking in classrooms",abstract,ok
27+4387355843,TimeLLM: Time Series Forecasting by Reprogramming Large Language Models,2026,book-chapter,foundation_models,10.1201/9781003616719-7,2310.01728,197.0,188.6,method,Time-LLM: reprogramming LLMs for time-series forecasting (book-chapter record of the ICLR 2024 paper),abstract,ok
28+4392947521,Towards a general-purpose foundation model for computational pathology,2024,article,foundation_models,10.1038/s41591-024-02857-3,,1505.0,186.9,method,UNI: general-purpose self-supervised foundation model for computational pathology,metadata_only,no_abstract
29+7124890304,Semi-supervised cross-domain fault diagnosis via contrastive pre-training and annotation-efficient alignment strategy,2026,article,foundation_models,10.1016/j.jii.2026.101076,,42.0,46.5,method,Semi-supervised cross-domain fault diagnosis via contrastive pre-training,metadata_only,no_abstract
30+4410551352,A foundation model for the Earth system,2025,article,foundation_models,10.1038/s41586-025-09005-y,2405.13063,249.0,69.7,method,Aurora: foundation model for the Earth system trained on >1M hours of geophysical data; abstract reports outperforming operational forecasts,abstract,ok
31+4322832366,TimeMAE: Self-Supervised Representations of Time Series with Decoupled Masked Autoencoders,2026,conference-paper,foundation_models,10.1145/3773966.3778007,2303.00320,35.0,126.3,method,TimeMAE: decoupled masked autoencoders for time-series representation,abstract,ok
32+7133233472,QLoRA: Efficient Finetuning of Quantized LLMs,2023,conference-paper,foundation_models,10.52202/075280-0441,2305.14314,532.0,122.7,method,QLoRA: 4-bit quantised base + low-rank adapters; 65B model finetuned on one 48 GB GPU,abstract,ok
33+2990704537,SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systems,2019,article,foundation_models,,,668.0,101.6,benchmark,SuperGLUE: harder successor to GLUE after models exceeded non-expert human level,abstract,ok
34+7125271461,ChemBERTa-3: an open source training framework for chemical foundation models,2026,article,foundation_models,10.1039/d5dd00348b,,25.0,70.6,engineering,ChemBERTa-3: open training framework for chemical foundation models,abstract,ok
35+7119541039,On the foundations of Earth foundation models,2026,article,foundation_models,10.1038/s43247-025-03127-x,,24.0,26.7,theory,Position paper defining eleven desirable features of Earth foundation models,abstract,ok
36+4393138529,Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey,2024,preprint,foundation_models,10.48550/arxiv.2403.14608,2403.14608,95.0,82.9,survey,Survey of parameter-efficient fine-tuning (PEFT) for large models; preprint,abstract,ok
37+4390690043,Data-Centric Foundation Models in Computational Healthcare: A Survey,2026,article,foundation_models,10.1145/3800677,2401.02458,22.0,10.0,survey,Survey of data-centric foundation-model approaches in computational healthcare,abstract,ok
38+7128029160,A generalizable foundation model for analysis of human brain MRI,2026,article,foundation_models,10.1038/s41593-026-02202-6,,20.0,22.3,method,BrainIAC: self-supervised foundation model for brain MRI,abstract,ok
39+4402220065,A pathology foundation model for cancer diagnosis and prognosis prediction,2024,article,foundation_models,10.1038/s41586-024-07894-z,,562.0,69.8,method,Pathology foundation model for cancer diagnosis and prognosis (Nature 2024),metadata_only,no_abstract
40+7118170538,Applications of Artificial Intelligence in Fisheries: From Data to Decisions,2026,article,foundation_models,10.3390/bdcc10010019,,17.0,17.9,survey,Review of AI applications in fisheries,abstract,ok
41+4411932262,A foundation model to predict and capture human cognition,2025,article,foundation_models,10.1038/s41586-025-09215-4,,110.0,26.9,method,Centaur: language model fine-tuned on Psych-101 (10M choices) predicting human behaviour across experiments,abstract,ok
42+4411099958,A multimodal vision foundation model for clinical dermatology,2025,article,foundation_models,10.1038/s41591-025-03747-y,2410.15038,108.0,18.4,method,"PanDerm: multimodal dermatology foundation model pretrained on 2M images, 28 benchmarks",abstract,ok
43+4361229539,LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention,2023,preprint,foundation_models,10.48550/arxiv.2303.16199,2303.16199,161.0,53.5,method,LLaMA-Adapter: 1.2M-parameter adaption prompts with zero-init attention for instruction following; preprint,abstract,ok
44+4407135816,"Foundation Models in Radiology: What, How, Why, and Why Not",2025,article,foundation_models,10.1148/radiol.240597,2411.18730,103.0,6.9,survey,Narrative review of foundation models for radiology,abstract,ok
45+4385948838,A visuallanguage foundation model for pathology image analysis using medical Twitter,2023,article,foundation_models,10.1038/s41591-023-02504-3,,471.0,49.1,method,PLIP: visual-language pathology foundation model trained on medical Twitter data,metadata_only,no_abstract
46+4406643700,Towards the next generation of Geospatial Artificial Intelligence,2025,article,foundation_models,10.1016/j.jag.2025.104368,,91.0,19.1,survey,Quantitative literature analysis and agenda for geospatial AI,abstract,ok
47+4406472599,Learning from models beyond fine-tuning,2025,article,foundation_models,10.1038/s42256-024-00961-0,2310.08184,86.0,29.3,survey,Perspective on 'learning from models' beyond fine-tuning,metadata_only,no_abstract
48+4393153153,TimesURL: Self-Supervised Contrastive Learning for Universal Time Series Representation Learning,2024,conference-paper,foundation_models,10.1609/aaai.v38i12.29299,2312.15709,113.0,41.8,method,TimesURL: self-supervised contrastive universal time-series representations,abstract,ok
49+4402727669,mPLUG-OwI2: Revolutionizing Multi-modal Large Language Model with Modality Collaboration,2024,conference-paper,foundation_models,10.1109/cvpr52733.2024.01239,,184.0,38.6,method,mPLUG-Owl2: modality-collaborative multimodal LLM with shared modules,abstract,ok
50+4401397240,A generalist visionlanguage foundation model for diverse biomedical tasks,2024,article,foundation_models,10.1038/s41591-024-03185-2,2305.17100,325.0,38.5,method,BiomedGPT: generalist vision-language foundation model for biomedical tasks,metadata_only,no_abstract
51+4385285835,Scaling Self-Supervised Learning for Histopathology with Masked Image Modeling,2023,preprint,foundation_models,10.1101/2023.07.21.23292757,,107.0,35.6,method,Masked image modelling at scale for histopathology; medRxiv preprint,abstract,ok
52+4318718936,BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models,2023,preprint,multimodal_ai,10.48550/arxiv.2301.12597,2301.12597,875.0,273.1,method,BLIP-2: Q-Former bridging frozen image encoders and frozen LLMs; preprint,abstract,ok
53+4393148714,T2I-Adapter: Learning Adapters to Dig Out More Controllable Ability for Text-to-Image Diffusion Models,2024,conference-paper,multimodal_ai,10.1609/aaai.v38i5.28226,2302.08453,923.0,193.7,method,T2I-Adapter: lightweight adapters adding structural/colour control to text-to-image diffusion,abstract,ok
54+4396599905,Multimodal fusion on low-quality data: A comprehensive survey,2026,article,multimodal_ai,10.1016/j.inffus.2026.104437,2404.18947,48.0,128.7,survey,"Survey of multimodal fusion under low-quality data (noise, missing modalities, imbalance, dynamics)",abstract,ok
55+4396952261,Multimodal Deep Learning,2024,book-chapter,multimodal_ai,10.1007/978-3-031-53092-0_10,,189.0,147.2,survey,Textbook on multimodal deep learning,metadata_only,no_abstract
56+7133250524,LAION-5B: An Open Large-Scale Dataset for Training Next Generation Image-Text Models,2022,conference-paper,multimodal_ai,10.52202/068431-1833,,828.0,111.9,benchmark,LAION-5B: open 5.85B image-text pair dataset for training CLIP-style models,metadata_only,no_abstract
57+4392947532,A visual-language foundation model for computational pathology,2024,article,multimodal_ai,10.1038/s41591-024-02856-4,,938.0,111.1,method,CONCH: visual-language foundation model for computational pathology,metadata_only,no_abstract
58+7138474371,OpenDriveVLA: Towards End-to-end Autonomous Driving with Large Vision Language Action Model,2026,conference-paper,multimodal_ai,10.1609/aaai.v40i16.38386,,31.0,159.4,method,OpenDriveVLA: vision-language-action model for end-to-end driving on open LLMs,abstract,ok
59+7127319731,Med-R1: Reinforcement Learning for Generalizable Medical Reasoning in Vision-Language Models,2026,article,multimodal_ai,10.1109/tmi.2026.3661001,,29.0,77.7,method,Med-R1: RL-enhanced vision-language model for medical reasoning,abstract,ok
60+2619383789,Multimodal Machine Learning: A Survey and Taxonomy,2018,article,multimodal_ai,10.1109/tpami.2018.2798607,,587.0,103.1,survey,"Taxonomy of multimodal machine learning (representation, translation, alignment, fusion, co-learning)",abstract,ok
61+7148475777,Capabilities of GPT5 on multimodal medical reasoning,2026,conference-paper,multimodal_ai,10.1117/12.3086794,2508.08224,26.0,73.2,empirical,Zero-shot evaluation of GPT-5 on multimodal medical QA/VQA,abstract,ok
62+7160627408,Generative Artificial Intelligence for Cross-Domain Knowledge Synthesis and Intelligent Automation,2026,conference-paper,multimodal_ai,10.1109/csnt69054.2026.11502110,,23.0,193.4,method,Generative cross-domain knowledge synthesis framework (conference paper),abstract,ok
63+2995460200,VL-BERT: Pre-training of Generic Visual-Linguistic Representations,2020,conference-paper,multimodal_ai,,,670.0,81.1,method,VL-BERT: single-stream Transformer pre-trained on image regions and words,abstract,ok
64+4415913530,A multimodal whole-slide foundation model for pathology,2025,article,multimodal_ai,10.1038/s41591-025-03982-3,,157.0,53.5,method,"TITAN: multimodal whole-slide pathology foundation model (335,645 WSIs)",abstract,ok
65+4289785095,An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversion,2022,preprint,multimodal_ai,10.48550/arxiv.2208.01618,2208.01618,366.0,79.4,method,Textual inversion: personalising text-to-image diffusion with learned pseudo-words from 3-5 images; preprint,abstract,ok
66+7125233009,SemCap: Sentiment-aware semantic captioning for multimodal aspect-based sentiment analysis,2026,article,multimodal_ai,10.1007/s44443-026-00472-5,,22.0,24.5,method,SemCap: sentiment-aware captions as visual proxy for multimodal aspect-based sentiment analysis,abstract,ok
67+7143512276,Multimodal Machine Learning Approaches in Predictive Healthcare Analytics: A Comprehensive Survey,2026,article,multimodal_ai,10.1007/s11831-026-10560-4,,21.0,23.4,survey,Survey of multimodal ML in predictive healthcare analytics,abstract,ok
68+4413447271,Towards generalist foundation model for radiology by leveraging web-scale 2D&3D medical data,2025,article,multimodal_ai,10.1038/s41467-025-62385-7,2308.02463,142.0,46.4,method,"RadFM: generalist radiology foundation model on 2D/3D data (MedMD, 13M images)",abstract,ok
69+7131830459,A Survey on Multimodal Recommender Systems: Recent Advances and Future Directions,2026,article,multimodal_ai,10.1109/tmm.2026.3668620,2502.15711,20.0,142.1,survey,Survey of multimodal recommender systems,abstract,ok
70+7206206818,Lingshu: Generalist Foundation Model for Unified Multimodal Medical Understanding and Reasoning,2026,article,multimodal_ai,10.1109/tpami.2026.3730310,,18.0,48.2,method,Lingshu: generalist medical multimodal LLM,abstract,ok
71+7133220561,InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning,2023,conference-paper,multimodal_ai,10.52202/075280-2142,,454.0,63.4,method,InstructBLIP: instruction tuning of vision-language models,metadata_only,no_abstract
72+7164501397,Multimodal foundation models exploit text to make medical image predictions,2026,article,multimodal_ai,10.1038/s41467-026-74207-5,2311.05591,17.0,45.6,empirical,"Evaluation of 8 multimodal foundation models on 1,090 medical cases: image predictions largely driven by text",abstract,ok
73+7165561001,"M3-HGMA: A Multi-modal, Multi-agent Hierarchical Graph-Mamba Architecture for Explainable and Adaptive Clinical Intelligence",2026,conference-paper,multimodal_ai,10.1007/978-3-032-29501-9_39,,17.0,143.0,method,Hierarchical graph-Mamba multi-agent clinical architecture,metadata_only,no_abstract
74+7138470740,Cross-modal Proxy Evolving for OOD Detection with Vision-Language Models,2026,conference-paper,multimodal_ai,10.1609/aaai.v40i18.38608,,17.0,130.5,method,CoEvo: training-free test-time OOD detection with vision-language models,abstract,ok
75+7138363754,AdaptCLIP: Adapting CLIP for Universal Visual Anomaly Detection,2026,conference-paper,multimodal_ai,10.1609/aaai.v40i6.42404,,17.0,130.5,method,AdaptCLIP: adapting CLIP for universal visual anomaly detection,abstract,ok
76+4410882873,Navigating the landscape of multimodal AI in medicine: A scoping review on technical challenges and clinical applications,2025,article,multimodal_ai,10.1016/j.media.2025.103621,2411.03782,119.0,7.9,survey,Scoping review of 432 multimodal medical AI papers (2018-2024),abstract,ok
77+4400228804,The Conglomeration of Everything: Shifting Animation Production Practices due to Artificial Intelligence,2024,dissertation,generative_ai,10.46569/df65vg26z,,2.0,540.0,empirical,Dissertation on generative AI in animation production; c36=2 with NCS36=540 is a normalisation artefact,abstract,normalisation_artefact
78+7124262017,OECD Digital Education Outlook 2026,2026,book,generative_ai,10.1787/062a7394-en,,87.0,783.7,survey,OECD Digital Education Outlook 2026 (policy report on generative AI in education),abstract,ok
79+4406472463,A generative model for inorganic materials design,2025,article,generative_ai,10.1038/s41586-025-08628-5,,498.0,77.1,method,MatterGen: diffusion model generating stable inorganic materials under property constraints,abstract,ok
80+7123342962,Structure-Preserving Frequency-Regularized Text-Guided Optimal Transport for Unpaired Rain Streaks and Raindrops Removal,2026,article,generative_ai,10.1109/tmm.2026.3651132,,66.0,176.9,method,Text-guided optimal transport for unpaired rain removal,abstract,ok
81+4383312437,A comprehensive AI policy education framework for university teaching and learning,2023,article,generative_ai,10.1186/s41239-023-00408-3,2305.00280,988.0,231.6,empirical,AI Ecological Education Policy Framework from 457 students and 180 staff (Hong Kong),abstract,ok
82+7133185348,Photorealistic Text-To-Image Diffusion Models with Deep Language Understanding,2022,conference-paper,generative_ai,10.52202/068431-2643,,1706.0,230.5,method,Imagen: photorealistic text-to-image diffusion with large frozen text encoders,metadata_only,no_abstract
83+3195296536,Generative Adversarial Trainer Defense to Adversarial Perturbations with GAN,2020,dissertation,generative_ai,,,48.0,221.7,method,Dissertation on GAN-based adversarial defence,metadata_only,no_abstract
84+4322617812,Generative AI and the future of education: Ragnarok or reformation? A paradoxical perspective from management educators,2023,article,generative_ai,10.1016/j.ijme.2023.100790,,873.0,204.6,theory,Paradox-theory essay on generative AI in education,abstract,ok
85+3096831136,Generative adversarial networks,2020,article,generative_ai,10.1145/3422622,,1473.0,198.6,survey,CACM review of generative adversarial networks,abstract,ok
86+4400770292,Generative AI Can Harm Learning,2024,preprint,generative_ai,10.2139/ssrn.4895486,,198.0,174.2,empirical,Field experiment: GPT-4 tutoring access improved practice but harmed unassisted exam performance; SSRN preprint,metadata_only,no_abstract
87+7127274868,Parental AI Investment Behaviours and AI Mediated Informal Digital Learning of English: The Chain Mediating Role of Children's Perceived AI ,2026,article,generative_ai,10.1002/jcal.70198,,48.0,341.0,empirical,Survey study of parental AI investment and informal English learning,abstract,ok
88+7124974472,The Rapid Adoption of Generative AI,2026,article,generative_ai,10.1287/mnsc.2025.02523,,42.0,46.7,empirical,Nationally representative US surveys of generative-AI adoption (45% of adults 18-64 by late 2024),abstract,ok
89+7118187240,Patch Time Series TransformerBased ShortTerm Photovoltaic Power Prediction Enhanced by Artificial Fish,2026,article,generative_ai,10.3390/en19010284,,40.0,44.5,engineering,PatchTST-based photovoltaic forecasting with GAN features (application),abstract,off_scope
90+4406199489,Generative AI in Higher Education: Balancing Innovation and Integrity,2025,article,generative_ai,10.3389/bjbs.2024.14048,,278.0,241.4,survey,Review of generative AI and integrity in higher education,abstract,ok
91+4398243313,A scoping review on how generative artificial intelligence transforms assessment in higher education,2024,article,generative_ai,10.1186/s41239-024-00468-z,,463.0,139.4,survey,Scoping review of 32 studies on generative AI and assessment,abstract,ok
92+4409269614,Generative AI and Academic Integrity in Higher Education: A Systematic Review and Research Agenda,2025,review,generative_ai,10.3390/info16040296,,271.0,14.2,survey,Systematic review (41 studies) of generative AI and academic integrity,abstract,ok
93+3193917007,Generative Adversarial Networks,2021,book-chapter,generative_ai,10.1007/978-1-4842-7341-8_10,,195.0,127.1,survey,Textbook chapter on GANs,abstract,ok
94+4288099666,Classifier-Free Diffusion Guidance,2022,preprint,generative_ai,10.48550/arxiv.2207.12598,2207.12598,575.0,124.8,method,Classifier-free diffusion guidance: joint conditional/unconditional training; preprint,abstract,ok
95+4394934695,The impact of generative AI on higher education learning and teaching: A study of educators perspectives,2024,article,generative_ai,10.1016/j.caeai.2024.100221,,394.0,118.6,empirical,Educator interviews on generative AI in higher education,abstract,ok
96+4321648656,From Paintbrush to Pixel: A Review of Deep Neural Networks in AI-Generated Art,2026,article,generative_ai,10.1163/22134913-bja10078,2302.10913,33.0,88.5,survey,Review of neural architectures for AI-generated art,abstract,ok
97+4395957952,Guidance for generative AI in education and research,2023,book,generative_ai,10.54675/ewzm9535,,400.0,116.7,survey,UNESCO guidance for generative AI in education and research,metadata_only,no_abstract
98+4394863715,How Generative AI Can Augment Human Creativity,2024,preprint,generative_ai,10.2139/ssrn.4759930,,131.0,114.3,empirical,SSRN preprint on generative AI and human creativity,metadata_only,no_abstract
99+4407778399,"Investigating the higher education institutions guidelines and policies regarding the use of generative AI in teaching, learning, research, ",2025,article,generative_ai,10.1186/s41239-025-00507-3,,213.0,184.9,empirical,Analysis of generative-AI guidelines at 50 US universities,abstract,ok
100+7124432017,Human-Centric Artificial Intelligence Pedagogy (HCAP) framework developed from TPACK through integration of artificial intelligence literacy,2026,article,generative_ai,10.1080/10494820.2026.2615818,,30.0,213.1,theory,Intelligent-TPACK pedagogy framework,abstract,ok
101+7140170686,Pedagogical partnerships with generative AI in higher education: how dual cognitive pathways paradoxically enable transformative learning,2026,article,generative_ai,10.1186/s41239-026-00585-x,,30.0,213.1,empirical,Mixed-methods study of human-GenAI pedagogical partnerships,abstract,ok
102+4406728221,Agentic AI: Autonomous Intelligence for Complex GoalsA Comprehensive Survey,2025,article,ai_agents,10.1109/access.2025.3532853,,694.0,236.4,survey,Survey of agentic AI concepts and applications,abstract,ok
103+7161734976,A multi-agent system for automating scientific discovery,2026,article,ai_agents,10.1038/s41586-026-10652-y,,62.0,175.1,engineering,Robin: multi-agent system automating hypothesis generation and data analysis for experimental biology,abstract,ok
104+7135409568,The role of agentic artificial intelligence in healthcare: a scoping review,2026,article,ai_agents,10.1038/s41746-026-02517-5,,52.0,23.7,survey,Scoping review of agentic AI in healthcare,abstract,ok
105+7134187809,Dynamic event-triggered approximate optimal consensus control for unknown nonlinear multi-agent systems via adaptive dynamic programming,2026,article,ai_agents,10.1016/j.isatra.2026.03.010,,46.0,129.9,theory,Event-triggered optimal consensus control for nonlinear multi-agent systems (control theory),metadata_only,off_scope
106+4390730779,"Exploring Large Language ModelBased Intelligent Agents: Definitions, Methods, and Prospects",2026,article,ai_agents,10.1002/widm.70111,2401.03428,44.0,49.0,survey,"Survey of LLM-based agents: definitions, methods, prospects",abstract,ok
107+4225355126,Introduction to Multi-agent Systems,2022,book-chapter,ai_agents,10.1007/978-981-19-0493-6_1,,268.0,143.6,survey,Textbook introduction to multi-agent systems,metadata_only,no_abstract
108+4407570917,AgentSociety: Large-scale simulation of LLM-driven generative agents advances understanding of human behaviors and society,2026,article,ai_agents,10.26599/if.2026.9710004,2502.08691,40.0,44.5,engineering,AgentSociety: large-scale simulation of LLM-driven generative social agents,abstract,ok
109+4410197190,The role of agentic AI in shaping a smart future: A systematic review,2025,review,ai_agents,10.1016/j.array.2025.100399,,267.0,33.1,survey,Narrative review of agentic AI in organisations,abstract,ok
110+4413427262,"AI Agents vs. Agentic AI: A Conceptual taxonomy, applications and challenges",2025,article,ai_agents,10.1016/j.inffus.2025.103599,2505.10468,258.0,87.9,survey,Conceptual taxonomy distinguishing AI agents from agentic AI,abstract,ok
111+4414991289,DIVE into hydrogen storage materials discovery with AI agents,2026,article,ai_agents,10.1039/d5sc09921h,2508.13251,31.0,23.9,engineering,DIVE: multi-agent workflow extracting experimental data from figures for hydrogen-storage materials,abstract,ok
112+7122857583,"Agentic AI systems in the age of generative models: architectures, cloud scalability, and real-world applications",2026,article,ai_agents,10.1007/s10462-025-11458-6,,30.0,33.4,theory,Conceptual agentic-AI framework (literature synthesis),abstract,ok
113+4401023556,Large Language Model Based Multi-agents: A Survey of Progress and Challenges,2024,conference-paper,ai_agents,10.24963/ijcai.2024/890,,371.0,103.1,survey,Survey of LLM-based multi-agent systems,abstract,ok
114+4414023998,"The Rise of Agentic AI: A Review of Definitions, Frameworks, Architectures, Applications, Evaluation Metrics, and Challenges",2025,article,ai_agents,10.3390/fi17090404,,189.0,64.4,survey,Review of 143 primary studies on agentic AI systems,abstract,ok
115+7118083939,Agentic information systems,2026,article,ai_agents,10.1007/s12525-025-00861-0,,26.0,32.7,theory,Conceptual overview of agentic information systems,abstract,ok
116+4412505619,"AI Agents vs. Agentic AI: A Conceptual Taxonomy, Applications and Challenges",2025,article,ai_agents,10.70777/si.v2i3.15161,,181.0,61.7,survey,Duplicate record of the AI agents vs agentic AI taxonomy (different venue),abstract,ok
117+4416209823,"Agentic AI: a comprehensive survey of architectures, applications, and future directions",2025,article,ai_agents,10.1007/s10462-025-11422-4,2510.25445,165.0,56.2,survey,PRISMA-based review (90 studies) with symbolic vs neural dual-paradigm framework for agentic AI,abstract,ok
118+7155930820,Adaptive Prescribed Performance Secure Consensus Control for High-Order Constrained Multi-Agent Systems With Time-Varying Powers and Arbitra,2026,article,ai_agents,10.1109/tase.2026.3686874,,23.0,80.7,theory,Prescribed-performance consensus control for multi-agent systems (control theory),metadata_only,off_scope
119+7162257503,Predefined-time sliding-mode formation control for second-order nonlinear multi-agent systems under deception attacks,2026,article,ai_agents,10.1016/j.ejcon.2026.101534,,22.0,77.2,theory,Predefined-time sliding-mode formation control (control theory),metadata_only,off_scope
120+4412709685,The Virtual Lab of AI agents designs new SARS-CoV-2 nanobodies,2025,article,ai_agents,10.1038/s41586-025-09442-9,,150.0,33.5,engineering,Virtual Lab: LLM agents designing SARS-CoV-2 nanobodies with experimental validation,metadata_only,no_abstract
121+7128165235,Multi-agent AI,2026,article,ai_agents,10.1007/s12525-025-00862-z,,21.0,23.4,theory,Five-component framework for multi-agent AI in knowledge work,abstract,ok
122+7118190454,Redefining Elderly Care With Agentic AI: Challenges and Opportunities,2026,article,ai_agents,10.1109/ojcs.2026.3650842,2507.14912,21.0,22.2,survey,Perspective on agentic AI for elderly care,abstract,ok
123+7130481397,Benchmarking large language model-based agent systems for clinical decision tasks,2026,article,ai_agents,10.1038/s41746-026-02443-6,,21.0,9.6,benchmark,"Benchmarking two agent systems (OpenManus, Manus) on clinical decision tasks",abstract,ok
124+4390638955,Artificial intelligence (AI) learning tools in K-12 education: A scoping review,2024,article,ai_agents,10.1007/s40692-023-00304-9,,240.0,72.3,survey,Scoping review of AI learning tools in K-12 education,abstract,off_scope
125+4412203333,A Survey on the Memory Mechanism of Large Language Model-based Agents,2025,article,ai_agents,10.1145/3748302,2404.13501,139.0,47.4,survey,Survey of memory mechanisms of LLM-based agents,abstract,ok
126+4399982941,Value engineering for autonomous agents,2024,dissertation,ai_agents,10.5821/dissertation-2117-410409,,8.0,68.2,theory,Dissertation on value engineering for autonomous agents (normative MAS),abstract,normalisation_artefact
127+2911283634,10. Reinforcement learning,2020,book-chapter,reinforcement_learning,10.1515/9783110595567-011,,532.0,276.0,survey,Textbook chapter on reinforcement learning,metadata_only,no_abstract
128+4389473504,TRUST-REGION BASED POLICY OPTIMIZATION FOR EFFICIENT REINFORCEMENT LEARNING,2023,dissertation,reinforcement_learning,10.23860/diss-li-hepfeng-2023,,28.0,232.9,method,Dissertation on trust-region policy optimisation,abstract,ok
129+2964083594,Large-Scale Study of Curiosity-Driven Learning,2025,conference-paper,reinforcement_learning,10.57702/mspz8fly,1808.04355,414.0,367.9,empirical,Large-scale study of curiosity-driven learning across 54 environments (ICLR 2019; OpenAlex year 2025 is a repository copy),abstract,year_implausible
130+2098774185,Maximum Entropy Inverse Reinforcement Learning,2018,article,reinforcement_learning,10.1184/r1/6555512,,1077.0,203.3,method,Maximum-entropy inverse reinforcement learning (AAAI 2008; OpenAlex year 2018),abstract,year_implausible
131+1591675293,Bayesian Inverse Reinforcement Learning,2025,book-chapter,reinforcement_learning,10.1017/9781009449441.027,,395.0,665.0,method,Bayesian inverse reinforcement learning (IJCAI 2007; OpenAlex year 2025 book chapter),abstract,year_implausible
132+7123365826,Event-Triggered Zero-Sum Game for Input Saturated Nonlinear Systems With State Constraints,2026,article,reinforcement_learning,10.1109/tce.2026.3652706,,55.0,155.3,theory,Event-triggered zero-sum game control for saturated nonlinear systems (control theory),abstract,off_scope
133+2991046523,Multi-Agent Reinforcement Learning: A Selective Overview of Theories and Algorithms,2021,book-chapter,reinforcement_learning,10.1007/978-3-030-60990-0_12,1911.10635,530.0,178.4,survey,Selective overview of multi-agent RL theory and algorithms,metadata_only,no_abstract
134+2973229164,Continuous control for robot based on deep reinforcement learning,2019,dissertation,reinforcement_learning,10.32657/10356/90191,,47.0,171.2,method,Dissertation on deep RL for continuous robot control,abstract,ok
135+2982316857,Grandmaster level in StarCraft II using multi-agent reinforcement learning,2019,article,reinforcement_learning,10.1038/s41586-019-1724-z,,1021.0,155.2,engineering,AlphaStar: grandmaster level in StarCraft II with multi-agent RL,metadata_only,no_abstract
136+2902907165,"A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play",2018,article,reinforcement_learning,10.1126/science.aar6404,,733.0,138.3,method,"AlphaZero: one self-play RL algorithm for chess, shogi and Go",abstract,ok
137+7133481537,Filter-based predefined-time optimal fault-tolerant consensus control for nonlinear multi-agent systems via reinforcement learning,2026,article,reinforcement_learning,10.1016/j.neucom.2026.133224,,35.0,122.9,theory,Predefined-time fault-tolerant consensus control (control theory),metadata_only,off_scope
138+7133226216,The Surprising Effectiveness of PPO in Cooperative Multi-Agent Games,2022,conference-paper,reinforcement_learning,10.52202/068431-1787,2103.01955,482.0,95.8,empirical,MAPPO: PPO is surprisingly strong in cooperative multi-agent benchmarks,abstract,ok
139+3041807988,Post-Covid-19 Education and Education Technology Solutionism: a Sellers Market,2020,article,reinforcement_learning,10.1007/s42438-020-00164-x,,322.0,93.8,theory,Critical essay on ed-tech solutionism after Covid-19 (not RL),abstract,off_scope
140+2781726626,Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor,2018,preprint,reinforcement_learning,10.48550/arxiv.1801.01290,1801.01290,775.0,90.8,method,Soft actor-critic: off-policy maximum-entropy RL; preprint record,abstract,ok
141+2963923407,Addressing Function Approximation Error in Actor-Critic Methods,2018,article,reinforcement_learning,,1802.09477,472.0,89.1,method,TD3: clipped double-Q and delayed policy updates against overestimation,abstract,ok
142+7122543576,SMSBased EventTriggered Fault Tolerant Control of Nonlinear Systems Using Adaptive Dynamic Programming,2026,article,reinforcement_learning,10.1002/rnc.70380,,25.0,70.6,theory,ADP-based event-triggered fault-tolerant control (control theory),abstract,off_scope
143+2899138465,Learning to Learn with Gradients,2018,dissertation,reinforcement_learning,,,20.0,88.6,method,Dissertation on gradient-based meta-learning (MAML lineage),abstract,ok
144+4391631327,Axiom-World: A Pre-Registered Study of Two-Stage Post-Training for Rule-Grounded Planning in a Verifiable Toy World,2024,preprint,reinforcement_learning,10.48550/arxiv.2402.03300,2402.03300,99.0,86.4,empirical,Pre-registered toy study of two-stage post-training; record carries arXiv 2402.03300 (DeepSeekMath) and its citations,abstract,identifier_mismatch
145+4409147637,Mastering diverse control tasks through world models,2025,article,reinforcement_learning,10.1038/s41586-025-08744-2,,170.0,57.9,method,DreamerV3: world-model RL mastering 150+ tasks with one configuration (Nature),abstract,ok
146+2895974005,Exploration and Exploitation in Reinforcement Learning,2018,dissertation,reinforcement_learning,,,18.0,79.8,empirical,Master's thesis on exploration/exploitation parameters,abstract,normalisation_artefact
147+7134057781,A fuzzy-TD3 hybrid reinforcement learning framework for robust trajectory tracking of the Mitsubishi RV-2AJ robotic arm,2026,article,reinforcement_learning,10.1038/s41598-026-42615-8,,21.0,23.4,engineering,Fuzzy-TD3 hybrid control for a 5-DOF manipulator (application),abstract,ok
148+2754517384,Deep Reinforcement Learning That Matters,2018,conference-paper,reinforcement_learning,10.1609/aaai.v32i1.11694,1709.06560,452.0,70.0,empirical,Deep RL That Matters: reproducibility and variance of deep RL results,abstract,ok
149+2761873684,Rainbow: Combining Improvements in Deep Reinforcement Learning,2018,conference-paper,reinforcement_learning,10.1609/aaai.v32i1.11796,,442.0,68.4,method,"Rainbow: combining six DQN extensions, with ablations",abstract,ok
150+7133208539,Direct Preference Optimization: Your Language Model is Secretly a Reward Model,2023,conference-paper,reinforcement_learning,10.52202/075280-2338,,293.0,67.6,method,Direct preference optimisation: closed-form RLHF objective without a separate reward model,metadata_only,no_abstract
151+7128436456,Synthesis of model predictive control and reinforcement learning: Survey and classification,2026,article,reinforcement_learning,10.1016/j.arcontrol.2026.101045,2502.02133,19.0,21.0,survey,Survey classifying combinations of MPC and RL,abstract,ok
152+2962862931,On a Method to Measure Supervised Multiclass Models Interpretability: Application to Degradation Diagnosis (Short Paper),2024,conference-paper,ml_theory,10.4230/oasics.dx.2024.27,,12586.0,3498.4,method,"Interpretability measure for degradation diagnosis (HAL); c36=12,586 cannot belong to this record",abstract,identifier_mismatch
153+1698155719,Support Vector Machines,2020,book-chapter,ml_theory,10.1017/9781108564175.026,,1212.0,971.7,survey,Textbook chapter on support vector machines,abstract,ok
154+4288283362,Spiral Time: A Geometric Reframing of Temporal Structure and Its Applications in Machine Learning,2026,article,ml_theory,10.5281/zenodo.20632938,,174.0,576.9,method,'Spiral Time' geometric time encoding with a small LSTM ablation; no venue; NCS36 implausible,abstract,normalisation_artefact
155+46659105,"Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation",2020,preprint,ml_theory,10.4230/lipics.esem.2026.21,2010.16061,3906.0,608.2,theory,Title = Powers (2011) evaluation-measures paper (arXiv 2010.16061) but the stored abstract is an LLM-as-judge study of program repair,abstract,abstract_mismatch
156+3145506661,Introduction to Machine Learning,2019,book-chapter,ml_theory,10.1007/978-3-030-10546-4_1,,1581.0,491.8,survey,Textbook: Introduction to Machine Learning (Springer Briefs),metadata_only,no_abstract
157+2291445757,Support Vector Regression,2024,book-chapter,ml_theory,10.1017/9781009006767.012,,769.0,426.8,survey,Textbook chapter on support vector regression,abstract,ok
158+2100415632,Efficient learning by implicit exploration in bandit problems with side observations,2026,preprint,ml_theory,10.48550/arxiv.2604.24555,2604.24555,116.0,667.5,theory,Bandits with side observations (2014 paper carrying a 2026 arXiv id and year),abstract,year_implausible
159+1568555062,Equivalence and Synthesis of Causal Models,2022,book-chapter,ml_theory,10.1145/3501714.3501732,,576.0,308.7,theory,"Verma & Pearl: equivalence and synthesis of causal models (reprint chapter, 2022)",abstract,ok
160+2751318774,Introduction to Machine Learning,2019,book-chapter,ml_theory,10.1142/9789811201967_0001,,978.0,304.2,survey,Textbook: Introduction to Machine Learning (Alpaydin),abstract,ok
161+3208688400,Anomaly Detection: A Survey,2021,book-chapter,ml_theory,10.1007/978-981-16-2102-4_36,,833.0,280.5,survey,Anomaly detection survey (book chapter),metadata_only,no_abstract
162+3105488314,Risk bounds for statistical learning,2020,dissertation,ml_theory,,,59.0,272.4,theory,Dissertation on risk bounds for statistical learning,metadata_only,no_abstract
163+57706852,Stochastic simultaneous optimistic optimization,2026,preprint,ml_theory,10.48550/arxiv.2604.24537,2604.24537,72.0,414.3,theory,StoSOO stochastic optimistic optimisation (2013 paper carrying a 2026 arXiv id and year),abstract,year_implausible
164+2946444988,Towards Accurate Model Selection in Deep Unsupervised Domain Adaptation,2026,preprint,ml_theory,10.48550/arxiv.2606.04665,2606.04665,68.0,2646.0,method,Deep Embedded Validation for UDA model selection (2019 paper carrying a 2026 arXiv id; NCS36 implausible),abstract,year_implausible
165+2945976633,Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead,2019,article,ml_theory,10.1038/s42256-019-0048-x,1811.10154,1556.0,236.6,theory,Rudin: argument for interpretable models over post-hoc explanation in high-stakes decisions,metadata_only,no_abstract
166+4400823820,Random Forest Algorithm Overview,2024,article,ml_theory,10.58496/bjml/2024/007,,782.0,235.4,survey,Overview of random forests (non-indexed journal),abstract,ok
167+2999309192,The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation,2020,article,ml_theory,10.1186/s12864-019-6413-7,,1781.0,222.8,theory,Advantages of MCC over F1 and accuracy for imbalanced binary classification,abstract,ok
168+2999615587,From local explanations to global understanding with explainable AI for trees,2020,article,ml_theory,10.1038/s42256-019-0138-9,,1767.0,221.0,method,TreeExplainer: exact SHAP values for tree ensembles (NMI),metadata_only,no_abstract
169+2439568532,The Mythos of Model Interpretability,2018,article,ml_theory,10.1145/3233231,,1085.0,204.8,theory,Lipton: the mythos of model interpretability,abstract,ok
170+3194464139,Penerapan Algoritma K-Nearest Neighbor untuk Penentuan Resiko Kredit Kepemilikan Kendaraan Bermotor,2020,dissertation,ml_theory,,,21.0,197.5,engineering,Indonesian thesis applying kNN to credit risk; NCS36 implausible for c36=21,abstract,normalisation_artefact
171+3041133507,A Comprehensive Survey on Transfer Learning,2020,article,ml_theory,10.1109/jproc.2020.3004555,1911.02685,1538.0,192.4,survey,Comprehensive survey of transfer learning (Proc. IEEE),abstract,ok
172+2436219157,Optimization Methods for Large-Scale Machine Learning,2018,article,ml_theory,10.1137/16m1080173,1606.04838,956.0,180.4,survey,"Bottou, Curtis, Nocedal: optimisation methods for large-scale ML (SIAM Review)",abstract,ok
173+3135028703,"Machine Learning: Algorithms, Real-World Applications and Research Directions",2021,article,ml_theory,10.1007/s42979-021-00592-x,,1518.0,176.9,survey,Overview of ML algorithms and applications (SN Computer Science),metadata_only,no_abstract
174+7124426637,A review of machine learning with small and limited data,2026,article,ml_theory,10.1186/s40537-025-01346-9,,46.0,51.2,survey,Review of machine learning with small and limited data,abstract,ok
175+2165856466,Spectral bandits for smooth graph functions,2026,preprint,ml_theory,10.48550/arxiv.2604.18420,2604.18420,45.0,259.0,theory,Spectral bandits for smooth graph functions (2014 paper carrying a 2026 arXiv id and year),abstract,year_implausible
176+4402843978,Interpretable machine learning,2020,report,ml_theory,10.58248/pn633,,455.0,158.3,survey,UK POSTnote on interpretable ML,abstract,ok
177+2964121744,AutoResearch Lite: a reproducible paper-inspired experiment runner (Adam mechanism reproduction and convergence-speed comparison),2026,article,deep_learning_architectures,10.4230/oasics.dx.2026.15,2108.04578,45775.0,50938.5,engineering,"'AutoResearch Lite' Adam reproduction runner; c36=45,775 and arXiv 2108.04578 cannot belong to this record",abstract,identifier_mismatch
178+1710476689,Generative Adversarial Networks,2022,book-chapter,deep_learning_architectures,10.1017/9781108891530.013,,2385.0,1018.2,survey,Textbook chapter on GANs (The Science of Deep Learning),abstract,ok
179+4322614756,Identifying Resilient Communities in Road Networks: A Path-Based Embedding Approach,2025,conference-paper,deep_learning_architectures,10.4230/lipics.giscience.2025.9,,1542.0,1370.5,method,"Path-based embedding for road-network community detection; c36=1,542 implausible for a LIPIcs paper",abstract,identifier_mismatch
180+2963399829,Mixup: Beyond empirical risk minimization,2024,preprint,deep_learning_architectures,10.57702/dcy1c3gw,1710.09412,773.0,674.3,method,mixup: training on convex combinations of examples (ICLR 2018; OpenAlex year 2024),abstract,year_implausible
181+4382203079,Are Transformers Effective for Time Series Forecasting?,2023,conference-paper,deep_learning_architectures,10.1609/aaai.v37i9.26317,2205.13504,2278.0,621.0,empirical,Zeng et al.: simple linear models beat Transformers on long-term time-series forecasting benchmarks,abstract,ok
182+4406170795,Accurate predictions on small data with a tabular foundation model,2025,article,deep_learning_architectures,10.1038/s41586-024-08328-6,,974.0,331.8,method,TabPFN: tabular foundation model (prior-data fitted network) for small data (Nature),abstract,ok
183+146900863,Neural Information Processing Systems,2018,book-chapter,deep_learning_architectures,10.7551/mitpress/11474.003.0014,,999.0,464.0,survey,NeurIPS proceedings volume record (MIT Press),metadata_only,no_abstract
184+4396821501,KAN: Kolmogorov-Arnold Networks,2024,preprint,deep_learning_architectures,10.48550/arxiv.2404.19756,2404.19756,531.0,463.2,method,KAN: Kolmogorov-Arnold networks with learnable spline edge functions; preprint,abstract,ok
185+2950067852,A Recurrent Latent Variable Model for Sequential Data,2026,preprint,deep_learning_architectures,10.5281/zenodo.19983308,,119.0,4630.6,method,Variational RNN (NeurIPS 2015) recorded as a 2026 preprint; NCS36 implausible,abstract,year_implausible
186+4221150249,Graph Neural Networks for Graphs With Heterophily: A Survey,2026,article,deep_learning_architectures,10.1109/tkde.2026.3680353,2202.07082,107.0,119.1,survey,Survey of GNNs for heterophilic graphs,abstract,ok
187+7148890333,Spatial-channel collaborative multi-scale graph interaction deep transfer learning for unsupervised rotating machinery fault diagnosis,2026,article,deep_learning_architectures,10.1016/j.engappai.2026.114691,,102.0,113.0,method,Multi-scale graph transfer learning for rotating-machinery diagnosis,metadata_only,no_abstract
188+3177318507,Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting,2021,conference-paper,deep_learning_architectures,10.1609/aaai.v35i12.17325,2012.07436,1344.0,315.3,method,Informer: ProbSparse attention for long-sequence forecasting,abstract,ok
189+4389326242,Mamba: Linear-Time Sequence Modeling with Selective State Spaces,2023,preprint,deep_learning_architectures,10.48550/arxiv.2312.00752,2312.00752,912.0,303.2,method,Mamba: selective state-space model with linear-time sequence modelling; preprint,abstract,ok
190+7118302872,A frame work to predict the influencing nodes in social networking platform using graph neural network model,2026,article,deep_learning_architectures,10.1007/s13278-025-01574-0,,85.0,87.6,method,GNN + ant colony optimisation for influence maximisation,abstract,ok
191+3137147200,A Survey of Quantization Methods for Efficient Neural Network Inference,2022,book-chapter,deep_learning_architectures,10.1201/9781003162810-13,,698.0,298.0,survey,Survey of quantisation methods for efficient inference (book chapter),abstract,ok
192+3140854437,"Review of deep learning: concepts, CNN architectures, challenges, applications, future directions",2021,article,deep_learning_architectures,10.1186/s40537-021-00444-8,,2452.0,293.1,survey,Review of deep learning concepts and CNN architectures,abstract,ok
193+4210257598,A Comprehensive Survey on Graph Neural Networks,2020,article,deep_learning_architectures,,,2239.0,280.0,survey,Comprehensive survey on GNNs (repository duplicate of the TNNLS paper),abstract,ok
194+2907492528,A Comprehensive Survey on Graph Neural Networks,2020,article,deep_learning_architectures,10.1109/tnnls.2020.2978386,1901.00596,2230.0,278.9,survey,Comprehensive survey on GNNs (TNNLS),abstract,ok
195+3152893301,Graph neural networks: A review of methods and applications,2020,article,deep_learning_architectures,10.1016/j.aiopen.2021.01.001,1812.08434,2023.0,253.0,survey,GNNs: review of methods and applications (AI Open),abstract,ok
196+7110333790,Enriching Location Representation with Detailed Semantic Information,2025,article,deep_learning_architectures,10.4230/lipics.giscience.2025.3,,496.0,169.0,method,CaLLiPer+: POI-name-enriched location embeddings; c36=496 implausible for a LIPIcs paper,abstract,identifier_mismatch
197+4385763767,Transformers in Time Series: A Survey,2023,conference-paper,deep_learning_architectures,10.24963/ijcai.2023/759,2202.07125,921.0,251.1,survey,Survey of Transformers in time series,abstract,ok
198+3124006607,SIMPLE SPECTRAL GRAPH CONVOLUTION,2026,conference-paper,deep_learning_architectures,10.26190/unsworks/32608,,69.0,529.6,method,Simple spectral graph convolution (ICLR 2021; OpenAlex year 2026),abstract,year_implausible
199+4285723986,Graph Attention Networks,2020,book-chapter,deep_learning_architectures,10.1007/978-3-031-01587-8_7,,471.0,244.4,survey,Synthesis lecture on graph attention networks,metadata_only,no_abstract
200+2892035503,Deep learning for time series classification: a review,2019,article,deep_learning_architectures,10.1007/s10618-019-00619-1,1809.04356,908.0,192.7,survey,Review of deep learning for time-series classification,metadata_only,no_abstract
201+2962776659,Fine-grained Analysis of Sentence Embeddings Using Auxiliary Prediction Tasks,2024,preprint,deep_learning_architectures,10.57702/tcr06vve,,212.0,184.9,empirical,Auxiliary-task analysis of sentence embeddings (ICLR 2017; OpenAlex year 2024),abstract,year_implausible
202+4406063062,Responsible artificial intelligence governance: A review and research framework,2025,article,ai_safety,10.1016/j.jsis.2024.101885,,534.0,74.5,survey,Review and research framework for responsible AI governance (IS),abstract,ok
203+7130596601,"Human in the loop artificial intelligence in healthcare: applications, outcomes, and implementation challenges",2026,article,ai_safety,10.1016/j.ijmedinf.2026.106362,,65.0,29.6,survey,Review of human-in-the-loop AI in healthcare,abstract,ok
204+3084163594,Age of Surveillance Capitalism The Fight for a Human Future at the New Frontier of Power.,2020,article,ai_safety,10.26522/brocked.v29i2.849,,775.0,225.7,survey,Book review of Zuboff's Surveillance Capitalism for educators,abstract,off_scope
205+2963808661,Scaling of End-To-End Governance Risk Assessments for AI Systems (Practitioner Track),2025,conference-paper,ai_safety,10.4230/oasics.saia.2024.4,,390.0,126.3,engineering,Governance risk-assessment tooling (OASIcs practitioner track); c36=390 implausible,abstract,identifier_mismatch
206+7133236347,"TRiSM for Agentic AI: A review of Trust, Risk, and Security Management in LLM-based Agentic Multi-Agent Systems",2026,article,ai_safety,10.1016/j.aiopen.2026.02.006,2506.04133,51.0,44.5,survey,"TRiSM (trust, risk, security management) for LLM-based multi-agent systems",abstract,ok
207+7128782475,Large language models provide unsafe answers to patient-posed medical questions,2026,article,ai_safety,10.1038/s41746-026-02428-5,,51.0,23.2,empirical,Physician-led red-teaming of four chatbots on 222 patient questions (HealthAdvice),abstract,ok
208+4406571986,Generative artificial intelligence in higher education: Evidence from an analysis of institutional policies and guidelines,2025,article,ai_safety,10.1016/j.chbah.2025.100121,2402.01659,347.0,301.3,empirical,Analysis of generative-AI policies at 116 US R1 universities,abstract,ok
209+2981731882,"Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI",2019,article,ai_safety,10.1016/j.inffus.2019.12.012,1910.10045,1097.0,166.8,survey,"Arrieta et al.: XAI concepts, taxonomies and responsible AI",metadata_only,no_abstract
210+4402698709,"The AI risk repository: A meta-review, database, and taxonomy of risks from artificial intelligence",2026,article,ai_safety,10.1016/j.patter.2026.101517,2408.12622,44.0,38.4,survey,"AI Risk Repository: 1,725 risks from 74 taxonomies with causal and domain taxonomies",abstract,ok
211+4408314709,Bias recognition and mitigation strategies in artificial intelligence healthcare applications,2025,article,ai_safety,10.1038/s41746-025-01503-7,,303.0,42.3,survey,Review of bias recognition and mitigation in healthcare AI,abstract,ok
212+4407243994,"AI Ethics: Integrating Transparency, Fairness, and Privacy in AI Development",2025,article,ai_safety,10.1080/08839514.2025.2463722,,300.0,41.8,theory,Comparative ethical framework (EU/US/China policies),abstract,ok
213+4330336443,GPTs are GPTs: An Early Look at the Labor Market Impact Potential of Large Language Models,2023,preprint,ai_safety,10.48550/arxiv.2303.10130,2303.10130,495.0,146.6,empirical,Title = 'GPTs are GPTs' (arXiv 2303.10130) but the stored abstract describes a TRIPS/Anthropic Economic Index comparison,abstract,abstract_mismatch
214+7128600296,Advancing healthcare AI governance through a comprehensive maturity model based on systematic review,2026,review,ai_safety,10.1038/s41746-026-02418-7,,41.0,12.8,survey,Systematic review of 35 healthcare-AI governance frameworks and a maturity model,abstract,ok
215+4399450035,Power Hungry Processing: Watts Driving the Cost of AI Deployment?,2024,conference-paper,ai_safety,10.1145/3630106.3658542,2311.16863,359.0,138.5,empirical,Power Hungry Processing: measured inference energy of task-specific vs general-purpose models,abstract,ok
216+4318586239,Artificial Intelligence: Definition and Background,2023,book-chapter,ai_safety,10.1007/978-3-031-21448-6_2,,247.0,136.6,survey,Policy book chapter defining AI,abstract,ok
217+4360615722,Chatbots in Education and Research: A Critical Examination of Ethical Implications and Solutions,2023,article,ai_safety,10.3390/su15075614,,557.0,130.5,theory,Qualitative examination of ethical implications of chatbots in education,abstract,ok
218+2670253439,Explanation in artificial intelligence: Insights from the social sciences,2018,article,ai_safety,10.1016/j.artint.2018.07.007,1706.07269,676.0,127.6,survey,Miller: explanation in AI through the lens of the social sciences,metadata_only,no_abstract
219+7118631510,Towards trustworthy AI in Industry 5.0: Ante-hoc interpretability with deep learning,2026,article,ai_safety,10.1016/j.cie.2026.111805,,35.0,38.9,method,Ante-hoc interpretability for Industry 5.0,metadata_only,no_abstract
220+7127143535,"Industrial information integration through autonomous AI agents: paradoxical effects on transparency, dehumanization, and responsible retail",2026,article,ai_safety,10.1016/j.jii.2026.101089,,33.0,50.0,empirical,Autonomous AI agents in industrial information integration (paradoxical effects),metadata_only,no_abstract
221+4409249263,Ethical challenges and evolving strategies in the integration of artificial intelligence into clinical practice,2025,article,ai_safety,10.1371/journal.pdig.0000810,2412.03576,218.0,14.5,survey,Ethical challenges of AI in clinical practice,abstract,ok
222+4366262984,Explainable Artificial Intelligence (XAI): What we know and what is left to attain Trustworthy Artificial Intelligence,2023,article,ai_safety,10.1016/j.inffus.2023.101805,,1197.0,109.9,survey,XAI: what is known and what remains for trustworthy AI (Information Fusion),abstract,ok
223+2969896603,A Survey on Bias and Fairness in Machine Learning,2021,article,ai_safety,10.1145/3457607,1908.09635,1794.0,102.3,survey,Survey on bias and fairness in machine learning (CSUR),abstract,ok
224+4406152291,Medical large language models are vulnerable to data-poisoning attacks,2025,article,ai_safety,10.1038/s41591-024-03445-1,,199.0,67.8,empirical,Data-poisoning threat assessment: 0.001% of tokens with medical misinformation yields harmful LLMs,abstract,ok
225+4410609100,Jailbreaking Black Box Large Language Models in Twenty Queries,2025,conference-paper,ai_safety,10.1109/satml64287.2025.00010,2310.08419,193.0,171.5,method,PAIR: black-box semantic jailbreaks in about twenty queries,abstract,ok
226+4411542009,Expressing stigma and inappropriate responses prevents LLMs from safely replacing mental health providers.,2025,conference-paper,ai_safety,10.1145/3715275.3732039,2504.18412,191.0,61.8,empirical,Experiments showing LLMs express stigma and respond inappropriately as replacement therapists,abstract,ok
227+4360620450,"Opinion Paper: So what if ChatGPT wrote it? Multidisciplinary perspectives on opportunities, challenges and implications of generative conve",2023,article,llm,10.1016/j.ijinfomgt.2023.102642,,3137.0,113.5,survey,Dwivedi et al.: 43 expert contributions on ChatGPT opportunities and challenges,abstract,ok
228+3185341429,"Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing",2022,article,llm,10.1145/3560815,2107.13586,2462.0,253.7,survey,Liu et al.: systematic survey of prompt-based learning,abstract,ok
229+4384071683,Large language models encode clinical knowledge,2023,article,llm,10.1038/s41586-023-06291-2,2212.13138,2448.0,38.1,benchmark,Med-PaLM: MultiMedQA benchmark and human evaluation framework for clinical LLM answers,abstract,ok
230+4384918448,Behavioral Safety and Context Retention of Large Language Models in a Longitudinal ICU Simulation under Offline Conditions,2023,preprint,llm,10.48550/arxiv.2307.09288,2307.09288,2431.0,88.5,empirical,Longitudinal ICU simulation of 23 open-weight models; record carries arXiv 2307.09288 (Llama 2) and its citations,abstract,identifier_mismatch
231+4393065402,A survey on large language model based autonomous agents,2024,article,llm,10.1007/s11704-024-40231-1,2308.11432,1820.0,226.1,survey,Wang et al.: survey of LLM-based autonomous agents,abstract,ok
232+4319083882,ChatGPT: five priorities for research,2023,article,llm,10.1038/d41586-023-00288-7,,1590.0,146.0,theory,van Dis et al.: five priorities for research on ChatGPT (Nature comment),metadata_only,no_abstract
233+4399790131,The effects of over-reliance on AI dialogue systems on students' cognitive abilities: a systematic review,2024,review,ai_safety,10.1186/s40561-024-00316-7,,1484.0,81.8,survey,Systematic review of over-reliance on AI dialogue systems and cognition,abstract,ok
234+3133702157,On the Dangers of Stochastic Parrots,2021,conference-paper,llm,10.1145/3442188.3445922,,1464.0,232.1,theory,Bender et al.: risks of ever-larger language models ('stochastic parrots'),abstract,ok
235+4321499901,What if the devil is my guardian angel: ChatGPT as a case study of using chatbots in education,2023,article,llm,10.1186/s40561-023-00237-x,,1389.0,325.5,empirical,Qualitative case study of ChatGPT among early adopters in education,abstract,ok
added literature_synthesis/agent_E/make_bib_part05.py +117 −0
@@ -0,0 +1,117 @@
1+#!/usr/bin/env python3
2+"""Agent E, WP44 — build manuscript/references_part05.bib + references_verification_part05.csv.
3+Each entry is resolved by a single Crossref (DOI) or arXiv (id) lookup; the identifier comes from the ai_ml landmark CSV
4+(column doi / arxiv_id) or from an explicit override when the corpus record has none. Nothing is written without a
5+successful lookup. Usage: python3 make_bib_part05.py <repo_root>"""
6+import csv, json, sys, re, time, unicodedata, urllib.request, xml.etree.ElementTree as ET
7+R = sys.argv[1]
8+L = {r["work_id"]: r for r in csv.DictReader(open(f"{R}/literature_synthesis/landmarks/ai_ml_landmarks.csv"))}
9+MAIL = "spbou4@icloud.com"
10+# work_id: (keyword, override identifier or None)
11+W = {
12+"7124974472": ("rapidadoption", None), "4323655724": ("chatgptgood", None), "2990704537": ("superglue", "arxiv:1905.00537"),
13+"4393138529": ("peftsurvey", "arxiv:2403.14608"), "4410551352": ("aurora", None), "4392947521": ("uni", None), "4402220065": ("chief", None),
14+"4401397240": ("biomedgpt", None), "4411932262": ("centaur", None), "7133220561": ("instructblip", "arxiv:2305.06500"),
15+"7133250524": ("laion5b", "arxiv:2210.08402"), "2995460200": ("vlbert", "arxiv:1908.08530"), "4393148714": ("t2iadapter", None),
16+"4289785095": ("textualinversion", "arxiv:2208.01618"), "7164501397": ("textexploit", None), "4396599905": ("lowqualityfusion", None),
17+"4288099666": ("cfg", "arxiv:2207.12598"), "4406472463": ("mattergen", None), "4400770292": ("harmlearning", None),
18+"4406728221": ("agenticsurvey", None), "7161734976": ("robin", None), "4412709685": ("virtuallab", None), "4407570917": ("agentsociety", None),
19+"4413427262": ("agentsvsagentic", None), "4412203333": ("agentmemory", None), "7130481397": ("clinicalagents", None),
20+"2781726626": ("sac", "arxiv:1801.01290"), "2963923407": ("td3", "arxiv:1802.09477"), "2982316857": ("alphastar", None),
21+"2761873684": ("rainbow", None), "7133226216": ("mappo", "arxiv:2103.01955"), "4409147637": ("dreamerv3", None),
22+"2964083594": ("curiosity", "arxiv:1808.04355"), "2991046523": ("marloverview", None), "2945976633": ("stopexplaining", None),
23+"2439568532": ("mythos", None), "2999615587": ("treeexplainer", None), "2999309192": ("mcc", None), "3041133507": ("transfersurvey", None),
24+"2436219157": ("optimization", None), "4382203079": ("transformerstsf", None), "3177318507": ("informer", None),
25+"4385763767": ("tstransformers", None), "4406170795": ("tabpfn", None), "4396821501": ("kan", "arxiv:2404.19756"),
26+"2907492528": ("gnnsurvey", None), "3152893301": ("gnnreview", None), "2963399829": ("mixup", "arxiv:1710.09412"),
27+"2981731882": ("xai", None), "2969896603": ("biasfairness", None), "2670253439": ("explanation", None), "4366262984": ("xaitrust", None),
28+"4402698709": ("riskrepository", None), "4410609100": ("pair", None), "4406152291": ("poisoning", None), "7128782475": ("unsafeanswers", None),
29+"4411542009": ("stigma", None), "4406063062": ("raigovernance", None), "4406571986": ("hepolicies", None), "4366420437": ("rapidreview", None),
30+"4384464487": ("students", None), "4383312437": ("policyframework", None), "4389437528": ("chatgptedu", None), "4317910584": ("assessment", None),
31+"4322617812": ("ragnarok", None), "4413447271": ("radfm", None), "4415913530": ("titan", None), "4392947532": ("conch", None),
32+"4385948838": ("plip", None), "4411099958": ("panderm", None), "7128029160": ("brainiac", None), "7119541039": ("earthfm", None),
33+"4384071683": ("clinical", None), "3185341429": ("prompting", None), "4319083882": ("priorities", None), "4390730779": ("agentsdefinitions", None),
34+"7133236347": ("trism", None), "4407135816": ("fmradiology", None), "4390690043": ("datacentric", None), "4406472599": ("learningfrommodels", None),
35+"4410882873": ("multimodalmedicine", None), "4399790131": ("overreliance", None), "4398243313": ("assessmentreview", None),
36+"4409269614": ("integrity", None), "4387355843": ("timellm", "arxiv:2310.01728"), "4393153153": ("timesurl", None), "4402727669": ("mplugowl2", None),
37+"7148475777": ("gpt5medical", "arxiv:2508.08224"), "7127319731": ("medr1", None), "4414991289": ("dive", None), "4414023998": ("agenticreview", None),
38+"4416209823": ("agenticparadigms", None), "7128165235": ("multiagentai", None), "4408314709": ("biashealthcare", None),
39+"7140170686": ("pedagogical", None), "7135409568": ("agentichealthcare", None), "3137147200": ("quantization", None), "4361229539": ("llamaadapter", "arxiv:2303.16199"),
40+"7133233472": None, # QLoRA already in references.bib (dettmers2023qlora)
41+}
42+def get(url):
43+ req = urllib.request.Request(url, headers={"User-Agent": f"WP44-agentE (mailto:{MAIL})"})
44+ return urllib.request.urlopen(req, timeout=40).read()
45+def clean(s):
46+ s = unicodedata.normalize("NFKC", s)
47+ s = s.replace("‐", "-").replace("‑", "-").replace("‒", "-").replace("–", "--").replace("—", "---").replace("‘", "'").replace("’", "'").replace("“", "``").replace("”", "''").replace(" ", " ").replace("−", "-")
48+ s = "".join(c if ord(c) < 0x250 else unicodedata.normalize("NFKD", c)[0] if unicodedata.normalize("NFKD", c)[0].isascii() else "" for c in s)
49+ s = s.replace("&", "\\&").replace("%", "\\%").replace("#", "\\#").replace("$", "\\$")
50+ return re.sub(r"\s+", " ", s).strip()
51+def toks(s): return set(re.findall(r"[a-z0-9]+", unicodedata.normalize("NFKD", s).lower()))
52+def crossref(doi):
53+ m = json.loads(get(f"https://api.crossref.org/works/{doi}?mailto={MAIL}"))["message"]
54+ auth = [a for a in m.get("author", []) if a.get("family")]
55+ year = (m.get("issued") or m.get("published") or {}).get("date-parts", [[None]])[0][0]
56+ return {"type": m.get("type"), "title": m.get("title", [""])[0], "subtitle": (m.get("subtitle") or [""])[0], "authors": [(a.get("family", ""), a.get("given", "")) for a in auth], "container": (m.get("container-title") or [""])[0],
57+ "volume": m.get("volume"), "issue": m.get("issue"), "page": m.get("page"), "year": year, "publisher": m.get("publisher"), "doi": m["DOI"], "event": (m.get("event") or {}).get("name")}
58+def arxiv(aid):
59+ root = ET.fromstring(get(f"https://export.arxiv.org/api/query?id_list={aid}"))
60+ ns = {"a": "http://www.w3.org/2005/Atom"}; e = root.find("a:entry", ns)
61+ title = re.sub(r"\s+", " ", e.find("a:title", ns).text.strip())
62+ if title.startswith("Error"): raise RuntimeError(aid)
63+ authors = [a.find("a:name", ns).text for a in e.findall("a:author", ns)]
64+ year = int(e.find("a:published", ns).text[:4])
65+ return {"type": "arxiv", "title": title, "authors": [(n.split()[-1], " ".join(n.split()[:-1])) for n in authors], "year": year, "arxiv": aid, "doi": None}
66+def bib(key, m):
67+ A = " and ".join(f"{clean(f)}, {clean(g)}" if g else clean(f) for f, g in m["authors"]) or "Anonymous"
68+ T = clean(m["title"] + (": " + m["subtitle"] if m.get("subtitle") else ""))
69+ f = [f" title = {{{{{T}}}}}", f" author = {{{A}}}", f" year = {{{m['year']}}}"]
70+ t = m["type"]
71+ if t == "arxiv":
72+ et = "misc"; f += [f" eprint = {{{m['arxiv']}}}", " archiveprefix = {arXiv}", f" howpublished = {{arXiv preprint arXiv:{m['arxiv']}}}", " note = {Preprint, not peer reviewed}"]
73+ elif t == "journal-article":
74+ et = "article"; f += [f" journal = {{{clean(m['container'])}}}"]
75+ if m.get("volume"): f.append(f" volume = {{{m['volume']}}}")
76+ if m.get("issue"): f.append(f" number = {{{m['issue']}}}")
77+ if m.get("page"): f.append(f" pages = {{{clean(m['page']).replace('-', '--').replace('----', '--')}}}")
78+ elif t == "proceedings-article":
79+ et = "inproceedings"; f += [f" booktitle = {{{clean(m['container'] or m.get('event') or 'Proceedings')}}}"]
80+ if m.get("page"): f.append(f" pages = {{{clean(m['page']).replace('-', '--').replace('----', '--')}}}")
81+ if m.get("publisher"): f.append(f" publisher = {{{clean(m['publisher'])}}}")
82+ elif t == "book-chapter":
83+ et = "incollection"; f += [f" booktitle = {{{clean(m['container'])}}}"]
84+ if m.get("publisher"): f.append(f" publisher = {{{clean(m['publisher'])}}}")
85+ if m.get("page"): f.append(f" pages = {{{clean(m['page']).replace('-', '--').replace('----', '--')}}}")
86+ elif t == "posted-content":
87+ et = "misc"; f += [f" howpublished = {{Preprint ({clean(m.get('publisher') or 'repository')})}}", " note = {Preprint, not peer reviewed}"]
88+ elif t in ("book", "monograph", "edited-book", "reference-book"):
89+ et = "book"; f += [f" publisher = {{{clean(m.get('publisher') or '')}}}"]
90+ elif t == "report":
91+ et = "techreport"; f += [f" institution = {{{clean(m.get('publisher') or '')}}}"]
92+ else:
93+ et = "misc"; f += [f" howpublished = {{{clean(m.get('container') or m.get('publisher') or '')}}}"]
94+ if m.get("doi"): f.append(f" doi = {{{m['doi']}}}")
95+ return f"@{et}{{{key},\n" + ",\n".join(f) + "\n}\n"
96+entries, ver, keys = [], [], {}
97+for wid, spec in W.items():
98+ if spec is None: continue
99+ kw, ovr = spec; r = L[wid]
100+ ident = ovr or (f"doi:{r['doi']}" if r["doi"] else (f"arxiv:{r['arxiv_id']}" if r["arxiv_id"] else None))
101+ if not ident: print("NO IDENTIFIER", wid, r["title"][:60]); continue
102+ try:
103+ m = crossref(ident[4:]) if ident.startswith("doi:") else arxiv(ident[6:])
104+ except Exception as e:
105+ print("LOOKUP FAILED", wid, ident, e); continue
106+ surname = re.sub(r"[^a-z]", "", unicodedata.normalize("NFKD", m["authors"][0][0]).encode("ascii", "ignore").decode().lower()) if m["authors"] else "anon"
107+ key = f"{surname}{m['year']}{kw}"
108+ j = len(toks(r["title"]) & toks(m["title"])) / max(1, len(toks(r["title"]) | toks(m["title"])))
109+ match = "yes" if j >= 0.5 else f"check (jaccard={j:.2f})"
110+ keys[wid] = key; entries.append(bib(key, m))
111+ ver.append({"key": key, "openalex_id": wid, "doi_or_arxiv": ident, "verified_via": "crossref" if ident.startswith("doi:") else "arxiv", "verified_title": clean(m["title"])[:160], "corpus_title": clean(r["title"])[:160], "match": match, "entry_type": m["type"]})
112+ print(key, "|", match, "|", clean(m["title"])[:70]); time.sleep(0.3)
113+open(f"{R}/manuscript/references_part05.bib", "w").write("% references_part05.bib — Part V (The Artificial Intelligence Transformation), WP44. Generated by\n% literature_synthesis/agent_E/make_bib_part05.py on 2026-11-03: every entry resolved by one Crossref (DOI) or arXiv lookup;\n% see references_verification_part05.csv. Keys of manuscript/references.bib are not redefined here.\n\n" + "\n".join(entries))
114+with open(f"{R}/manuscript/references_verification_part05.csv", "w", newline="") as f:
115+ w = csv.DictWriter(f, fieldnames=list(ver[0].keys())); w.writeheader(); w.writerows(ver)
116+json.dump(keys, open(f"{R}/literature_synthesis/agent_E/bib_keys.json", "w"), indent=1)
117+print(len(entries), "entries written")
added literature_synthesis/agent_E/make_contributions.py +264 −0
@@ -0,0 +1,264 @@
1+#!/usr/bin/env python3
2+"""Agent E (AI Research Specialist), WP44 Phase 5 — contributions.csv for the ai_ml landmark set.
3+Reads literature_synthesis/landmarks/ai_ml_landmarks.csv + ai_ml_abstracts.jsonl; the judgements below were made by the
4+specialist agent from the abstract (judged_from=abstract) or from title/venue/type only (metadata_only). data_quality_flag:
5+ ok | identifier_mismatch (title/abstract do not match the arXiv id or the citation count attached to the record)
6+ | abstract_mismatch (abstract belongs to another work) | year_implausible (OpenAlex year contradicts the known first
7+ publication) | normalisation_artefact (tiny c36 but very large NCS36: young cohort / rare document type)
8+ | off_scope (classified into an AI category but not an AI contribution) | no_abstract (metadata only).
9+Usage: python3 literature_synthesis/agent_E/make_contributions.py <repo_root>"""
10+import csv, json, sys, os, unicodedata
11+R = sys.argv[1] if len(sys.argv) > 1 else "."
12+L = {r["work_id"]: r for r in csv.DictReader(open(f"{R}/literature_synthesis/landmarks/ai_ml_landmarks.csv"))}
13+A = {str(json.loads(l)["work_id"]): json.loads(l).get("abstract") or "" for l in open(f"{R}/literature_synthesis/landmarks/ai_ml_abstracts.jsonl")}
14+def asc(s): return unicodedata.normalize("NFKD", s).encode("ascii", "ignore").decode()
15+# work_id: (contribution_type, description, data_quality_flag)
16+J = {
17+"7172491899": ("engineering","Text-to-MongoDB query translation (GAT + RAG + T5); record carries arXiv 2005.11401 (RAG, Lewis et al.) and its citations","identifier_mismatch"),
18+"7154011589": ("empirical","Portuguese-language essay on LLM pre-training data and copyright; record carries arXiv 2005.14165 (GPT-3) and its citations","identifier_mismatch"),
19+"4362515116": ("survey","Survey of LLM techniques (pre-training, adaptation, utilisation, evaluation); journal version of the 2023 arXiv survey","ok"),
20+"4288253152": ("engineering","Dissertation: fine-tuned multilingual sentence-transformer for Dewey Decimal classification; NCS36 implausible for a dissertation","normalisation_artefact"),
21+"4322718191": ("engineering","LLaMA: 7B-65B open-weight foundation language models trained on public data; the abstract reports LLaMA-13B outperforms GPT-3 on most benchmarks","ok"),
22+"4323655724": ("survey","Position paper on opportunities and challenges of LLMs for education (student and teacher perspectives)","ok"),
23+"4367369313": ("survey","Practitioner guide to prompt engineering for ChatGPT","ok"),
24+"4226278401": ("method","Proposal of a 'companion curriculum' for LM training; record carries arXiv 2203.02155 (InstructGPT) and its citations","identifier_mismatch"),
25+"4396531244": ("engineering","Dissertation: multilingual semantic search + LLMs for news retrieval; NCS36 implausible for c36=20","normalisation_artefact"),
26+"4221143046": ("theory","Essay on 'theatrical compliance' failure mode; record carries arXiv 2201.11903 (chain-of-thought) and its citations","identifier_mismatch"),
27+"4327810158": ("engineering","Title = GPT-4 Technical Report (arXiv 2303.08774) but the stored abstract describes an unrelated 'MFOUR Vibe Framework'","abstract_mismatch"),
28+"7135096569": ("empirical","Preregistered field experiment (758 BCG consultants) on the uneven ('jagged') effect of GPT-4 assistance on knowledge-work tasks","ok"),
29+"4414281281": ("method","DeepSeek-R1: reasoning incentivised by pure reinforcement learning without human-annotated reasoning traces (Nature version)","ok"),
30+"4389437528": ("survey","Review of potential benefits of ChatGPT for teaching and learning","ok"),
31+"4292779060": ("theory","Speculative 'Aion Framework' on AI consciousness; c36=13,333 cannot belong to this record","identifier_mismatch"),
32+"4360836968": ("empirical","'Sparks of AGI': qualitative experiments on an early GPT-4; contested interpretation; preprint","ok"),
33+"4406800520": ("survey","Survey of LLM-based agents (brain/perception/action framework); journal version of arXiv 2309.07864","no_abstract"),
34+"4385223296": ("survey","Book chapter on responsible use of ChatGPT in education","no_abstract"),
35+"4287689466": ("empirical","LLM essay classification for writing placement (450 essays, four models); c36=722 implausible for an OASIcs paper","identifier_mismatch"),
36+"4366420437": ("survey","Rapid review of 50 early articles on ChatGPT in education (Dec 2022-Feb 2023)","ok"),
37+"4317910584": ("survey","Commentary on ChatGPT and assessment in higher education","ok"),
38+"4313564799": ("empirical","SSRN preprint on ChatGPT user experience in education","no_abstract"),
39+"4382618722": ("engineering","Chatlaw: multi-agent legal assistant with role-aligned mixture-of-experts (Chinese law)","ok"),
40+"4384464487": ("empirical","Survey of 399 Hong Kong students on generative AI perceptions","ok"),
41+"4392773210": ("survey","Discussion of AI literacy, prompt engineering and critical thinking in classrooms","ok"),
42+"4387355843": ("method","Time-LLM: reprogramming LLMs for time-series forecasting (book-chapter record of the ICLR 2024 paper)","ok"),
43+"4392947521": ("method","UNI: general-purpose self-supervised foundation model for computational pathology","no_abstract"),
44+"7124890304": ("method","Semi-supervised cross-domain fault diagnosis via contrastive pre-training","no_abstract"),
45+"4410551352": ("method","Aurora: foundation model for the Earth system trained on >1M hours of geophysical data; abstract reports outperforming operational forecasts","ok"),
46+"4322832366": ("method","TimeMAE: decoupled masked autoencoders for time-series representation","ok"),
47+"7133233472": ("method","QLoRA: 4-bit quantised base + low-rank adapters; 65B model finetuned on one 48 GB GPU","ok"),
48+"2990704537": ("benchmark","SuperGLUE: harder successor to GLUE after models exceeded non-expert human level","ok"),
49+"7125271461": ("engineering","ChemBERTa-3: open training framework for chemical foundation models","ok"),
50+"7119541039": ("theory","Position paper defining eleven desirable features of Earth foundation models","ok"),
51+"4393138529": ("survey","Survey of parameter-efficient fine-tuning (PEFT) for large models; preprint","ok"),
52+"4390690043": ("survey","Survey of data-centric foundation-model approaches in computational healthcare","ok"),
53+"7128029160": ("method","BrainIAC: self-supervised foundation model for brain MRI","ok"),
54+"4402220065": ("method","Pathology foundation model for cancer diagnosis and prognosis (Nature 2024)","no_abstract"),
55+"7118170538": ("survey","Review of AI applications in fisheries","ok"),
56+"4411932262": ("method","Centaur: language model fine-tuned on Psych-101 (10M choices) predicting human behaviour across experiments","ok"),
57+"4411099958": ("method","PanDerm: multimodal dermatology foundation model pretrained on 2M images, 28 benchmarks","ok"),
58+"4361229539": ("method","LLaMA-Adapter: 1.2M-parameter adaption prompts with zero-init attention for instruction following; preprint","ok"),
59+"4407135816": ("survey","Narrative review of foundation models for radiology","ok"),
60+"4385948838": ("method","PLIP: visual-language pathology foundation model trained on medical Twitter data","no_abstract"),
61+"4406643700": ("survey","Quantitative literature analysis and agenda for geospatial AI","ok"),
62+"4406472599": ("survey","Perspective on 'learning from models' beyond fine-tuning","no_abstract"),
63+"4393153153": ("method","TimesURL: self-supervised contrastive universal time-series representations","ok"),
64+"4402727669": ("method","mPLUG-Owl2: modality-collaborative multimodal LLM with shared modules","ok"),
65+"4401397240": ("method","BiomedGPT: generalist vision-language foundation model for biomedical tasks","no_abstract"),
66+"4385285835": ("method","Masked image modelling at scale for histopathology; medRxiv preprint","ok"),
67+"4318718936": ("method","BLIP-2: Q-Former bridging frozen image encoders and frozen LLMs; preprint","ok"),
68+"4393148714": ("method","T2I-Adapter: lightweight adapters adding structural/colour control to text-to-image diffusion","ok"),
69+"4396599905": ("survey","Survey of multimodal fusion under low-quality data (noise, missing modalities, imbalance, dynamics)","ok"),
70+"4396952261": ("survey","Textbook on multimodal deep learning","no_abstract"),
71+"7133250524": ("benchmark","LAION-5B: open 5.85B image-text pair dataset for training CLIP-style models","no_abstract"),
72+"4392947532": ("method","CONCH: visual-language foundation model for computational pathology","no_abstract"),
73+"7138474371": ("method","OpenDriveVLA: vision-language-action model for end-to-end driving on open LLMs","ok"),
74+"7127319731": ("method","Med-R1: RL-enhanced vision-language model for medical reasoning","ok"),
75+"2619383789": ("survey","Taxonomy of multimodal machine learning (representation, translation, alignment, fusion, co-learning)","ok"),
76+"7148475777": ("empirical","Zero-shot evaluation of GPT-5 on multimodal medical QA/VQA","ok"),
77+"7160627408": ("method","Generative cross-domain knowledge synthesis framework (conference paper)","ok"),
78+"2995460200": ("method","VL-BERT: single-stream Transformer pre-trained on image regions and words","ok"),
79+"4415913530": ("method","TITAN: multimodal whole-slide pathology foundation model (335,645 WSIs)","ok"),
80+"4289785095": ("method","Textual inversion: personalising text-to-image diffusion with learned pseudo-words from 3-5 images; preprint","ok"),
81+"7125233009": ("method","SemCap: sentiment-aware captions as visual proxy for multimodal aspect-based sentiment analysis","ok"),
82+"7143512276": ("survey","Survey of multimodal ML in predictive healthcare analytics","ok"),
83+"4413447271": ("method","RadFM: generalist radiology foundation model on 2D/3D data (MedMD, 13M images)","ok"),
84+"7131830459": ("survey","Survey of multimodal recommender systems","ok"),
85+"7206206818": ("method","Lingshu: generalist medical multimodal LLM","ok"),
86+"7133220561": ("method","InstructBLIP: instruction tuning of vision-language models","no_abstract"),
87+"7164501397": ("empirical","Evaluation of 8 multimodal foundation models on 1,090 medical cases: image predictions largely driven by text","ok"),
88+"7165561001": ("method","Hierarchical graph-Mamba multi-agent clinical architecture","no_abstract"),
89+"7138470740": ("method","CoEvo: training-free test-time OOD detection with vision-language models","ok"),
90+"7138363754": ("method","AdaptCLIP: adapting CLIP for universal visual anomaly detection","ok"),
91+"4410882873": ("survey","Scoping review of 432 multimodal medical AI papers (2018-2024)","ok"),
92+"4400228804": ("empirical","Dissertation on generative AI in animation production; c36=2 with NCS36=540 is a normalisation artefact","normalisation_artefact"),
93+"7124262017": ("survey","OECD Digital Education Outlook 2026 (policy report on generative AI in education)","ok"),
94+"4406472463": ("method","MatterGen: diffusion model generating stable inorganic materials under property constraints","ok"),
95+"7123342962": ("method","Text-guided optimal transport for unpaired rain removal","ok"),
96+"4383312437": ("empirical","AI Ecological Education Policy Framework from 457 students and 180 staff (Hong Kong)","ok"),
97+"7133185348": ("method","Imagen: photorealistic text-to-image diffusion with large frozen text encoders","no_abstract"),
98+"3195296536": ("method","Dissertation on GAN-based adversarial defence","no_abstract"),
99+"4322617812": ("theory","Paradox-theory essay on generative AI in education","ok"),
100+"3096831136": ("survey","CACM review of generative adversarial networks","ok"),
101+"4400770292": ("empirical","Field experiment: GPT-4 tutoring access improved practice but harmed unassisted exam performance; SSRN preprint","no_abstract"),
102+"7127274868": ("empirical","Survey study of parental AI investment and informal English learning","ok"),
103+"7124974472": ("empirical","Nationally representative US surveys of generative-AI adoption (45% of adults 18-64 by late 2024)","ok"),
104+"7118187240": ("engineering","PatchTST-based photovoltaic forecasting with GAN features (application)","off_scope"),
105+"4406199489": ("survey","Review of generative AI and integrity in higher education","ok"),
106+"4398243313": ("survey","Scoping review of 32 studies on generative AI and assessment","ok"),
107+"4409269614": ("survey","Systematic review (41 studies) of generative AI and academic integrity","ok"),
108+"3193917007": ("survey","Textbook chapter on GANs","ok"),
109+"4288099666": ("method","Classifier-free diffusion guidance: joint conditional/unconditional training; preprint","ok"),
110+"4394934695": ("empirical","Educator interviews on generative AI in higher education","ok"),
111+"4321648656": ("survey","Review of neural architectures for AI-generated art","ok"),
112+"4395957952": ("survey","UNESCO guidance for generative AI in education and research","no_abstract"),
113+"4394863715": ("empirical","SSRN preprint on generative AI and human creativity","no_abstract"),
114+"4407778399": ("empirical","Analysis of generative-AI guidelines at 50 US universities","ok"),
115+"7124432017": ("theory","Intelligent-TPACK pedagogy framework","ok"),
116+"7140170686": ("empirical","Mixed-methods study of human-GenAI pedagogical partnerships","ok"),
117+"4406728221": ("survey","Survey of agentic AI concepts and applications","ok"),
118+"7161734976": ("engineering","Robin: multi-agent system automating hypothesis generation and data analysis for experimental biology","ok"),
119+"7135409568": ("survey","Scoping review of agentic AI in healthcare","ok"),
120+"7134187809": ("theory","Event-triggered optimal consensus control for nonlinear multi-agent systems (control theory)","off_scope"),
121+"4390730779": ("survey","Survey of LLM-based agents: definitions, methods, prospects","ok"),
122+"4225355126": ("survey","Textbook introduction to multi-agent systems","no_abstract"),
123+"4407570917": ("engineering","AgentSociety: large-scale simulation of LLM-driven generative social agents","ok"),
124+"4410197190": ("survey","Narrative review of agentic AI in organisations","ok"),
125+"4413427262": ("survey","Conceptual taxonomy distinguishing AI agents from agentic AI","ok"),
126+"4414991289": ("engineering","DIVE: multi-agent workflow extracting experimental data from figures for hydrogen-storage materials","ok"),
127+"7122857583": ("theory","Conceptual agentic-AI framework (literature synthesis)","ok"),
128+"4401023556": ("survey","Survey of LLM-based multi-agent systems","ok"),
129+"4414023998": ("survey","Review of 143 primary studies on agentic AI systems","ok"),
130+"7118083939": ("theory","Conceptual overview of agentic information systems","ok"),
131+"4412505619": ("survey","Duplicate record of the AI agents vs agentic AI taxonomy (different venue)","ok"),
132+"4416209823": ("survey","PRISMA-based review (90 studies) with symbolic vs neural dual-paradigm framework for agentic AI","ok"),
133+"7155930820": ("theory","Prescribed-performance consensus control for multi-agent systems (control theory)","off_scope"),
134+"7162257503": ("theory","Predefined-time sliding-mode formation control (control theory)","off_scope"),
135+"4412709685": ("engineering","Virtual Lab: LLM agents designing SARS-CoV-2 nanobodies with experimental validation","no_abstract"),
136+"7128165235": ("theory","Five-component framework for multi-agent AI in knowledge work","ok"),
137+"7118190454": ("survey","Perspective on agentic AI for elderly care","ok"),
138+"7130481397": ("benchmark","Benchmarking two agent systems (OpenManus, Manus) on clinical decision tasks","ok"),
139+"4390638955": ("survey","Scoping review of AI learning tools in K-12 education","off_scope"),
140+"4412203333": ("survey","Survey of memory mechanisms of LLM-based agents","ok"),
141+"4399982941": ("theory","Dissertation on value engineering for autonomous agents (normative MAS)","normalisation_artefact"),
142+"2911283634": ("survey","Textbook chapter on reinforcement learning","no_abstract"),
143+"4389473504": ("method","Dissertation on trust-region policy optimisation","ok"),
144+"2964083594": ("empirical","Large-scale study of curiosity-driven learning across 54 environments (ICLR 2019; OpenAlex year 2025 is a repository copy)","year_implausible"),
145+"2098774185": ("method","Maximum-entropy inverse reinforcement learning (AAAI 2008; OpenAlex year 2018)","year_implausible"),
146+"1591675293": ("method","Bayesian inverse reinforcement learning (IJCAI 2007; OpenAlex year 2025 book chapter)","year_implausible"),
147+"7123365826": ("theory","Event-triggered zero-sum game control for saturated nonlinear systems (control theory)","off_scope"),
148+"2991046523": ("survey","Selective overview of multi-agent RL theory and algorithms","no_abstract"),
149+"2973229164": ("method","Dissertation on deep RL for continuous robot control","ok"),
150+"2982316857": ("engineering","AlphaStar: grandmaster level in StarCraft II with multi-agent RL","no_abstract"),
151+"2902907165": ("method","AlphaZero: one self-play RL algorithm for chess, shogi and Go","ok"),
152+"7133481537": ("theory","Predefined-time fault-tolerant consensus control (control theory)","off_scope"),
153+"7133226216": ("empirical","MAPPO: PPO is surprisingly strong in cooperative multi-agent benchmarks","ok"),
154+"3041807988": ("theory","Critical essay on ed-tech solutionism after Covid-19 (not RL)","off_scope"),
155+"2781726626": ("method","Soft actor-critic: off-policy maximum-entropy RL; preprint record","ok"),
156+"2963923407": ("method","TD3: clipped double-Q and delayed policy updates against overestimation","ok"),
157+"7122543576": ("theory","ADP-based event-triggered fault-tolerant control (control theory)","off_scope"),
158+"2899138465": ("method","Dissertation on gradient-based meta-learning (MAML lineage)","ok"),
159+"4391631327": ("empirical","Pre-registered toy study of two-stage post-training; record carries arXiv 2402.03300 (DeepSeekMath) and its citations","identifier_mismatch"),
160+"4409147637": ("method","DreamerV3: world-model RL mastering 150+ tasks with one configuration (Nature)","ok"),
161+"2895974005": ("empirical","Master's thesis on exploration/exploitation parameters","normalisation_artefact"),
162+"7134057781": ("engineering","Fuzzy-TD3 hybrid control for a 5-DOF manipulator (application)","ok"),
163+"2754517384": ("empirical","Deep RL That Matters: reproducibility and variance of deep RL results","ok"),
164+"2761873684": ("method","Rainbow: combining six DQN extensions, with ablations","ok"),
165+"7133208539": ("method","Direct preference optimisation: closed-form RLHF objective without a separate reward model","no_abstract"),
166+"7128436456": ("survey","Survey classifying combinations of MPC and RL","ok"),
167+"2962862931": ("method","Interpretability measure for degradation diagnosis (HAL); c36=12,586 cannot belong to this record","identifier_mismatch"),
168+"1698155719": ("survey","Textbook chapter on support vector machines","ok"),
169+"4288283362": ("method","'Spiral Time' geometric time encoding with a small LSTM ablation; no venue; NCS36 implausible","normalisation_artefact"),
170+"46659105": ("theory","Title = Powers (2011) evaluation-measures paper (arXiv 2010.16061) but the stored abstract is an LLM-as-judge study of program repair","abstract_mismatch"),
171+"3145506661": ("survey","Textbook: Introduction to Machine Learning (Springer Briefs)","no_abstract"),
172+"2291445757": ("survey","Textbook chapter on support vector regression","ok"),
173+"2100415632": ("theory","Bandits with side observations (2014 paper carrying a 2026 arXiv id and year)","year_implausible"),
174+"1568555062": ("theory","Verma & Pearl: equivalence and synthesis of causal models (reprint chapter, 2022)","ok"),
175+"2751318774": ("survey","Textbook: Introduction to Machine Learning (Alpaydin)","ok"),
176+"3208688400": ("survey","Anomaly detection survey (book chapter)","no_abstract"),
177+"3105488314": ("theory","Dissertation on risk bounds for statistical learning","no_abstract"),
178+"57706852": ("theory","StoSOO stochastic optimistic optimisation (2013 paper carrying a 2026 arXiv id and year)","year_implausible"),
179+"2946444988": ("method","Deep Embedded Validation for UDA model selection (2019 paper carrying a 2026 arXiv id; NCS36 implausible)","year_implausible"),
180+"2945976633": ("theory","Rudin: argument for interpretable models over post-hoc explanation in high-stakes decisions","no_abstract"),
181+"4400823820": ("survey","Overview of random forests (non-indexed journal)","ok"),
182+"2999309192": ("theory","Advantages of MCC over F1 and accuracy for imbalanced binary classification","ok"),
183+"2999615587": ("method","TreeExplainer: exact SHAP values for tree ensembles (NMI)","no_abstract"),
184+"2439568532": ("theory","Lipton: the mythos of model interpretability","ok"),
185+"3194464139": ("engineering","Indonesian thesis applying kNN to credit risk; NCS36 implausible for c36=21","normalisation_artefact"),
186+"3041133507": ("survey","Comprehensive survey of transfer learning (Proc. IEEE)","ok"),
187+"2436219157": ("survey","Bottou, Curtis, Nocedal: optimisation methods for large-scale ML (SIAM Review)","ok"),
188+"3135028703": ("survey","Overview of ML algorithms and applications (SN Computer Science)","no_abstract"),
189+"7124426637": ("survey","Review of machine learning with small and limited data","ok"),
190+"2165856466": ("theory","Spectral bandits for smooth graph functions (2014 paper carrying a 2026 arXiv id and year)","year_implausible"),
191+"4402843978": ("survey","UK POSTnote on interpretable ML","ok"),
192+"2964121744": ("engineering","'AutoResearch Lite' Adam reproduction runner; c36=45,775 and arXiv 2108.04578 cannot belong to this record","identifier_mismatch"),
193+"1710476689": ("survey","Textbook chapter on GANs (The Science of Deep Learning)","ok"),
194+"4322614756": ("method","Path-based embedding for road-network community detection; c36=1,542 implausible for a LIPIcs paper","identifier_mismatch"),
195+"2963399829": ("method","mixup: training on convex combinations of examples (ICLR 2018; OpenAlex year 2024)","year_implausible"),
196+"4382203079": ("empirical","Zeng et al.: simple linear models beat Transformers on long-term time-series forecasting benchmarks","ok"),
197+"4406170795": ("method","TabPFN: tabular foundation model (prior-data fitted network) for small data (Nature)","ok"),
198+"146900863": ("survey","NeurIPS proceedings volume record (MIT Press)","no_abstract"),
199+"4396821501": ("method","KAN: Kolmogorov-Arnold networks with learnable spline edge functions; preprint","ok"),
200+"2950067852": ("method","Variational RNN (NeurIPS 2015) recorded as a 2026 preprint; NCS36 implausible","year_implausible"),
201+"4221150249": ("survey","Survey of GNNs for heterophilic graphs","ok"),
202+"7148890333": ("method","Multi-scale graph transfer learning for rotating-machinery diagnosis","no_abstract"),
203+"3177318507": ("method","Informer: ProbSparse attention for long-sequence forecasting","ok"),
204+"4389326242": ("method","Mamba: selective state-space model with linear-time sequence modelling; preprint","ok"),
205+"7118302872": ("method","GNN + ant colony optimisation for influence maximisation","ok"),
206+"3137147200": ("survey","Survey of quantisation methods for efficient inference (book chapter)","ok"),
207+"3140854437": ("survey","Review of deep learning concepts and CNN architectures","ok"),
208+"4210257598": ("survey","Comprehensive survey on GNNs (repository duplicate of the TNNLS paper)","ok"),
209+"2907492528": ("survey","Comprehensive survey on GNNs (TNNLS)","ok"),
210+"3152893301": ("survey","GNNs: review of methods and applications (AI Open)","ok"),
211+"7110333790": ("method","CaLLiPer+: POI-name-enriched location embeddings; c36=496 implausible for a LIPIcs paper","identifier_mismatch"),
212+"4385763767": ("survey","Survey of Transformers in time series","ok"),
213+"3124006607": ("method","Simple spectral graph convolution (ICLR 2021; OpenAlex year 2026)","year_implausible"),
214+"4285723986": ("survey","Synthesis lecture on graph attention networks","no_abstract"),
215+"2892035503": ("survey","Review of deep learning for time-series classification","no_abstract"),
216+"2962776659": ("empirical","Auxiliary-task analysis of sentence embeddings (ICLR 2017; OpenAlex year 2024)","year_implausible"),
217+"4406063062": ("survey","Review and research framework for responsible AI governance (IS)","ok"),
218+"7130596601": ("survey","Review of human-in-the-loop AI in healthcare","ok"),
219+"3084163594": ("survey","Book review of Zuboff's Surveillance Capitalism for educators","off_scope"),
220+"2963808661": ("engineering","Governance risk-assessment tooling (OASIcs practitioner track); c36=390 implausible","identifier_mismatch"),
221+"7133236347": ("survey","TRiSM (trust, risk, security management) for LLM-based multi-agent systems","ok"),
222+"7128782475": ("empirical","Physician-led red-teaming of four chatbots on 222 patient questions (HealthAdvice)","ok"),
223+"4406571986": ("empirical","Analysis of generative-AI policies at 116 US R1 universities","ok"),
224+"2981731882": ("survey","Arrieta et al.: XAI concepts, taxonomies and responsible AI","no_abstract"),
225+"4402698709": ("survey","AI Risk Repository: 1,725 risks from 74 taxonomies with causal and domain taxonomies","ok"),
226+"4408314709": ("survey","Review of bias recognition and mitigation in healthcare AI","ok"),
227+"4407243994": ("theory","Comparative ethical framework (EU/US/China policies)","ok"),
228+"4330336443": ("empirical","Title = 'GPTs are GPTs' (arXiv 2303.10130) but the stored abstract describes a TRIPS/Anthropic Economic Index comparison","abstract_mismatch"),
229+"7128600296": ("survey","Systematic review of 35 healthcare-AI governance frameworks and a maturity model","ok"),
230+"4399450035": ("empirical","Power Hungry Processing: measured inference energy of task-specific vs general-purpose models","ok"),
231+"4318586239": ("survey","Policy book chapter defining AI","ok"),
232+"4360615722": ("theory","Qualitative examination of ethical implications of chatbots in education","ok"),
233+"2670253439": ("survey","Miller: explanation in AI through the lens of the social sciences","no_abstract"),
234+"7118631510": ("method","Ante-hoc interpretability for Industry 5.0","no_abstract"),
235+"7127143535": ("empirical","Autonomous AI agents in industrial information integration (paradoxical effects)","no_abstract"),
236+"4409249263": ("survey","Ethical challenges of AI in clinical practice","ok"),
237+"4366262984": ("survey","XAI: what is known and what remains for trustworthy AI (Information Fusion)","ok"),
238+"2969896603": ("survey","Survey on bias and fairness in machine learning (CSUR)","ok"),
239+"4406152291": ("empirical","Data-poisoning threat assessment: 0.001% of tokens with medical misinformation yields harmful LLMs","ok"),
240+"4410609100": ("method","PAIR: black-box semantic jailbreaks in about twenty queries","ok"),
241+"4411542009": ("empirical","Experiments showing LLMs express stigma and respond inappropriately as replacement therapists","ok"),
242+"4360620450": ("survey","Dwivedi et al.: 43 expert contributions on ChatGPT opportunities and challenges","ok"),
243+"3185341429": ("survey","Liu et al.: systematic survey of prompt-based learning","ok"),
244+"4384071683": ("benchmark","Med-PaLM: MultiMedQA benchmark and human evaluation framework for clinical LLM answers","ok"),
245+"4384918448": ("empirical","Longitudinal ICU simulation of 23 open-weight models; record carries arXiv 2307.09288 (Llama 2) and its citations","identifier_mismatch"),
246+"4393065402": ("survey","Wang et al.: survey of LLM-based autonomous agents","ok"),
247+"4319083882": ("theory","van Dis et al.: five priorities for research on ChatGPT (Nature comment)","no_abstract"),
248+"4399790131": ("survey","Systematic review of over-reliance on AI dialogue systems and cognition","ok"),
249+"3133702157": ("theory","Bender et al.: risks of ever-larger language models ('stochastic parrots')","ok"),
250+"4321499901": ("empirical","Qualitative case study of ChatGPT among early adopters in education","ok"),
251+}
252+rows = []
253+for wid, r in L.items():
254+ ct, desc, flag = J[wid]
255+ has = bool(A.get(wid, "").strip())
256+ if flag == "ok" and not has: flag = "no_abstract"
257+ rows.append({"openalex_id": wid, "title": asc(r["title"])[:140], "year": r["year"], "type": r["type"], "primary_category": r["primary_category"], "doi": r["doi"], "arxiv_id": r["arxiv_id"], "c36": r["c36"], "ncs36": round(float(r["ncs36"]), 1),
258+ "contribution_type": ct, "description": desc, "judged_from": "abstract" if has else "metadata_only", "data_quality_flag": flag})
259+missing = set(L) - set(J); assert not missing, missing
260+out = f"{R}/literature_synthesis/agent_E/contributions.csv"
261+with open(out, "w", newline="") as f:
262+ w = csv.DictWriter(f, fieldnames=list(rows[0].keys())); w.writeheader(); w.writerows(rows)
263+from collections import Counter
264+print(len(rows), "rows ->", out); print(Counter(r["data_quality_flag"] for r in rows)); print(Counter(r["contribution_type"] for r in rows)); print(Counter(r["judged_from"] for r in rows))
added literature_synthesis/agent_H/review.md +111 −0
@@ -0,0 +1,111 @@
1+# Agent H — Security, privacy and cryptography: disciplinary synthesis (Phase 5, 2026-11-03)
2+
3+Area: `security_privacy` (categories `cybersecurity`, `applied_cryptography`, `blockchain`, `privacy_preserving_computation`,
4+`federated_learning`). Inputs: `literature_synthesis/landmarks/security_privacy_landmarks.csv` (125 works, 25 per category) and
5+`security_privacy_abstracts.jsonl` (92 abstracts); `results/graph/category_counts_yearly.csv`, `category_flows_yearly.csv`,
6+`category_cooccurrence_yearly.csv`; `results/bibliometrics/cagr_by_category.csv`; `results/econometrics/breaks.csv` and
7+`its_results.json` (rows `cat_*`). No knowledge-graph extraction file exists for this area (`data/processed/graph/extraction/` is
8+empty) and no emerging-topic row concerns these categories, so every statement below rests on metadata, abstracts, or corpus
9+statistics; nothing rests on a full text. Numbers are quoted here for the record and enter the manuscript only through the
10+`\Sec...` macros appended to `manuscript/tables/numbers.tex` by `tables/make_numbers.py` and the four tables
11+`tab_security_counts`, `tab_security_summary`, `tab_security_llm_cooccurrence`, `tab_security_landmarks` (`tables/make_tables.py`).
12+
13+## 0. Reading rules and what the inputs can and cannot support
14+
15+The landmark set of this area is an automatic selection (`literature_synthesis/select_landmarks.py`): the top works by
16+field-normalised 36-month citations (NCS36) within each primary category, with 2025–2026 works ranked by 12-month citations
17+relative to their cohort. Three consequences shape everything that follows. First, the selection is blind to record quality:
18+duplicated, misdated and misclassified records rank as high as genuine landmarks when OpenAlex attaches citations to them.
19+Second, the NCS36 criterion favours survey and application papers in categories whose citing community is broad (blockchain,
20+federated learning), and favours standards and reports in categories where the influential documents are not journal
21+articles (applied cryptography). Third, 33 of the 125 records are dated 2026 and ranked on a few months of citations; their
22+normalised scores (up to 9,806 for an SSRN preprint without abstract) measure cohort position, not durable influence. The
23+synthesis therefore treats the landmark list as a *sample of what is highly cited under the corpus's own metric*, documents
24+its artefacts explicitly (Section 7), and complements it with the verified references of `manuscript/references.bib` where
25+the automatic selection missed works that any specialist would consider foundational (NIST FIPS 205, prompt-injection and
26+jailbreak papers, DP-SGD, secure aggregation). Abstract-level evidence is marked "the abstract reports"; works without
27+abstract (33 of 125) are discussed from their metadata only.
28+
29+## 1. Volume and structural change across the five categories (corpus statistics)
30+
31+Counts are eligible works carrying the category (multi-label) per publication year, summed over the pre/baseline/post labels
32+of `category_counts_yearly.csv`; primary-category counts are reported alongside in `tab_security_counts`. Cybersecurity is the
33+largest category of the area (78,082 works in 2022; 139,134 in 2025; 829,538 over 2018–2026), applied cryptography the second
34+(29,699 in 2018; 51,759 in 2025), blockchain (10,872 → 33,426), privacy-preserving computation (11,378 → 49,266) and federated
35+learning (1,040 → 18,093) follow. Growth profiles differ sharply. Cybersecurity and applied cryptography were nearly flat over
36+2018–2022 (multi-label CAGR 3.7 % and 1.9 %) and accelerated after 2023 (24.7 % and 20.4 % over 2023–2025). Privacy-preserving
37+computation grew throughout (15.2 % then 39.8 %). Federated learning had the steepest baseline growth of the area (54.3 %) and
38+remained the fastest-growing category after 2023 (45.8 %), but its growth rate fell. Blockchain is the exception: its growth was
39+front-loaded (16.7 % baseline, 17.3 % post, with the 2020 Bai–Perron break), and its share of primary classifications peaked
40+around 2020 (1.76 %) before settling (2.02 % in 2025 on a larger base; `category_share_ci_yearly.csv` places the 2023 share
41+above the 2025 one when confidence intervals are read).
42+
43+The interrupted-time-series estimates at the 2022-12 reference month (`breaks.csv`, HAC p-values) show no level shift for any
44+of the five categories except federated learning (level −0.126, p = 0.007), and positive slope changes for cybersecurity
45+(+0.015 per month, p < 0.001), applied cryptography (+0.013), privacy-preserving computation (+0.012) and none for blockchain
46+(−0.002, p = 0.20) or federated learning (−0.006, p = 0.001, i.e. a deceleration). The placebo distribution over 65 alternative
47+event months qualifies these: the slope t-statistic at 2022-12 ranks 3rd of 65 for applied cryptography (placebo p = 0.046),
48+4th for cybersecurity (0.062), 9th for privacy-preserving computation (0.138), and is unremarkable for blockchain (0.508) and
49+federated learning (0.385); the federated-learning level drop is likewise unremarkable against placebos (0.246). Bai–Perron
50+(BIC) dates a single break at 2022-07 for cybersecurity, applied cryptography and privacy-preserving computation — five months
51+*before* the reference event — and at 2020-04 (blockchain) and 2020-02 (federated learning). PELT places changes at 2022-11
52+and 2024-12 for the two security categories, at 2020-12 and 2023-12 for privacy, and at 2020-02 and 2023-05 for federated
53+learning. The donut specification (three months excluded on each side) turns the cybersecurity and applied-cryptography
54+level estimates positive and significant (+0.051, p = 0.049; +0.066, p = 0.001) while leaving slopes unchanged, which indicates
55+that the months around the reference date are themselves unusual (indexing, not necessarily publication). Alternative event
56+dates tell the same story: the o1 date (2024-10) carries a large positive level coefficient for cybersecurity (+0.149, p = 0.001),
57+applied cryptography (+0.142) and privacy (+0.198), consistent with a late-2024 acceleration that PELT also detects (2024-12).
58+
59+The reading that these results support is modest. The two "security" categories and privacy-preserving computation grew faster
60+after 2023 than before, the acceleration is dated by the break detectors to mid-2022 and to late 2024 rather than to a single
61+month, and the growth is partly compositional (Section 6: the share of their works co-classified with the LLM category rose from
62+1.7 % in 2022 to 7.6 % in 2025 for cybersecurity and from 1.4 % to 9.3 % for privacy). Blockchain and federated learning are
63+stories of the baseline period: both broke upward in early 2020 and neither shows a post-2022 acceleration. None of this is
64+causal evidence about ChatGPT; the manuscript's structural-change design (Part IV) is explicit that the reference month is a
65+convention.
66+
67+## 2. Cybersecurity
68+
69+**What the landmark set contains.** The 25 cybersecurity landmarks are a heterogeneous set: ten are grey literature (five
70+dissertations, four book chapters, one report), five have no DOI, and nine are dated 2026. The highest-ranked record is a 2026
71+SSRN preprint on persistent homology as a prompt-injection defence (no abstract; NCS36 9,806 on 252 citations within months),
72+which cannot be assessed from metadata and is flagged for PI review. Two records are misclassified: *Fedformer* (a long-term
73+time-series forecasting Transformer, reissued as a 2026 CRC chapter) and a 2018 book chapter on *Enigma* (a secure-computation
74+platform that belongs to privacy-preserving computation). A 2023 IT Professional record titled "IEEE Symposium on Security and
75+Privacy" (853 citations in 36 months, no abstract) is a venue-level record, not a paper. These artefacts are consistent with the
76+Part III finding that OpenAlex attaches citations to reprints and container records.
77+
78+**Pre-window landmarks.** The genuine baseline landmarks are the adversarial-examples chapter of Kurakin, Goodfellow and Bengio
79+(a 2018 reprint of the 2017 workshop paper; the abstract reports that adversarial examples survive printing and camera capture,
80+i.e. transfer to the physical world), the CIC-IDS2017 intrusion-detection dataset paper (no abstract; 534 citations in 36 months,
81+the single most reused benchmark of the network-intrusion literature), two 2018 IEEE surveys on blockchain for IoT and on the
82+security and privacy of Bitcoin (both abstracts describe surveys), and a 2018 healthcare-cloud blockchain position paper. The
83+baseline cybersecurity literature, as sampled by NCS36, was thus dominated by (i) machine-learning-based detection on shared
84+datasets and (ii) blockchain-as-security-solution surveys — not by the microarchitectural attacks (Spectre, Meltdown) that the
85+verified reference list records as the field's 2018 landmarks; those are classified in computer architecture or receive their
86+citations outside the 36-month window.
87+
88+**Post-window landmarks and the LLM-security turn.** The post-2023 landmarks with abstracts cluster on large language models
89+used *for* security and on the security *of* LLM systems. On the first side: *vul-rag* (ACM TOSEM 2026; the abstract reports
90+that LLMs distinguish vulnerable from patched code with 0.06–0.14 accuracy and that knowledge-level retrieval raises accuracy by
91+16–24 points), *HuntGPT* (anomaly detection with explainability and an LLM front-end), and a RoBERTa-based phishing detector
92+(Computer Networks 2026). On the second side: the ACM Computing Surveys survey of security and privacy challenges of LLMs
93+(jailbreaking, data poisoning, PII leakage; `das2025llmsecurity`), and the ACM TOSEM study of the Model Context Protocol
94+(a lifecycle of MCP servers in four phases and sixteen activities with a threat taxonomy; `hou2026mcp`). The verified reference
95+list adds the works that the automatic selection missed because they are preprints or workshop papers: indirect prompt injection
96+(`greshake2023injection`), prompt-injection attacks on integrated applications (`liu2023promptinjection`, preprint), the
97+"ignore previous prompt" attack (`perez2022ignore`), the jailbreak analysis of Wei et al. (`wei2023jailbroken`, preprint), the
98+universal adversarial suffix (`zou2023universal`, preprint), the one-day-exploit agent study (`fang2024oneday`, preprint),
99+PentestGPT (`deng2023pentestgpt`, preprint) and the surveys of Yao et al. and Shayegani et al. The corpus statistics locate this
100+turn: works co-classified cybersecurity × LLM rose from 1,337 (2022) to 10,549 (2025) and 14,410 in the incomplete 2026, i.e. from
101+1.7 % to 7.6 % and 12.0 % of cybersecurity works; co-classification with AI agents reached 5.3 % in 2026. Citation flows show the
102+same thing more weakly: 0.7 % of references from cybersecurity works went to LLM-category works in 2025 (1.1 % in 2026), against
103+47.5 % to cybersecurity itself.
104+
105+**Assessment.** Established: the security-of-ML literature predates the window (adversarial examples, membership inference,
106+training-data extraction), and the post-2022 LLM-security literature extends it to a new attack surface (instructions in data,
107+tool use, agent protocols). Contested: whether LLM-based detection improves on fine-tuned encoders or classical features at equal
108+cost — the vul-rag abstract itself reports very low base accuracy, and the phishing and HuntGPT abstracts report gains on
109+specific datasets. Speculative: claims that defences against prompt injection can be made robust; the only landmark on the
110+question (the 2026 SSRN preprint) argues a negative result and has not been reviewed. The corpus cannot yet tell whether the
111+LLM-security literature is a durable subfield or a labelling wave; its first 36-month citation windows close in 2026–2027.
modified manuscript/figures/fig_category_growth.pdf +0 −0

Binary file not shown.

added manuscript/references_part06.bib +712 −0
@@ -0,0 +1,712 @@
1+% references_part06.bib — WP44 Part VI (Agent F, Systems Specialist). Every entry verified 2026-11-03 by a Crossref or arXiv
2+% single lookup (see references_verification_part06.csv). Keys already in references.bib are not redefined here.
3+
4+@inproceedings{chen2024rcacopilot,
5+ title = {Automatic Root Cause Analysis via Large Language Models for Cloud Incidents},
6+ author = {Yinfang Chen and Huaibing Xie and Minghua Ma and Yu Kang and Xin Gao and Liu Shi and Yunjie Cao and Xuedong Gao and Hao Fan and Ming Wen and Jun Zeng and Supriyo Ghosh and Xuchao Zhang and Chaoyun Zhang and Qingwei Lin and Saravan Rajmohan and Dongmei Zhang and Tianyin Xu},
7+ booktitle = {Proceedings of the Nineteenth European Conference on Computer Systems (EuroSys '24)},
8+ year = {2024},
9+ pages = {674--688},
10+ publisher = {ACM},
11+ doi = {10.1145/3627703.3629553},
12+ note = {verified: Crossref 2026-11-03}
13+}
14+
15+@article{zhou2024drlcloud,
16+ title = {Deep reinforcement learning-based methods for resource scheduling in cloud computing: a review and future directions},
17+ author = {Guangyao Zhou and Wenhong Tian and Rajkumar Buyya and Ruini Xue and Liang Song},
18+ journal = {Artificial Intelligence Review},
19+ volume = {57},
20+ number = {5},
21+ year = {2024},
22+ doi = {10.1007/s10462-024-10756-9},
23+ note = {verified: Crossref 2026-11-03}
24+}
25+
26+@article{wang2023road6g,
27+ title = {On the Road to {6G}: Visions, Requirements, Key Technologies, and Testbeds},
28+ author = {Cheng-Xiang Wang and Xiaohu You and Xiqi Gao and Xiuming Zhu and Zixin Li and Chuan Zhang and Haiming Wang and Yongming Huang and Yunfei Chen and Harald Haas and John S. Thompson and Erik G. Larsson and Marco Di Renzo and Wen Tong and Peiying Zhu and Xuemin Shen and H. Vincent Poor and Lajos Hanzo},
29+ journal = {IEEE Communications Surveys \& Tutorials},
30+ volume = {25},
31+ number = {2},
32+ pages = {905--974},
33+ year = {2023},
34+ doi = {10.1109/COMST.2023.3249835},
35+ note = {verified: Crossref 2026-11-03}
36+}
37+
38+@article{liu2022isac,
39+ title = {Integrated Sensing and Communications: Toward Dual-Functional Wireless Networks for {6G} and Beyond},
40+ author = {Fan Liu and Yuanhao Cui and Christos Masouros and Jie Xu and Tony Xiao Han and Yonina C. Eldar and Stefano Buzzi},
41+ journal = {IEEE Journal on Selected Areas in Communications},
42+ volume = {40},
43+ number = {6},
44+ pages = {1728--1767},
45+ year = {2022},
46+ doi = {10.1109/JSAC.2022.3156632},
47+ note = {verified: Crossref 2026-11-03}
48+}
49+
50+@article{qu2025meillm,
51+ title = {Mobile Edge Intelligence for Large Language Models: A Contemporary Survey},
52+ author = {Guanqiao Qu and Qiyuan Chen and Wei Wei and Zheng Lin and Xianhao Chen and Kaibin Huang},
53+ journal = {IEEE Communications Surveys \& Tutorials},
54+ volume = {27},
55+ number = {6},
56+ pages = {3820--3860},
57+ year = {2025},
58+ doi = {10.1109/COMST.2025.3527641},
59+ note = {verified: Crossref 2026-11-03}
60+}
61+
62+@article{zhang2026edgegeneral,
63+ title = {Toward Edge General Intelligence With Agentic {AI} and Agentification: Concepts, Technologies, and Future Directions},
64+ author = {Ruichen Zhang and Guangyuan Liu and Yinqiu Liu and Changyuan Zhao and Jiacheng Wang and Yunting Xu and Dusit Niyato and Jiawen Kang and Yonghui Li and Shiwen Mao and Sumei Sun and Xuemin Shen and Dong In Kim},
65+ journal = {IEEE Communications Surveys \& Tutorials},
66+ volume = {28},
67+ pages = {4285--4318},
68+ year = {2026},
69+ doi = {10.1109/COMST.2026.3651702},
70+ note = {verified: Crossref 2026-11-03}
71+}
72+
73+@article{liu2026eccc,
74+ title = {Edge-Cloud Collaborative Computing on Distributed Intelligence and Model Optimization: A Survey},
75+ author = {Jing Liu and Yao Du and Kun Yang and Jiaqi Wu and Yan Wang and Xiping Hu and Zehua Wang and Yang Liu and Peng Sun and Azzedine Boukerche and Victor C. M. Leung},
76+ journal = {IEEE Communications Surveys \& Tutorials},
77+ volume = {28},
78+ pages = {5049--5080},
79+ year = {2026},
80+ doi = {10.1109/COMST.2026.3669216},
81+ note = {verified: Crossref 2026-11-03}
82+}
83+
84+@article{jiang2026lamagentic,
85+ title = {From Large {AI} Models to Agentic {AI}: A Tutorial on Future Intelligent Communications},
86+ author = {Feibo Jiang and Cunhua Pan and Kezhi Wang and Pietro Michiardi and Octavia A. Dobre and Merouane Debbah},
87+ journal = {IEEE Journal on Selected Areas in Communications},
88+ volume = {44},
89+ pages = {3507--3540},
90+ year = {2026},
91+ doi = {10.1109/JSAC.2026.3660010},
92+ note = {verified: Crossref 2026-11-03}
93+}
94+
95+@article{douze2026faiss,
96+ title = {The {Faiss} Library},
97+ author = {Matthijs Douze and Alexandr Guzhva and Chengqi Deng and Jeff Johnson and Gergely Szilvasy and Pierre-Emmanuel Mazar{\'e} and Maria Lomeli and Lucas Hosseini and Herv{\'e} J{\'e}gou},
98+ journal = {IEEE Transactions on Big Data},
99+ volume = {12},
100+ number = {2},
101+ pages = {346--361},
102+ year = {2026},
103+ doi = {10.1109/TBDATA.2025.3618474},
104+ note = {verified: Crossref 2026-11-03}
105+}
106+
107+@article{zha2025datacentric,
108+ title = {Data-centric Artificial Intelligence: A Survey},
109+ author = {Daochen Zha and Zaid Pervaiz Bhat and Kwei-Herng Lai and Fan Yang and Zhimeng Jiang and Shaochen Zhong and Xia Hu},
110+ journal = {ACM Computing Surveys},
111+ volume = {57},
112+ number = {5},
113+ pages = {1--42},
114+ year = {2025},
115+ doi = {10.1145/3711118},
116+ note = {verified: Crossref 2026-11-03}
117+}
118+
119+@article{mohammed2025dataquality,
120+ title = {The effects of data quality on machine learning performance on tabular data},
121+ author = {Sedir Mohammed and Lukas Budach and Moritz Feuerpfeil and Nina Ihde and Andrea Nathansen and Nele Noack and Hendrik Patzlaff and Felix Naumann and Hazar Harmouch},
122+ journal = {Information Systems},
123+ volume = {132},
124+ pages = {102549},
125+ year = {2025},
126+ doi = {10.1016/j.is.2025.102549},
127+ note = {verified: Crossref 2026-11-03}
128+}
129+
130+@article{dinh2018blockbench,
131+ title = {Untangling Blockchain: A Data Processing View of Blockchain Systems},
132+ author = {Tien Tuan Anh Dinh and Rui Liu and Meihui Zhang and Gang Chen and Beng Chin Ooi and Ji Wang},
133+ journal = {IEEE Transactions on Knowledge and Data Engineering},
134+ volume = {30},
135+ number = {7},
136+ pages = {1366--1385},
137+ year = {2018},
138+ doi = {10.1109/TKDE.2017.2781227},
139+ note = {verified: Crossref 2026-11-03}
140+}
141+
142+@inproceedings{fan2024ragsurvey,
143+ title = {A Survey on {RAG} Meeting {LLMs}: Towards Retrieval-Augmented Large Language Models},
144+ author = {Wenqi Fan and Yujuan Ding and Liangbo Ning and Shijie Wang and Hengyun Li and Dawei Yin and Tat-Seng Chua and Qing Li},
145+ booktitle = {Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD '24)},
146+ pages = {6491--6501},
147+ year = {2024},
148+ publisher = {ACM},
149+ doi = {10.1145/3637528.3671470},
150+ note = {verified: Crossref 2026-11-03}
151+}
152+
153+@article{wu2024llmrec,
154+ title = {A survey on large language models for recommendation},
155+ author = {Likang Wu and Zhi Zheng and Zhaopeng Qiu and Hao Wang and Hongchao Gu and Tingjia Shen and Chuan Qin and Chen Zhu and Hengshu Zhu and Qi Liu and Hui Xiong and Enhong Chen},
156+ journal = {World Wide Web},
157+ volume = {27},
158+ number = {5},
159+ year = {2024},
160+ doi = {10.1007/s11280-024-01291-2},
161+ note = {verified: Crossref 2026-11-03}
162+}
163+
164+@article{zhao2024recsysllm,
165+ title = {Recommender Systems in the Era of Large Language Models ({LLMs})},
166+ author = {Zihuai Zhao and Wenqi Fan and Jiatong Li and Yunqing Liu and Xiaowei Mei and Yiqi Wang and Zhen Wen and Fei Wang and Xiangyu Zhao and Jiliang Tang and Qing Li},
167+ journal = {IEEE Transactions on Knowledge and Data Engineering},
168+ volume = {36},
169+ number = {11},
170+ pages = {6889--6907},
171+ year = {2024},
172+ doi = {10.1109/TKDE.2024.3392335},
173+ note = {verified: Crossref 2026-11-03}
174+}
175+
176+@inproceedings{xiao2024cpack,
177+ title = {{C-Pack}: Packed Resources For General {Chinese} Embeddings},
178+ author = {Shitao Xiao and Zheng Liu and Peitian Zhang and Niklas Muennighoff and Defu Lian and Jian-Yun Nie},
179+ booktitle = {Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2024)},
180+ pages = {641--649},
181+ year = {2024},
182+ publisher = {ACM},
183+ doi = {10.1145/3626772.3657878},
184+ note = {verified: Crossref 2026-11-03}
185+}
186+
187+@article{zhao2026ragaigc,
188+ title = {Retrieval-Augmented Generation for {AI}-Generated Content: A Survey},
189+ author = {Penghao Zhao and Hailin Zhang and Qinhan Yu and Zhengren Wang and Yunteng Geng and Fangcheng Fu and Ling Yang and Wentao Zhang and Jie Jiang and Bin Cui},
190+ journal = {Data Science and Engineering},
191+ volume = {11},
192+ number = {1},
193+ pages = {1--29},
194+ year = {2026},
195+ doi = {10.1007/s41019-025-00335-5},
196+ note = {verified: Crossref 2026-11-03}
197+}
198+
199+@article{huang2026ragsurvey,
200+ title = {A Survey on Retrieval-Augmented Text Generation for Large Language Models},
201+ author = {Yizheng Huang and Jimmy Xiangji Huang},
202+ journal = {ACM Computing Surveys},
203+ volume = {58},
204+ number = {12},
205+ pages = {1--38},
206+ year = {2026},
207+ doi = {10.1145/3805774},
208+ note = {verified: Crossref 2026-11-03}
209+}
210+
211+@inproceedings{he2020lightgcn,
212+ title = {{LightGCN}: Simplifying and Powering Graph Convolution Network for Recommendation},
213+ author = {Xiangnan He and Kuan Deng and Xiang Wang and Yan Li and YongDong Zhang and Meng Wang},
214+ booktitle = {Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR '20)},
215+ pages = {639--648},
216+ year = {2020},
217+ publisher = {ACM},
218+ doi = {10.1145/3397271.3401063},
219+ note = {verified: Crossref 2026-11-03 (Crossref title field is the short form ``LightGCN'')}
220+}
221+
222+@article{wu2023gnnrec,
223+ title = {Graph Neural Networks in Recommender Systems: A Survey},
224+ author = {Shiwen Wu and Fei Sun and Wentao Zhang and Xu Xie and Bin Cui},
225+ journal = {ACM Computing Surveys},
226+ volume = {55},
227+ number = {5},
228+ pages = {1--37},
229+ year = {2023},
230+ doi = {10.1145/3535101},
231+ note = {verified: Crossref 2026-11-03}
232+}
233+
234+@article{gao2023gnnrecsurvey,
235+ title = {A Survey of Graph Neural Networks for Recommender Systems: Challenges, Methods, and Directions},
236+ author = {Chen Gao and Yu Zheng and Nian Li and Yinfeng Li and Yingrong Qin and Jinghua Piao and Yuhan Quan and Jianxin Chang and Depeng Jin and Xiangnan He and Yong Li},
237+ journal = {ACM Transactions on Recommender Systems},
238+ volume = {1},
239+ number = {1},
240+ pages = {1--51},
241+ year = {2023},
242+ doi = {10.1145/3568022},
243+ note = {verified: Crossref 2026-11-03}
244+}
245+
246+@article{davies2018loihi,
247+ title = {{Loihi}: A Neuromorphic Manycore Processor with On-Chip Learning},
248+ author = {Mike Davies and Narayan Srinivasa and Tsung-Han Lin and Gautham Chinya and Yongqiang Cao and Sri Harsha Choday and Georgios Dimou and Prasad Joshi and Nabil Imam and Shweta Jain and Yuyun Liao and Chit-Kwan Lin and Andrew Lines and Ruokun Liu and Deepak Mathaikutty and Steven McCoy and Arnab Paul and Jonathan Tse and Guruguhanathan Venkataramanan and Yi-Hsin Weng and Andreas Wild and Yoonseok Yang and Hong Wang},
249+ journal = {IEEE Micro},
250+ volume = {38},
251+ number = {1},
252+ pages = {82--99},
253+ year = {2018},
254+ doi = {10.1109/MM.2018.112130359},
255+ note = {verified: Crossref 2026-11-03}
256+}
257+
258+@article{schuman2022neuromorphic,
259+ title = {Opportunities for neuromorphic computing algorithms and applications},
260+ author = {Catherine D. Schuman and Shruti R. Kulkarni and Maryam Parsa and J. Parker Mitchell and Prasanna Date and Bill Kay},
261+ journal = {Nature Computational Science},
262+ volume = {2},
263+ number = {1},
264+ pages = {10--19},
265+ year = {2022},
266+ doi = {10.1038/s43588-021-00184-y},
267+ note = {verified: Crossref 2026-11-03}
268+}
269+
270+@article{xu2024taichi,
271+ title = {Large-scale photonic chiplet {Taichi} empowers 160-{TOPS/W} artificial general intelligence},
272+ author = {Zhihao Xu and Tiankuang Zhou and Muzhou Ma and ChenChen Deng and Qionghai Dai and Lu Fang},
273+ journal = {Science},
274+ volume = {384},
275+ number = {6692},
276+ pages = {202--209},
277+ year = {2024},
278+ doi = {10.1126/science.adl1203},
279+ note = {verified: Crossref 2026-11-03}
280+}
281+
282+@article{ahmed2025photonic,
283+ title = {Universal photonic artificial intelligence acceleration},
284+ author = {Sufi R. Ahmed and Reza Baghdadi and Mikhail Bernadskiy and Nate Bowman and Ryan Braid and Jim Carr and Chen Chen and Pietro Ciccarella and Matthew Cole and John Cooke and Kishor Desai and Carlos Dorta and Jonathan Elmhurst and Bryce Gardiner and others},
285+ journal = {Nature},
286+ volume = {640},
287+ number = {8058},
288+ pages = {368--374},
289+ year = {2025},
290+ doi = {10.1038/s41586-025-08854-x},
291+ note = {verified: Crossref 2026-11-03}
292+}
293+
294+@article{fu2024onn,
295+ title = {Optical neural networks: progress and challenges},
296+ author = {Tingzhao Fu and Jianfa Zhang and Run Sun and Yuyao Huang and Wei Xu and Sigang Yang and Zhihong Zhu and Hongwei Chen},
297+ journal = {Light: Science \& Applications},
298+ volume = {13},
299+ number = {1},
300+ year = {2024},
301+ doi = {10.1038/s41377-024-01590-3},
302+ note = {verified: Crossref 2026-11-03}
303+}
304+
305+@article{lam2026neuromorphicphotonic,
306+ title = {Neuromorphic photonic computing with an electro-optic analog memory},
307+ author = {Sean Lam and Ahmed Khaled and Simon Bilodeau and Bicky A. Marquez and Paul R. Prucnal and Lukas Chrostowski and Bhavin J. Shastri and Sudip Shekhar},
308+ journal = {Nature Communications},
309+ volume = {17},
310+ number = {1},
311+ year = {2026},
312+ doi = {10.1038/s41467-026-69084-x},
313+ note = {verified: Crossref 2026-11-03}
314+}
315+
316+@inproceedings{patel2024splitwise,
317+ title = {{Splitwise}: Efficient Generative {LLM} Inference Using Phase Splitting},
318+ author = {Pratyush Patel and Esha Choukse and Chaojie Zhang and Aashaka Shah and {\'I}{\~n}igo Goiri and Saeed Maleki and Ricardo Bianchini},
319+ booktitle = {2024 ACM/IEEE 51st Annual International Symposium on Computer Architecture (ISCA)},
320+ pages = {118--132},
321+ year = {2024},
322+ publisher = {IEEE},
323+ doi = {10.1109/ISCA59077.2024.00019},
324+ note = {verified: Crossref 2026-11-03}
325+}
326+
327+@article{lin2025awq,
328+ title = {{AWQ}: Activation-aware Weight Quantization for On-Device {LLM} Compression and Acceleration},
329+ author = {Ji Lin and Jiaming Tang and Haotian Tang and Shang Yang and Guangxuan Xiao and Song Han},
330+ journal = {GetMobile: Mobile Computing and Communications},
331+ volume = {28},
332+ number = {4},
333+ pages = {12--17},
334+ year = {2025},
335+ doi = {10.1145/3714983.3714987},
336+ note = {verified: Crossref 2026-11-03}
337+}
338+
339+@article{silvano2025accelerators,
340+ title = {A Survey on Deep Learning Hardware Accelerators for Heterogeneous {HPC} Platforms},
341+ author = {Cristina Silvano and Daniele Ielmini and Fabrizio Ferrandi and Leandro Fiorin and Serena Curzel and Luca Benini and Francesco Conti and Angelo Garofalo and Cristian Zambelli and Enrico Calore and Sebastiano Schifano and Maurizio Palesi and Giuseppe Ascia and Davide Patti and others},
342+ journal = {ACM Computing Surveys},
343+ volume = {57},
344+ number = {11},
345+ pages = {1--39},
346+ year = {2025},
347+ doi = {10.1145/3729215},
348+ note = {verified: Crossref 2026-11-03}
349+}
350+
351+@incollection{mutlu2023pim,
352+ title = {A Modern Primer on Processing in Memory},
353+ author = {Onur Mutlu and Saugata Ghose and Juan G{\'o}mez-Luna and Rachata Ausavarungnirun},
354+ booktitle = {Emerging Computing: From Devices to Systems},
355+ series = {Computer Architecture and Design Methodologies},
356+ pages = {171--243},
357+ year = {2023},
358+ publisher = {Springer Nature Singapore},
359+ doi = {10.1007/978-981-16-7487-7_7},
360+ note = {verified: Crossref 2026-11-03}
361+}
362+
363+@article{bennun2020demystifying,
364+ title = {Demystifying Parallel and Distributed Deep Learning: An In-depth Concurrency Analysis},
365+ author = {Tal Ben-Nun and Torsten Hoefler},
366+ journal = {ACM Computing Surveys},
367+ volume = {52},
368+ number = {4},
369+ pages = {1--43},
370+ year = {2020},
371+ doi = {10.1145/3320060},
372+ note = {verified: Crossref 2026-11-03}
373+}
374+
375+@inproceedings{song2024powerinfer,
376+ title = {{PowerInfer}: Fast Large Language Model Serving with a Consumer-grade {GPU}},
377+ author = {Yixin Song and Zeyu Mi and Haotong Xie and Haibo Chen},
378+ booktitle = {Proceedings of the ACM SIGOPS 30th Symposium on Operating Systems Principles (SOSP '24)},
379+ pages = {590--606},
380+ year = {2024},
381+ publisher = {ACM},
382+ doi = {10.1145/3694715.3695964},
383+ note = {verified: Crossref 2026-11-03}
384+}
385+
386+@inproceedings{strati2024orion,
387+ title = {{Orion}: Interference-aware, Fine-grained {GPU} Sharing for {ML} Applications},
388+ author = {Foteini Strati and Xianzhe Ma and Ana Klimovic},
389+ booktitle = {Proceedings of the Nineteenth European Conference on Computer Systems (EuroSys '24)},
390+ pages = {1075--1092},
391+ year = {2024},
392+ publisher = {ACM},
393+ doi = {10.1145/3627703.3629578},
394+ note = {verified: Crossref 2026-11-03}
395+}
396+
397+@inproceedings{prabhu2025vattention,
398+ title = {{vAttention}: Dynamic Memory Management for Serving {LLMs} without {PagedAttention}},
399+ author = {Ramya Prabhu and Ajay Nayak and Jayashree Mohan and Ramachandran Ramjee and Ashish Panwar},
400+ booktitle = {Proceedings of the 30th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS '25)},
401+ pages = {1133--1150},
402+ year = {2025},
403+ publisher = {ACM},
404+ doi = {10.1145/3669940.3707256},
405+ note = {verified: Crossref 2026-11-03}
406+}
407+
408+@inproceedings{lee2020keystone,
409+ title = {{Keystone}: an open framework for architecting trusted execution environments},
410+ author = {Dayeol Lee and David Kohlbrenner and Shweta Shinde and Krste Asanovi{\'c} and Dawn Song},
411+ booktitle = {Proceedings of the Fifteenth European Conference on Computer Systems (EuroSys '20)},
412+ pages = {1--16},
413+ year = {2020},
414+ publisher = {ACM},
415+ doi = {10.1145/3342195.3387532},
416+ note = {verified: Crossref 2026-11-03 (Crossref title field is the short form ``Keystone'')}
417+}
418+
419+@inproceedings{jia2021nightcore,
420+ title = {{Nightcore}: efficient and scalable serverless computing for latency-sensitive, interactive microservices},
421+ author = {Zhipeng Jia and Emmett Witchel},
422+ booktitle = {Proceedings of the 26th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS '21)},
423+ pages = {152--166},
424+ year = {2021},
425+ publisher = {ACM},
426+ doi = {10.1145/3445814.3446701},
427+ note = {verified: Crossref 2026-11-03}
428+}
429+
430+@article{googlequantum2023suppressing,
431+ title = {Suppressing quantum errors by scaling a surface code logical qubit},
432+ author = {{Google Quantum AI}},
433+ journal = {Nature},
434+ volume = {614},
435+ number = {7949},
436+ pages = {676--681},
437+ year = {2023},
438+ doi = {10.1038/s41586-022-05434-1},
439+ note = {verified: Crossref 2026-11-03}
440+}
441+
442+@article{bluvstein2024logical,
443+ title = {Logical quantum processor based on reconfigurable atom arrays},
444+ author = {Dolev Bluvstein and Simon J. Evered and Alexandra A. Geim and Sophie H. Li and Hengyun Zhou and Tom Manovitz and Sepehr Ebadi and Madelyn Cain and Marcin Kalinowski and Dominik Hangleiter and J. Pablo {Bonilla Ataides} and Nishad Maskara and Iris Cong and Xun Gao and others},
445+ journal = {Nature},
446+ volume = {626},
447+ number = {7997},
448+ pages = {58--65},
449+ year = {2024},
450+ doi = {10.1038/s41586-023-06927-3},
451+ note = {verified: Crossref 2026-11-03}
452+}
453+
454+@article{larocca2025barren,
455+ title = {Barren plateaus in variational quantum computing},
456+ author = {Mart{\'i}n Larocca and Supanut Thanasilp and Samson Wang and Kunal Sharma and Jacob Biamonte and Patrick J. Coles and Lukasz Cincio and Jarrod R. McClean and Zo{\"e} Holmes and M. Cerezo},
457+ journal = {Nature Reviews Physics},
458+ volume = {7},
459+ number = {4},
460+ pages = {174--189},
461+ year = {2025},
462+ doi = {10.1038/s42254-025-00813-9},
463+ note = {verified: Crossref 2026-11-03}
464+}
465+
466+@article{main2025distributedqc,
467+ title = {Distributed quantum computing across an optical network link},
468+ author = {D. Main and P. Drmota and D. P. Nadlinger and E. M. Ainley and A. Agrawal and B. C. Nichol and R. Srinivas and G. Araneda and D. M. Lucas},
469+ journal = {Nature},
470+ volume = {638},
471+ number = {8050},
472+ pages = {383--388},
473+ year = {2025},
474+ doi = {10.1038/s41586-024-08404-x},
475+ note = {verified: Crossref 2026-11-03}
476+}
477+
478+@article{aghaee2025aurora,
479+ title = {Scaling and networking a modular photonic quantum computer},
480+ author = {H. {Aghaee Rad} and T. Ainsworth and R. N. Alexander and B. Altieri and M. F. Askarani and R. Baby and L. Banchi and B. Q. Baragiola and J. E. Bourassa and R. S. Chadwick and I. Charania and H. Chen and M. J. Collins and P. Contu and others},
481+ journal = {Nature},
482+ volume = {638},
483+ number = {8052},
484+ pages = {912--919},
485+ year = {2025},
486+ doi = {10.1038/s41586-024-08406-9},
487+ note = {verified: Crossref 2026-11-03}
488+}
489+
490+@misc{javadiabhari2024qiskit,
491+ title = {Quantum computing with {Qiskit}},
492+ author = {Ali Javadi-Abhari and Matthew Treinish and Kevin Krsulich and Christopher J. Wood and Jake Lishman and Julien Gacon and Simon Martiel and Paul D. Nation and Lev S. Bishop and Andrew W. Cross and Blake R. Johnson and Jay M. Gambetta},
493+ year = {2024},
494+ eprint = {2405.08810},
495+ archiveprefix = {arXiv},
496+ howpublished = {arXiv preprint arXiv:2405.08810},
497+ note = {verified: arxiv API 2026-11-03}
498+}
499+
500+@article{abughanem2025ibmquantum,
501+ title = {{IBM} quantum computers: evolution, performance, and future directions},
502+ author = {Muhammad AbuGhanem},
503+ journal = {The Journal of Supercomputing},
504+ volume = {81},
505+ number = {5},
506+ year = {2025},
507+ doi = {10.1007/s11227-025-07047-7},
508+ note = {verified: Crossref 2026-11-03}
509+}
510+
511+@article{kanno2026qsci,
512+ title = {Quantum-selected configuration interaction: Classical diagonalization of {Hamiltonians} in subspaces selected by quantum computers},
513+ author = {Keita Kanno and Masaya Kohda and Ryosuke Imai and Sho Koh and Kosuke Mitarai and Wataru Mizukami and Yuya O. Nakagawa},
514+ journal = {Physical Review Research},
515+ volume = {8},
516+ number = {2},
517+ year = {2026},
518+ doi = {10.1103/dmn4-snfx},
519+ note = {verified: Crossref 2026-11-03}
520+}
521+
522+@inproceedings{dodge2022carbonintensity,
523+ title = {Measuring the Carbon Intensity of {AI} in Cloud Instances},
524+ author = {Jesse Dodge and Taylor Prewitt and Remi {Tachet des Combes} and Erika Odmark and Roy Schwartz and Emma Strubell and Alexandra Sasha Luccioni and Noah A. Smith and Nicole DeCario and Will Buchanan},
525+ booktitle = {2022 ACM Conference on Fairness, Accountability, and Transparency (FAccT '22)},
526+ pages = {1877--1894},
527+ year = {2022},
528+ publisher = {ACM},
529+ doi = {10.1145/3531146.3533234},
530+ note = {verified: Crossref 2026-11-03}
531+}
532+
533+@misc{li2023thirsty,
534+ title = {Making {AI} Less ``Thirsty'': Uncovering and Addressing the Secret Water Footprint of {AI} Models},
535+ author = {Pengfei Li and Jianyi Yang and Mohammad A. Islam and Shaolei Ren},
536+ year = {2023},
537+ eprint = {2304.03271},
538+ archiveprefix = {arXiv},
539+ howpublished = {arXiv preprint arXiv:2304.03271},
540+ note = {verified: arxiv API 2026-11-03}
541+}
542+
543+@article{li2025thirsty,
544+ title = {Making {AI} Less `Thirsty'},
545+ author = {Pengfei Li and Jianyi Yang and Mohammad A. Islam and Shaolei Ren},
546+ journal = {Communications of the ACM},
547+ volume = {68},
548+ number = {7},
549+ pages = {54--61},
550+ year = {2025},
551+ doi = {10.1145/3724499},
552+ note = {verified: Crossref 2026-11-03}
553+}
554+
555+@inproceedings{chien2023carbongenai,
556+ title = {Reducing the Carbon Impact of Generative {AI} Inference (today and in 2035)},
557+ author = {Andrew A. Chien and Liuzixuan Lin and Hai Nguyen and Varsha Rao and Tristan Sharma and Rajini Wijayawardana},
558+ booktitle = {Proceedings of the 2nd Workshop on Sustainable Computer Systems (HotCarbon '23)},
559+ pages = {1--7},
560+ year = {2023},
561+ publisher = {ACM},
562+ doi = {10.1145/3604930.3605705},
563+ note = {verified: Crossref 2026-11-03}
564+}
565+
566+@article{boloncanedo2024greenai,
567+ title = {A review of green artificial intelligence: Towards a more sustainable future},
568+ author = {Ver{\'o}nica Bol{\'o}n-Canedo and Laura Mor{\'a}n-Fern{\'a}ndez and Brais Cancela and Amparo Alonso-Betanzos},
569+ journal = {Neurocomputing},
570+ volume = {599},
571+ pages = {128096},
572+ year = {2024},
573+ doi = {10.1016/j.neucom.2024.128096},
574+ note = {verified: Crossref 2026-11-03}
575+}
576+
577+@article{devriesgao2026datacenters,
578+ title = {The carbon and water footprints of data centers and what this could mean for artificial intelligence},
579+ author = {Alex de Vries-Gao},
580+ journal = {Patterns},
581+ volume = {7},
582+ number = {1},
583+ pages = {101430},
584+ year = {2026},
585+ doi = {10.1016/j.patter.2025.101430},
586+ note = {verified: Crossref 2026-11-03}
587+}
588+
589+@article{thakur2025greenfl,
590+ title = {Green Federated Learning: A New Era of Green Aware {AI}},
591+ author = {Dipanwita Thakur and Antonella Guzzo and Giancarlo Fortino and Francesco Piccialli},
592+ journal = {ACM Computing Surveys},
593+ volume = {57},
594+ number = {8},
595+ pages = {1--36},
596+ year = {2025},
597+ doi = {10.1145/3718363},
598+ note = {verified: Crossref 2026-11-03}
599+}
600+
601+@inproceedings{luccioni2025jevons,
602+ title = {From Efficiency Gains to Rebound Effects: The Problem of {Jevons}' Paradox in {AI}'s Polarized Environmental Debate},
603+ author = {Alexandra Sasha Luccioni and Emma Strubell and Kate Crawford},
604+ booktitle = {Proceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency (FAccT '25)},
605+ pages = {76--88},
606+ year = {2025},
607+ publisher = {ACM},
608+ doi = {10.1145/3715275.3732007},
609+ note = {verified: Crossref 2026-11-03}
610+}
611+
612+@article{chen2026aielectricity,
613+ title = {Electricity demand and grid impacts of {AI} data centers: Challenges and prospects},
614+ author = {Xin Chen and Xiaoyang Wang and Ana Colacelli and Matt Lee and Le Xie},
615+ journal = {Nexus},
616+ volume = {3},
617+ number = {3},
618+ pages = {100162},
619+ year = {2026},
620+ doi = {10.1016/j.ynexs.2026.100162},
621+ note = {verified: Crossref 2026-11-03}
622+}
623+
624+@article{niu2026tokenpowerbench,
625+ title = {{TokenPowerBench}: Benchmarking the Power Consumption of {LLM} Inference},
626+ author = {Chenxu Niu and Wei Zhang and Jie Li and Yongjian Zhao and Tongyang Wang and Xi Wang and Yong Chen},
627+ journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
628+ volume = {40},
629+ number = {38},
630+ pages = {32582--32590},
631+ year = {2026},
632+ doi = {10.1609/aaai.v40i38.40535},
633+ note = {verified: Crossref 2026-11-03}
634+}
635+
636+@article{zhang2019dlwireless,
637+ title = {Deep Learning in Mobile and Wireless Networking: A Survey},
638+ author = {Chaoyun Zhang and Paul Patras and Hamed Haddadi},
639+ journal = {IEEE Communications Surveys \& Tutorials},
640+ volume = {21},
641+ number = {3},
642+ pages = {2224--2287},
643+ year = {2019},
644+ doi = {10.1109/COMST.2019.2904897},
645+ note = {verified: Crossref 2026-11-03}
646+}
647+
648+@techreport{ietf2024rfc9669,
649+ title = {{BPF} Instruction Set Architecture ({ISA})},
650+ author = {{Internet Engineering Task Force}},
651+ institution = {RFC Editor},
652+ number = {RFC 9669},
653+ year = {2024},
654+ doi = {10.17487/RFC9669},
655+ note = {verified: Crossref 2026-11-03 (Crossref record carries no author field)}
656+}
657+
658+@inproceedings{liao2026caps,
659+ title = {Multi-Objective Scheduling for Large Language Model Inference with Prompt-Level Cost Prediction and {SLO} Awareness},
660+ author = {Jiajing Liao and Feng Chang and Yihan Xue and Tianjian Xia and Zeyu Huang and Yuxiao Wang},
661+ booktitle = {2026 IEEE 8th International Conference on Communications, Information System and Computer Engineering (CISCE)},
662+ pages = {1606--1612},
663+ year = {2026},
664+ publisher = {IEEE},
665+ doi = {10.1109/CISCE69494.2026.11504731},
666+ note = {verified: Crossref 2026-11-03}
667+}
668+
669+@inproceedings{li2026deepservecb,
670+ title = {{DeepServeCB}: {SLO}-Aware and Cost-Aware Elastic Scheduling for Serverless Multi-Tenant {LLM} Inference},
671+ author = {Xuexian Li and Xiayuan Liu and Zilong Wang and Zijian Xiao and Chun-Yao Hsieh and Yixue Liu},
672+ booktitle = {2026 International Conference on Generative Artificial Intelligence and Information Security (GAIIS)},
673+ pages = {90--95},
674+ year = {2026},
675+ publisher = {IEEE},
676+ doi = {10.1109/GAIIS69281.2026.11519206},
677+ note = {verified: Crossref 2026-11-03}
678+}
679+
680+@article{an2026aiflow,
681+ title = {{AI} Flow: perspectives, scenarios, and approaches},
682+ author = {Hongjun An and Wenhan Hu and Sida Huang and Siqi Huang and Ruanjun Li and Yuanzhi Liang and Jiawei Shao and Yiliang Song and Zihan Wang and Cheng Yuan and Chi Zhang and Hongyuan Zhang and Wenhao Zhuang and Xuelong Li},
683+ journal = {Vicinagearth},
684+ volume = {3},
685+ number = {1},
686+ year = {2026},
687+ doi = {10.1007/s44336-025-00031-y},
688+ note = {verified: Crossref 2026-11-03}
689+}
690+
691+@article{cheikh2026massiveiot,
692+ title = {Energy, Scalability, Data, and Security in Massive {IoT}: Current Landscape and Future Directions},
693+ author = {Imane Cheikh and S{\'e}bastien Roy and Essaid Sabir and Rachid Aouami},
694+ journal = {IEEE Internet of Things Journal},
695+ volume = {13},
696+ number = {6},
697+ pages = {10127--10163},
698+ year = {2026},
699+ doi = {10.1109/JIOT.2026.3655796},
700+ note = {verified: Crossref 2026-11-03}
701+}
702+
703+@article{kumar2026sustainabledc,
704+ title = {Toward sustainable data center operation: A review on existing infrastructures, integrated smart energy management, and emerging opportunities},
705+ author = {Amit Kumar and Naran M. Pindoriya},
706+ journal = {Renewable and Sustainable Energy Reviews},
707+ volume = {230},
708+ pages = {116664},
709+ year = {2026},
710+ doi = {10.1016/j.rser.2025.116664},
711+ note = {verified: Crossref 2026-11-03 (title verified on its first clause; the subtitle follows the journal record)}
712+}
added manuscript/references_part07b.bib +539 −0
@@ -0,0 +1,539 @@
1+% references_part07b.bib — Part VII-b (security half of Part VII: cybersecurity, applied cryptography, blockchain,
2+% privacy-preserving computation, federated learning, LLM security). Agent H, 2026-11-03. Every entry verified by a single
3+% Crossref, DataCite or arXiv lookup; see references_verification_part07b.csv. Keys that already exist in references.bib
4+% (nist2024fips203, nist2024fips204, das2025llmsecurity, greshake2023injection, ...) are reused there and not redefined here.
5+
6+@techreport{rescorla2018tls13,
7+ author = {Rescorla, Eric},
8+ title = {{The Transport Layer Security (TLS) Protocol Version 1.3}},
9+ year = {2018},
10+ institution = {RFC Editor},
11+ doi = {10.17487/rfc8446}
12+}
13+
14+@inproceedings{androulaki2018fabric,
15+ author = {Androulaki, Elli and Barger, Artem and Bortnikov, Vita and Cachin, Christian and Christidis, Konstantinos and De Caro, Angelo and Enyeart, David and Ferris, Christopher and Laventman, Gennady and Manevich, Yacov and Muralidharan, Srinivasan and Murthy, Chet and Nguyen, Binh and Sethi, Manish and Singh, Gari and Smith, Keith and Sorniotti, Alessandro and Stathakopoulou, Chrysoula and Vukolić, Marko and Cocco, Sharon Weed and Yellick, Jason},
16+ title = {{Hyperledger fabric}},
17+ year = {2018},
18+ booktitle = {Proceedings of the Thirteenth EuroSys Conference},
19+ pages = {1--15},
20+ publisher = {ACM},
21+ doi = {10.1145/3190508.3190538}
22+}
23+
24+@article{yang2019fml,
25+ author = {Yang, Qiang and Liu, Yang and Chen, Tianjian and Tong, Yongxin},
26+ title = {{Federated Machine Learning}},
27+ year = {2019},
28+ journal = {ACM Transactions on Intelligent Systems and Technology},
29+ volume = {10},
30+ number = {2},
31+ pages = {1--19},
32+ doi = {10.1145/3298981}
33+}
34+
35+@incollection{zhu2020dlg,
36+ author = {Zhu, Ligeng and Han, Song},
37+ title = {{Deep Leakage from Gradients}},
38+ year = {2020},
39+ booktitle = {Lecture Notes in Computer Science},
40+ pages = {17--31},
41+ publisher = {Springer International Publishing},
42+ doi = {10.1007/978-3-030-63076-8_2}
43+}
44+
45+@article{tan2023personalizedfl,
46+ author = {Tan, Alysa Ziying and Yu, Han and Cui, Lizhen and Yang, Qiang},
47+ title = {{Towards Personalized Federated Learning}},
48+ year = {2023},
49+ journal = {IEEE Transactions on Neural Networks and Learning Systems},
50+ volume = {34},
51+ number = {12},
52+ pages = {9587--9603},
53+ doi = {10.1109/tnnls.2022.3160699}
54+}
55+
56+@article{ye2024heterogeneousfl,
57+ author = {Ye, Mang and Fang, Xiuwen and Du, Bo and Yuen, Pong C. and Tao, Dacheng},
58+ title = {{Heterogeneous Federated Learning: State-of-the-art and Research Challenges}},
59+ year = {2024},
60+ journal = {ACM Computing Surveys},
61+ volume = {56},
62+ number = {3},
63+ pages = {1--44},
64+ doi = {10.1145/3625558}
65+}
66+
67+@article{zhang2023fedala,
68+ author = {Zhang, Jianqing and Hua, Yang and Wang, Hao and Song, Tao and Xue, Zhengui and Ma, Ruhui and Guan, Haibing},
69+ title = {{FedALA: Adaptive Local Aggregation for Personalized Federated Learning}},
70+ year = {2023},
71+ journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
72+ volume = {37},
73+ number = {9},
74+ pages = {11237--11244},
75+ doi = {10.1609/aaai.v37i9.26330}
76+}
77+
78+@article{nguyen2025unlearning,
79+ author = {Nguyen, Thanh Tam and Huynh, Thanh Trung and Ren, Zhao and Nguyen, Phi Le and Liew, Alan Wee-Chung and Yin, Hongzhi and Nguyen, Quoc Viet Hung},
80+ title = {{A Survey of Machine Unlearning}},
81+ year = {2025},
82+ journal = {ACM Transactions on Intelligent Systems and Technology},
83+ volume = {16},
84+ number = {5},
85+ pages = {1--46},
86+ doi = {10.1145/3749987}
87+}
88+
89+@article{hou2026mcp,
90+ author = {Hou, Xinyi and Zhao, Yanjie and Wang, Shenao and Wang, Haoyu},
91+ title = {{Model Context Protocol (MCP): Landscape, Security Threats, and Future Research Directions}},
92+ year = {2026},
93+ journal = {ACM Transactions on Software Engineering and Methodology},
94+ volume = {35},
95+ number = {10},
96+ pages = {1--37},
97+ doi = {10.1145/3796519}
98+}
99+
100+@incollection{castryck2023sidh,
101+ author = {Castryck, Wouter and Decru, Thomas},
102+ title = {{An Efficient Key Recovery Attack on SIDH}},
103+ year = {2023},
104+ booktitle = {Lecture Notes in Computer Science},
105+ pages = {423--447},
106+ publisher = {Springer Nature Switzerland},
107+ doi = {10.1007/978-3-031-30589-4_15}
108+}
109+
110+@techreport{nist2019pqcround1,
111+ author = {Alagic, Gorjan and Alperin-Sheriff, Jacob and Apon, Daniel and Cooper, David and Dang, Quynh and Liu, Yi-Kai and Miller, Carl and Moody, Dustin and Peralta, Rene and Perlner, Ray and Robinson, Angela and Smith-Tone, Daniel},
112+ title = {{Status report on the first round of the NIST post-quantum cryptography standardization process}},
113+ year = {2019},
114+ institution = {National Institute of Standards and Technology},
115+ doi = {10.6028/nist.ir.8240}
116+}
117+
118+@techreport{nist2022pqcround3,
119+ author = {Alagic, Gorjan and Apon, Daniel and Cooper, David and Dang, Quynh and Dang, Thinh and Kelsey, John and Lichtinger, Jacob and Liu, Yi-Kai and Miller, Carl and Moody, Dustin and Peralta, Rene and Perlner, Ray and Robinson, Angela and Smith-Tone, Daniel},
120+ title = {{Status report on the third round of the NIST Post-Quantum Cryptography Standardization process}},
121+ year = {2022},
122+ institution = {National Institute of Standards and Technology (U.S.)},
123+ doi = {10.6028/nist.ir.8413}
124+}
125+
126+@article{foley2019bitcoin,
127+ author = {Foley, Sean and Karlsen, Jonathan R and Putniņš, Tālis J},
128+ title = {{Sex, Drugs, and Bitcoin: How Much Illegal Activity Is Financed through Cryptocurrencies?}},
129+ year = {2019},
130+ journal = {The Review of Financial Studies},
131+ volume = {32},
132+ number = {5},
133+ pages = {1798--1853},
134+ doi = {10.1093/rfs/hhz015}
135+}
136+
137+@article{cong2019smartcontracts,
138+ author = {Cong, Lin William and He, Zhiguo},
139+ title = {{Blockchain Disruption and Smart Contracts}},
140+ year = {2019},
141+ journal = {The Review of Financial Studies},
142+ volume = {32},
143+ number = {5},
144+ pages = {1754--1797},
145+ doi = {10.1093/rfs/hhz007}
146+}
147+
148+@techreport{yaga2018blockchain,
149+ author = {Yaga, Dylan and Mell, Peter and Roby, Nik and Scarfone, Karen},
150+ title = {{Blockchain technology overview}},
151+ year = {2018},
152+ institution = {National Institute of Standards and Technology},
153+ doi = {10.6028/nist.ir.8202}
154+}
155+
156+@article{wang2019consensus,
157+ author = {Wang, Wenbo and Hoang, Dinh Thai and Hu, Peizhao and Xiong, Zehui and Niyato, Dusit and Wang, Ping and Wen, Yonggang and Kim, Dong In},
158+ title = {{A Survey on Consensus Mechanisms and Mining Strategy Management in Blockchain Networks}},
159+ year = {2019},
160+ journal = {IEEE Access},
161+ volume = {7},
162+ pages = {22328--22370},
163+ doi = {10.1109/access.2019.2896108}
164+}
165+
166+@incollection{kurakin2018physical,
167+ author = {Kurakin, Alexey and Goodfellow, Ian J. and Bengio, Samy},
168+ title = {{Adversarial Examples in the Physical World}},
169+ year = {2018},
170+ booktitle = {Artificial Intelligence Safety and Security},
171+ pages = {99--112},
172+ publisher = {Chapman and Hall/CRC},
173+ doi = {10.1201/9781351251389-8}
174+}
175+
176+@article{thapa2022splitfed,
177+ author = {Thapa, Chandra and Mahawaga Arachchige, Pathum Chamikara and Camtepe, Seyit and Sun, Lichao},
178+ title = {{SplitFed: When Federated Learning Meets Split Learning}},
179+ year = {2022},
180+ journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
181+ volume = {36},
182+ number = {8},
183+ pages = {8485--8493},
184+ doi = {10.1609/aaai.v36i8.20825}
185+}
186+
187+@article{du2026vulrag,
188+ author = {Du, Xueying and Zheng, Geng and Wang, Kaixin and Zou, Yi and Wang, Yujia and Deng, Wentai and Feng, Jiayi and Liu, Mingwei and Chen, Bihuan and Peng, Xin and Ma, Tao and Lou, Yiling},
189+ title = {{vul-rag : Enhancing LLM-based Vulnerability Detection via Knowledge-level RAG}},
190+ year = {2026},
191+ journal = {ACM Transactions on Software Engineering and Methodology},
192+ volume = {35},
193+ number = {10},
194+ pages = {1--25},
195+ doi = {10.1145/3797277}
196+}
197+
198+@article{ahmed2020tracingapps,
199+ author = {Ahmed, Nadeem and Michelin, Regio A. and Xue, Wanli and Ruj, Sushmita and Malaney, Robert and Kanhere, Salil S. and Seneviratne, Aruna and Hu, Wen and Janicke, Helge and Jha, Sanjay K.},
200+ title = {{A Survey of COVID-19 Contact Tracing Apps}},
201+ year = {2020},
202+ journal = {IEEE Access},
203+ volume = {8},
204+ pages = {134577--134601},
205+ doi = {10.1109/access.2020.3010226}
206+}
207+
208+@article{sheller2020flmedicine,
209+ author = {Sheller, Micah J. and Edwards, Brandon and Reina, G. Anthony and Martin, Jason and Pati, Sarthak and Kotrotsou, Aikaterini and Milchenko, Mikhail and Xu, Weilin and Marcus, Daniel and Colen, Rivka R. and Bakas, Spyridon},
210+ title = {{Federated learning in medicine: facilitating multi-institutional collaborations without sharing patient data}},
211+ year = {2020},
212+ journal = {Scientific Reports},
213+ volume = {10},
214+ number = {1},
215+ doi = {10.1038/s41598-020-69250-1}
216+}
217+
218+@article{chen2020fedhealth,
219+ author = {Chen, Yiqiang and Qin, Xin and Wang, Jindong and Yu, Chaohui and Gao, Wen},
220+ title = {{FedHealth: A Federated Transfer Learning Framework for Wearable Healthcare}},
221+ year = {2020},
222+ journal = {IEEE Intelligent Systems},
223+ volume = {35},
224+ number = {4},
225+ pages = {83--93},
226+ doi = {10.1109/mis.2020.2988604}
227+}
228+
229+@article{zhang2024fedtgp,
230+ author = {Zhang, Jianqing and Liu, Yang and Hua, Yang and Cao, Jian},
231+ title = {{FedTGP: Trainable Global Prototypes with Adaptive-Margin-Enhanced Contrastive Learning for Data and Model Heterogeneity in Federated Learning}},
232+ year = {2024},
233+ journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
234+ volume = {38},
235+ number = {15},
236+ pages = {16768--16776},
237+ doi = {10.1609/aaai.v38i15.29617}
238+}
239+
240+@article{fu2026dpfl,
241+ author = {Fu, Jie and Hong, Yuan and Ling, Xinpeng and Wang, Leixia and Ran, Xun and Sun, Zhiyu and Wang, Wendy Hui and Chen, Zhili and Cao, Yang},
242+ title = {{Differentially Private Federated Learning: A Systematic Review}},
243+ year = {2026},
244+ journal = {ACM Computing Surveys},
245+ volume = {58},
246+ number = {11},
247+ pages = {1--38},
248+ doi = {10.1145/3801079}
249+}
250+
251+@article{jimenezgutierrez2026flsecurity,
252+ author = {Jimenez-Gutierrez, Daniel M. and Falkouskaya, Yelizaveta and Hernandez-Ramos, José L. and Anagnostopoulos, Aris and Chatzigiannakis, Ioannis and Vitaletti, Andrea},
253+ title = {{On the security and privacy of federated learning: A survey with attacks, defenses, frameworks, applications, and future directions}},
254+ year = {2026},
255+ journal = {Information Fusion},
256+ volume = {131},
257+ pages = {104155},
258+ doi = {10.1016/j.inffus.2026.104155}
259+}
260+
261+@article{davidson2018institutions,
262+ author = {Davidson, Sinclair and De Filippi, Primavera and Potts, Jason},
263+ title = {{Blockchains and the economic institutions of capitalism}},
264+ year = {2018},
265+ journal = {Journal of Institutional Economics},
266+ volume = {14},
267+ number = {4},
268+ pages = {639--658},
269+ doi = {10.1017/s1744137417000200}
270+}
271+
272+@article{mendling2018bpm,
273+ author = {Mendling, Jan and Weber, Ingo and Aalst, Wil Van Der and Brocke, Jan Vom and Cabanillas, Cristina and Daniel, Florian and Debois, Søren and Ciccio, Claudio Di and Dumas, Marlon and Dustdar, Schahram and Gal, Avigdor and García-Bañuelos, Luciano and Governatori, Guido and Hull, Richard and Rosa, Marcello La and Leopold, Henrik and Leymann, Frank and Recker, Jan and Reichert, Manfred and Reijers, Hajo A. and Rinderle-Ma, Stefanie and Solti, Andreas and Rosemann, Michael and Schulte, Stefan and Singh, Munindar P. and Slaats, Tijs and Staples, Mark and Weber, Barbara and Weidlich, Matthias and Weske, Mathias and Xu, Xiwei and Zhu, Liming},
274+ title = {{Blockchains for Business Process Management - Challenges and Opportunities}},
275+ year = {2018},
276+ journal = {ACM Transactions on Management Information Systems},
277+ volume = {9},
278+ number = {1},
279+ pages = {1--16},
280+ doi = {10.1145/3183367}
281+}
282+
283+@article{saberi2019supplychain,
284+ author = {Saberi, Sara and Kouhizadeh, Mahtab and Sarkis, Joseph and Shen, Lejia},
285+ title = {{Blockchain technology and its relationships to sustainable supply chain management}},
286+ year = {2019},
287+ journal = {International Journal of Production Research},
288+ volume = {57},
289+ number = {7},
290+ pages = {2117--2135},
291+ doi = {10.1080/00207543.2018.1533261}
292+}
293+
294+@article{conti2018bitcoinsecurity,
295+ author = {Conti, Mauro and Sandeep Kumar, E. and Lal, Chhagan and Ruj, Sushmita},
296+ title = {{A Survey on Security and Privacy Issues of Bitcoin}},
297+ year = {2018},
298+ journal = {IEEE Communications Surveys \& Tutorials},
299+ volume = {20},
300+ number = {4},
301+ pages = {3416--3452},
302+ doi = {10.1109/comst.2018.2842460}
303+}
304+
305+@article{fernandez2018biot,
306+ author = {Fernandez-Carames, Tiago M. and Fraga-Lamas, Paula},
307+ title = {{A Review on the Use of Blockchain for the Internet of Things}},
308+ year = {2018},
309+ journal = {IEEE Access},
310+ volume = {6},
311+ pages = {32979--33001},
312+ doi = {10.1109/access.2018.2842685}
313+}
314+
315+@article{huynh2023metaverse,
316+ author = {Huynh-The, Thien and Gadekallu, Thippa Reddy and Wang, Weizheng and Yenduri, Gokul and Ranaweera, Pasika and Pham, Quoc-Viet and da Costa, Daniel Benevides and Liyanage, Madhusanka},
317+ title = {{Blockchain for the metaverse: A Review}},
318+ year = {2023},
319+ journal = {Future Generation Computer Systems},
320+ volume = {143},
321+ pages = {401--419},
322+ doi = {10.1016/j.future.2023.02.008}
323+}
324+
325+@article{mohammadi2026dpmedical,
326+ author = {Mohammadi, Marziyeh and Vejdanihemmat, Mohsen and Lotfinia, Mahshad and Rusu, Mirabela and Truhn, Daniel and Maier, Andreas and Tayebi Arasteh, Soroosh},
327+ title = {{Differential privacy for medical deep learning: methods, tradeoffs, and deployment implications}},
328+ year = {2026},
329+ journal = {npj Digital Medicine},
330+ volume = {9},
331+ number = {1},
332+ doi = {10.1038/s41746-025-02280-z}
333+}
334+
335+@article{eden2025flgovernance,
336+ author = {Eden, Rebekah and Chukwudi, Ignatius and Bain, Chris and Barbieri, Sebastiano and Callaway, Leonie and de Jersey, Susan and George, Yasmeen and Gorse, Alain-Dominique and Lawley, Michael and Marendy, Peter and McPhail, Steven M. and Nguyen, Anthony and Samadbeik, Mahnaz and Sullivan, Clair},
337+ title = {{A scoping review of the governance of federated learning in healthcare}},
338+ year = {2025},
339+ journal = {npj Digital Medicine},
340+ volume = {8},
341+ number = {1},
342+ doi = {10.1038/s41746-025-01836-3}
343+}
344+
345+@article{zhu2021noniid,
346+ author = {Zhu, Hangyu and Xu, Jinjin and Liu, Shiqing and Jin, Yaochu},
347+ title = {{Federated learning on non-IID data: A survey}},
348+ year = {2021},
349+ journal = {Neurocomputing},
350+ volume = {465},
351+ pages = {371--390},
352+ doi = {10.1016/j.neucom.2021.07.098}
353+}
354+
355+@article{zhang2021flsurvey,
356+ author = {Zhang, Chen and Xie, Yu and Bai, Hang and Yu, Bin and Li, Weihong and Gao, Yuan},
357+ title = {{A survey on federated learning}},
358+ year = {2021},
359+ journal = {Knowledge-Based Systems},
360+ volume = {216},
361+ pages = {106775},
362+ doi = {10.1016/j.knosys.2021.106775}
363+}
364+
365+@article{xu2021flhealth,
366+ author = {Xu, Jie and Glicksberg, Benjamin S. and Su, Chang and Walker, Peter and Bian, Jiang and Wang, Fei},
367+ title = {{Federated Learning for Healthcare Informatics}},
368+ year = {2021},
369+ journal = {Journal of Healthcare Informatics Research},
370+ volume = {5},
371+ number = {1},
372+ pages = {1--19},
373+ doi = {10.1007/s41666-020-00082-4}
374+}
375+
376+@inproceedings{yang2024fedas,
377+ author = {Yang, Xiyuan and Huang, Wenke and Ye, Mang},
378+ title = {{FedAS: Bridging Inconsistency in Personalized Federated Learning}},
379+ year = {2024},
380+ booktitle = {2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
381+ pages = {11986--11995},
382+ publisher = {IEEE},
383+ doi = {10.1109/cvpr52733.2024.01139}
384+}
385+
386+@article{ali2026huntgpt,
387+ author = {Ali, Tarek and Kostakos, Panos and Sheikhi, Saeid},
388+ title = {{HuntGPT: Integrating Machine Learning-Based Anomaly Detection and Explainable AI with Large Language Models (LLMs)}},
389+ year = {2026},
390+ journal = {Telecom},
391+ volume = {7},
392+ number = {3},
393+ pages = {73},
394+ doi = {10.3390/telecom7030073}
395+}
396+
397+@inproceedings{bonawitz2017secagg,
398+ author = {Bonawitz, Keith and Ivanov, Vladimir and Kreuter, Ben and Marcedone, Antonio and McMahan, H. Brendan and Patel, Sarvar and Ramage, Daniel and Segal, Aaron and Seth, Karn},
399+ title = {{Practical Secure Aggregation for Privacy-Preserving Machine Learning}},
400+ year = {2017},
401+ booktitle = {Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security},
402+ pages = {1175--1191},
403+ publisher = {ACM},
404+ doi = {10.1145/3133956.3133982}
405+}
406+
407+@incollection{beullens2022rainbow,
408+ author = {Beullens, Ward},
409+ title = {{Breaking Rainbow Takes a Weekend on a Laptop}},
410+ year = {2022},
411+ booktitle = {Lecture Notes in Computer Science},
412+ pages = {464--479},
413+ publisher = {Springer Nature Switzerland},
414+ doi = {10.1007/978-3-031-15979-4_16}
415+}
416+
417+@article{uddin2026phishing,
418+ author = {Uddin, Mohammad Amaz and Mahiuddin, Md and Sarker, Iqbal H.},
419+ title = {{An explainable transformer-based model for phishing email detection: A large language model approach}},
420+ year = {2026},
421+ journal = {Computer Networks},
422+ volume = {277},
423+ pages = {112061},
424+ doi = {10.1016/j.comnet.2026.112061}
425+}
426+
427+@article{xu2025cbrfl,
428+ author = {Xu, Gang and Lei, Lele and Mao, Yanhui and Li, Zongpeng and Chen, Xiu-Bo and Zhang, Kejia},
429+ title = {{CBRFL: A framework for Committee-based Byzantine-Resilient Federated Learning}},
430+ year = {2025},
431+ journal = {Journal of Network and Computer Applications},
432+ volume = {238},
433+ pages = {104165},
434+ doi = {10.1016/j.jnca.2025.104165}
435+}
436+
437+@techreport{nist2024transition,
438+ author = {Regenscheid, Andrew},
439+ title = {{Transition to Post-Quantum Cryptography Standards}},
440+ year = {2024},
441+ institution = {National Institute of Standards and Technology},
442+ doi = {10.6028/NIST.IR.8547.ipd}
443+}
444+
445+@techreport{nist2025hqc,
446+ author = {Alagic, Gorjan and Bros, Maxime and Ciadoux, Pierre and Cooper, David and Dang, Quynh and Dang, Thinh and Kelsey, John and Lichtinger, Jacob and Liu, Yi-Kai and Miller, Carl and Moody, Dustin and Peralta, Rene and Perlner, Ray and Robinson, Angela and Silberg, Hamilton and Smith-Tone, Daniel and Waller, Noah},
447+ title = {{Status report on the fourth round of the NIST post-quantum cryptography standardization process}},
448+ year = {2025},
449+ institution = {National Institute of Standards and Technology (U.S.)},
450+ doi = {10.6028/NIST.IR.8545}
451+}
452+
453+@article{guo2022blockchainsecurity,
454+ author = {Guo, Huaqun and Yu, Xingjie},
455+ title = {{A survey on blockchain technology and its security}},
456+ year = {2022},
457+ journal = {Blockchain: Research and Applications},
458+ volume = {3},
459+ number = {2},
460+ pages = {100067},
461+ doi = {10.1016/j.bcra.2022.100067}
462+}
463+
464+@article{kim2018provenance,
465+ author = {Kim, Henry M. and Laskowski, Marek},
466+ title = {{Toward an ontology-driven blockchain design for supply-chain provenance}},
467+ year = {2018},
468+ journal = {Intelligent Systems in Accounting, Finance and Management},
469+ volume = {25},
470+ number = {1},
471+ pages = {18--27},
472+ doi = {10.1002/isaf.1424}
473+}
474+
475+@article{esposito2018healthcare,
476+ author = {Esposito, Christian and De Santis, Alfredo and Tortora, Genny and Chang, Henry and Choo, Kim-Kwang Raymond},
477+ title = {{Blockchain: A Panacea for Healthcare Cloud-Based Data Security and Privacy?}},
478+ year = {2018},
479+ journal = {IEEE Cloud Computing},
480+ volume = {5},
481+ number = {1},
482+ pages = {31--37},
483+ doi = {10.1109/mcc.2018.011791712}
484+}
485+
486+@misc{li2019qffl,
487+ author = {Tian Li and Maziar Sanjabi and Ahmad Beirami and Virginia Smith},
488+ title = {{Fair Resource Allocation in Federated Learning}},
489+ year = {2019},
490+ eprint = {1905.10497},
491+ archivePrefix = {arXiv},
492+ howpublished = {arXiv preprint arXiv:1905.10497}
493+}
494+
495+@misc{li2018fedprox,
496+ author = {Tian Li and Anit Kumar Sahu and Manzil Zaheer and Maziar Sanjabi and Ameet Talwalkar and Virginia Smith},
497+ title = {{Federated Optimization in Heterogeneous Networks}},
498+ year = {2018},
499+ eprint = {1812.06127},
500+ archivePrefix = {arXiv},
501+ howpublished = {arXiv preprint arXiv:1812.06127}
502+}
503+
504+@misc{wang2021nft,
505+ author = {Qin Wang and Rujia Li and Qi Wang and Shiping Chen},
506+ title = {{Non-Fungible Token (NFT): Overview, Evaluation, Opportunities and Challenges}},
507+ year = {2021},
508+ eprint = {2105.07447},
509+ archivePrefix = {arXiv},
510+ howpublished = {arXiv preprint arXiv:2105.07447}
511+}
512+
513+@misc{cho2020tracing,
514+ author = {Hyunghoon Cho and Daphne Ippolito and Yun William Yu},
515+ title = {{Contact Tracing Mobile Apps for COVID-19: Privacy Considerations and Related Trade-offs}},
516+ year = {2020},
517+ eprint = {2003.11511},
518+ archivePrefix = {arXiv},
519+ howpublished = {arXiv preprint arXiv:2003.11511}
520+}
521+
522+@misc{bagdasaryan2018backdoor,
523+ author = {Eugene Bagdasaryan and Andreas Veit and Yiqing Hua and Deborah Estrin and Vitaly Shmatikov},
524+ title = {{How To Backdoor Federated Learning}},
525+ year = {2018},
526+ eprint = {1807.00459},
527+ archivePrefix = {arXiv},
528+ howpublished = {arXiv preprint arXiv:1807.00459}
529+}
530+
531+@inproceedings{guesmi2026reentrancy,
532+ author = {Guesmi, Semia and Piazza, Carla and Gasparetto, Andrea and Rizzo, Matteo and Rossi, Sabina},
533+ title = {{Detecting Cross-Function Reentrancy from EVM Traces}},
534+ year = {2026},
535+ booktitle = {OASIcs, Volume 142, FMBC 2026 (6th International Workshop on Formal Methods for Blockchains)},
536+ pages = {8:1--8:15},
537+ publisher = {Schloss Dagstuhl -- Leibniz-Zentrum für Informatik},
538+ doi = {10.4230/oasics.fmbc.2026.8}
539+}
added manuscript/references_verification_part06.csv +63 −0
@@ -0,0 +1,63 @@
1+key,doi_or_arxiv,verified_via,verified_title,match,note
2+chen2024rcacopilot,10.1145/3627703.3629553,crossref,Automatic Root Cause Analysis via Large Language Models for Cloud Incidents,exact,EuroSys 2024; corpus work 4394946189
3+zhou2024drlcloud,10.1007/s10462-024-10756-9,crossref,Deep reinforcement learning-based methods for resource scheduling in cloud computing: a review and future directions,exact,corpus work 4395047770
4+wang2023road6g,10.1109/COMST.2023.3249835,crossref,"On the Road to 6G: Visions, Requirements, Key Technologies, and Testbeds",exact,corpus work 4322576964
5+liu2022isac,10.1109/JSAC.2022.3156632,crossref,Integrated Sensing and Communications: Toward Dual-Functional Wireless Networks for 6G and Beyond,exact,corpus work 3194620979
6+qu2025meillm,10.1109/COMST.2025.3527641,crossref,Mobile Edge Intelligence for Large Language Models: A Contemporary Survey,exact,corpus work 4406207083
7+zhang2026edgegeneral,10.1109/COMST.2026.3651702,crossref,"Toward Edge General Intelligence With Agentic AI and Agentification: Concepts, Technologies, and Future Directions",exact,corpus work 7123361055
8+liu2026eccc,10.1109/COMST.2026.3669216,crossref,Edge-Cloud Collaborative Computing on Distributed Intelligence and Model Optimization: A Survey,exact,corpus work 4415028496
9+jiang2026lamagentic,10.1109/JSAC.2026.3660010,crossref,From Large AI Models to Agentic AI: A Tutorial on Future Intelligent Communications,exact,corpus work 7127083419
10+douze2026faiss,10.1109/TBDATA.2025.3618474,crossref,The Faiss Library,exact,print 2026 (v12 n2); OpenAlex year 2025; corpus work 4415124086
11+zha2025datacentric,10.1145/3711118,crossref,Data-centric Artificial Intelligence: A Survey,exact,corpus work 4406085118
12+mohammed2025dataquality,10.1016/j.is.2025.102549,crossref,The effects of data quality on machine learning performance on tabular data,exact,corpus work 4408444355
13+dinh2018blockbench,10.1109/TKDE.2017.2781227,crossref,Untangling Blockchain: A Data Processing View of Blockchain Systems,exact,corpus work 2963507231
14+fan2024ragsurvey,10.1145/3637528.3671470,crossref,A Survey on RAG Meeting LLMs: Towards Retrieval-Augmented Large Language Models,exact,KDD 2024; corpus work 4401857375
15+wu2024llmrec,10.1007/s11280-024-01291-2,crossref,A survey on large language models for recommendation,exact,corpus work 4401834466
16+zhao2024recsysllm,10.1109/TKDE.2024.3392335,crossref,Recommender Systems in the Era of Large Language Models (LLMs),exact,corpus work 4394994587
17+xiao2024cpack,10.1145/3626772.3657878,crossref,C-Pack: Packed Resources For General Chinese Embeddings,exact,SIGIR 2024; corpus work 4400531953
18+zhao2026ragaigc,10.1007/s41019-025-00335-5,crossref,Retrieval-Augmented Generation for AI-Generated Content: A Survey,exact,corpus work 7118085507
19+huang2026ragsurvey,10.1145/3805774,crossref,A Survey on Retrieval-Augmented Text Generation for Large Language Models,exact,corpus work 4394947112
20+he2020lightgcn,10.1145/3397271.3401063,crossref,LightGCN,exact,Crossref title is the short form; full title from the ACM record; corpus work 3045200674
21+wu2023gnnrec,10.1145/3535101,crossref,Graph Neural Networks in Recommender Systems: A Survey,exact,print 2023 (v55 n5); OpenAlex year 2022; corpus work 3097300053
22+gao2023gnnrecsurvey,10.1145/3568022,crossref,"A Survey of Graph Neural Networks for Recommender Systems: Challenges, Methods, and Directions",exact,corpus work 4315977496
23+davies2018loihi,10.1109/MM.2018.112130359,crossref,Loihi: A Neuromorphic Manycore Processor with On-Chip Learning,exact,corpus work 2783525259
24+schuman2022neuromorphic,10.1038/s43588-021-00184-y,crossref,Opportunities for neuromorphic computing algorithms and applications,exact,corpus work 4210357113
25+xu2024taichi,10.1126/science.adl1203,crossref,Large-scale photonic chiplet Taichi empowers 160-TOPS/W artificial general intelligence,exact,corpus work 4394727812
26+ahmed2025photonic,10.1038/s41586-025-08854-x,crossref,Universal photonic artificial intelligence acceleration,exact,author list truncated with 'others'; corpus work 4409283612
27+fu2024onn,10.1038/s41377-024-01590-3,crossref,Optical neural networks: progress and challenges,exact,corpus work 4402694963
28+lam2026neuromorphicphotonic,10.1038/s41467-026-69084-x,crossref,Neuromorphic photonic computing with an electro-optic analog memory,exact,corpus work 7128293306
29+patel2024splitwise,10.1109/ISCA59077.2024.00019,crossref,Splitwise: Efficient Generative LLM Inference Using Phase Splitting,exact,ISCA 2024; corpus work 4401211704
30+lin2025awq,10.1145/3714983.3714987,crossref,AWQ: Activation-aware Weight Quantization for On-Device LLM Compression and Acceleration,exact,GetMobile reprint of the MLSys 2024 paper; corpus work 4406650295
31+silvano2025accelerators,10.1145/3729215,crossref,A Survey on Deep Learning Hardware Accelerators for Heterogeneous HPC Platforms,exact,author list truncated with 'others'; corpus work 4409568286
32+mutlu2023pim,10.1007/978-981-16-7487-7_7,crossref,A Modern Primer on Processing in Memory,exact,Crossref year 2023; OpenAlex year 2022; corpus work 3111721177
33+bennun2020demystifying,10.1145/3320060,crossref,Demystifying Parallel and Distributed Deep Learning,exact,Crossref title is the short form; print 2020 (v52 n4); OpenAlex year 2019; corpus work 2972087877
34+song2024powerinfer,10.1145/3694715.3695964,crossref,PowerInfer: Fast Large Language Model Serving with a Consumer-grade GPU,exact,SOSP 2024; corpus work 4404401018
35+strati2024orion,10.1145/3627703.3629578,crossref,"Orion: Interference-aware, Fine-grained GPU Sharing for ML Applications",exact,EuroSys 2024; corpus work 4394944658
36+prabhu2025vattention,10.1145/3669940.3707256,crossref,vAttention: Dynamic Memory Management for Serving LLMs without PagedAttention,exact,ASPLOS 2025; corpus work 4407217670
37+lee2020keystone,10.1145/3342195.3387532,crossref,Keystone,exact,Crossref title is the short form; subtitle from the ACM record; corpus work 3021475380
38+jia2021nightcore,10.1145/3445814.3446701,crossref,"Nightcore: efficient and scalable serverless computing for latency-sensitive, interactive microservices",exact,ASPLOS 2021; corpus work 3152809665
39+googlequantum2023suppressing,10.1038/s41586-022-05434-1,crossref,Suppressing quantum errors by scaling a surface code logical qubit,exact,collaboration author (Google Quantum AI); corpus work 4321610982
40+bluvstein2024logical,10.1038/s41586-023-06927-3,crossref,Logical quantum processor based on reconfigurable atom arrays,exact,print 2024 (v626); OpenAlex year 2023; corpus work 4389373471
41+larocca2025barren,10.1038/s42254-025-00813-9,crossref,Barren plateaus in variational quantum computing,exact,corpus work 4408847956
42+main2025distributedqc,10.1038/s41586-024-08404-x,crossref,Distributed quantum computing across an optical network link,exact,corpus work 4407158326
43+aghaee2025aurora,10.1038/s41586-024-08406-9,crossref,Scaling and networking a modular photonic quantum computer,exact,author list truncated with 'others'; corpus work 4406693713
44+javadiabhari2024qiskit,arXiv:2405.08810,arxiv,Quantum computing with Qiskit,exact,preprint; corpus work 4396986598
45+abughanem2025ibmquantum,10.1007/s11227-025-07047-7,crossref,"IBM quantum computers: evolution, performance, and future directions",exact,corpus work 4409126690
46+kanno2026qsci,10.1103/dmn4-snfx,crossref,Quantum-selected configuration interaction: Classical diagonalization of Hamiltonians in subspaces selected by quantum computers,exact,corpus work 4321649614
47+dodge2022carbonintensity,10.1145/3531146.3533234,crossref,Measuring the Carbon Intensity of AI in Cloud Instances,exact,FAccT 2022; corpus work 4283157527
48+li2023thirsty,arXiv:2304.03271,arxiv,"Making AI Less ""Thirsty"": Uncovering and Addressing the Secret Water Footprint of AI Models",exact,preprint; corpus work 4362707075
49+li2025thirsty,10.1145/3724499,crossref,Making AI Less 'Thirsty',exact,CACM version; corpus work 4411368489
50+chien2023carbongenai,10.1145/3604930.3605705,crossref,Reducing the Carbon Impact of Generative AI Inference (today and in 2035),exact,HotCarbon 2023 workshop; corpus work 4385481791
51+boloncanedo2024greenai,10.1016/j.neucom.2024.128096,crossref,A review of green artificial intelligence: Towards a more sustainable future,exact,corpus work 4399913870
52+devriesgao2026datacenters,10.1016/j.patter.2025.101430,crossref,The carbon and water footprints of data centers and what this could mean for artificial intelligence,exact,print 2026 (v7 n1); OpenAlex year 2025; corpus work 7115707641
53+thakur2025greenfl,10.1145/3718363,crossref,Green Federated Learning: A New Era of Green Aware AI,exact,corpus work 4407592994
54+luccioni2025jevons,10.1145/3715275.3732007,crossref,From Efficiency Gains to Rebound Effects: The Problem of Jevons' Paradox in AI's Polarized Environmental Debate,exact,FAccT 2025; corpus work 4411541921
55+chen2026aielectricity,10.1016/j.ynexs.2026.100162,crossref,Electricity demand and grid impacts of AI data centers: Challenges and prospects,exact,corpus work 4415057924
56+niu2026tokenpowerbench,10.1609/aaai.v40i38.40535,crossref,TokenPowerBench: Benchmarking the Power Consumption of LLM Inference,exact,AAAI 2026; corpus work 7138433923
57+zhang2019dlwireless,10.1109/COMST.2019.2904897,crossref,Deep Learning in Mobile and Wireless Networking: A Survey,exact,corpus work 2962883549
58+ietf2024rfc9669,10.17487/RFC9669,crossref,BPF Instruction Set Architecture (ISA),exact,RFC 9669; Crossref record has no author field; corpus work 4403931219
59+liao2026caps,10.1109/CISCE69494.2026.11504731,crossref,Multi-Objective Scheduling for Large Language Model Inference with Prompt-Level Cost Prediction and SLO Awareness,exact,corpus work 7160863705
60+li2026deepservecb,10.1109/GAIIS69281.2026.11519206,crossref,DeepServeCB: SLO-Aware and Cost-Aware Elastic Scheduling for Serverless Multi-Tenant LLM Inference,exact,corpus work 7161556515
61+an2026aiflow,10.1007/s44336-025-00031-y,crossref,"AI Flow: perspectives, scenarios, and approaches",exact,corpus work 7125650829
62+cheikh2026massiveiot,10.1109/JIOT.2026.3655796,crossref,"Energy, Scalability, Data, and Security in Massive IoT: Current Landscape and Future Directions",exact,corpus work 7125196921
63+kumar2026sustainabledc,10.1016/j.rser.2025.116664,crossref,"Toward sustainable data center operation: A review on existing infrastructures, integrated smart energy management, and emerging opportunities",exact,corpus work 7117963061
added manuscript/references_verification_part07b.csv +54 −0
@@ -0,0 +1,54 @@
1+key,doi_or_arxiv,verified_via,verified_title,match
2+rescorla2018tls13,10.17487/rfc8446,crossref,"The Transport Layer Security (TLS) Protocol Version 1.3",yes
3+androulaki2018fabric,10.1145/3190508.3190538,crossref,"Hyperledger fabric",yes
4+yang2019fml,10.1145/3298981,crossref,"Federated Machine Learning",yes
5+zhu2020dlg,10.1007/978-3-030-63076-8_2,crossref,"Deep Leakage from Gradients",yes
6+tan2023personalizedfl,10.1109/tnnls.2022.3160699,crossref,"Towards Personalized Federated Learning",yes
7+ye2024heterogeneousfl,10.1145/3625558,crossref,"Heterogeneous Federated Learning: State-of-the-art and Research Challenges",yes
8+zhang2023fedala,10.1609/aaai.v37i9.26330,crossref,"FedALA: Adaptive Local Aggregation for Personalized Federated Learning",yes
9+nguyen2025unlearning,10.1145/3749987,crossref,"A Survey of Machine Unlearning",yes
10+hou2026mcp,10.1145/3796519,crossref,"Model Context Protocol (MCP): Landscape, Security Threats, and Future Research Directions",yes
11+castryck2023sidh,10.1007/978-3-031-30589-4_15,crossref,"An Efficient Key Recovery Attack on SIDH",yes
12+nist2019pqcround1,10.6028/nist.ir.8240,crossref,"Status report on the first round of the NIST post-quantum cryptography standardization process",yes
13+nist2022pqcround3,10.6028/nist.ir.8413,crossref,"Status report on the third round of the NIST Post-Quantum Cryptography Standardization process",yes
14+foley2019bitcoin,10.1093/rfs/hhz015,crossref,"Sex, Drugs, and Bitcoin: How Much Illegal Activity Is Financed through Cryptocurrencies?",yes
15+cong2019smartcontracts,10.1093/rfs/hhz007,crossref,"Blockchain Disruption and Smart Contracts",yes
16+yaga2018blockchain,10.6028/nist.ir.8202,crossref,"Blockchain technology overview",yes
17+wang2019consensus,10.1109/access.2019.2896108,crossref,"A Survey on Consensus Mechanisms and Mining Strategy Management in Blockchain Networks",yes
18+kurakin2018physical,10.1201/9781351251389-8,crossref,"Adversarial Examples in the Physical World",yes
19+thapa2022splitfed,10.1609/aaai.v36i8.20825,crossref,"SplitFed: When Federated Learning Meets Split Learning",yes
20+du2026vulrag,10.1145/3797277,crossref,"vul-rag : Enhancing LLM-based Vulnerability Detection via Knowledge-level RAG",yes
21+ahmed2020tracingapps,10.1109/access.2020.3010226,crossref,"A Survey of COVID-19 Contact Tracing Apps",yes
22+sheller2020flmedicine,10.1038/s41598-020-69250-1,crossref,"Federated learning in medicine: facilitating multi-institutional collaborations without sharing patient data",yes
23+chen2020fedhealth,10.1109/mis.2020.2988604,crossref,"FedHealth: A Federated Transfer Learning Framework for Wearable Healthcare",yes
24+zhang2024fedtgp,10.1609/aaai.v38i15.29617,crossref,"FedTGP: Trainable Global Prototypes with Adaptive-Margin-Enhanced Contrastive Learning for Data and Model Heterogeneity in Federated Learning",yes
25+fu2026dpfl,10.1145/3801079,crossref,"Differentially Private Federated Learning: A Systematic Review",yes
26+jimenezgutierrez2026flsecurity,10.1016/j.inffus.2026.104155,crossref,"On the security and privacy of federated learning: A survey with attacks, defenses, frameworks, applications, and future directions",yes
27+davidson2018institutions,10.1017/s1744137417000200,crossref,"Blockchains and the economic institutions of capitalism",yes
28+mendling2018bpm,10.1145/3183367,crossref,"Blockchains for Business Process Management - Challenges and Opportunities",yes
29+saberi2019supplychain,10.1080/00207543.2018.1533261,crossref,"Blockchain technology and its relationships to sustainable supply chain management",yes
30+conti2018bitcoinsecurity,10.1109/comst.2018.2842460,crossref,"A Survey on Security and Privacy Issues of Bitcoin",yes
31+fernandez2018biot,10.1109/access.2018.2842685,crossref,"A Review on the Use of Blockchain for the Internet of Things",yes
32+huynh2023metaverse,10.1016/j.future.2023.02.008,crossref,"Blockchain for the metaverse: A Review",yes
33+mohammadi2026dpmedical,10.1038/s41746-025-02280-z,crossref,"Differential privacy for medical deep learning: methods, tradeoffs, and deployment implications",yes
34+eden2025flgovernance,10.1038/s41746-025-01836-3,crossref,"A scoping review of the governance of federated learning in healthcare",yes
35+zhu2021noniid,10.1016/j.neucom.2021.07.098,crossref,"Federated learning on non-IID data: A survey",yes
36+zhang2021flsurvey,10.1016/j.knosys.2021.106775,crossref,"A survey on federated learning",yes
37+xu2021flhealth,10.1007/s41666-020-00082-4,crossref,"Federated Learning for Healthcare Informatics",yes
38+yang2024fedas,10.1109/cvpr52733.2024.01139,crossref,"FedAS: Bridging Inconsistency in Personalized Federated Learning",yes
39+ali2026huntgpt,10.3390/telecom7030073,crossref,"HuntGPT: Integrating Machine Learning-Based Anomaly Detection and Explainable AI with Large Language Models (LLMs)",yes
40+bonawitz2017secagg,10.1145/3133956.3133982,crossref,"Practical Secure Aggregation for Privacy-Preserving Machine Learning",yes
41+beullens2022rainbow,10.1007/978-3-031-15979-4_16,crossref,"Breaking Rainbow Takes a Weekend on a Laptop",yes
42+uddin2026phishing,10.1016/j.comnet.2026.112061,crossref,"An explainable transformer-based model for phishing email detection: A large language model approach",yes
43+xu2025cbrfl,10.1016/j.jnca.2025.104165,crossref,"CBRFL: A framework for Committee-based Byzantine-Resilient Federated Learning",yes
44+nist2024transition,10.6028/NIST.IR.8547.ipd,crossref,"Transition to Post-Quantum Cryptography Standards",yes
45+nist2025hqc,10.6028/NIST.IR.8545,crossref,"Status report on the fourth round of the NIST post-quantum cryptography standardization process",yes
46+guo2022blockchainsecurity,10.1016/j.bcra.2022.100067,crossref,"A survey on blockchain technology and its security",yes
47+kim2018provenance,10.1002/isaf.1424,crossref,"Toward an ontology-driven blockchain design for supply-chain provenance",yes
48+esposito2018healthcare,10.1109/mcc.2018.011791712,crossref,"Blockchain: A Panacea for Healthcare Cloud-Based Data Security and Privacy?",yes
49+li2019qffl,arXiv:1905.10497,arxiv,"Fair Resource Allocation in Federated Learning",yes
50+li2018fedprox,arXiv:1812.06127,arxiv,"Federated Optimization in Heterogeneous Networks",yes
51+wang2021nft,arXiv:2105.07447,arxiv,"Non-Fungible Token (NFT): Overview, Evaluation, Opportunities and Challenges",yes
52+cho2020tracing,arXiv:2003.11511,arxiv,"Contact Tracing Mobile Apps for COVID-19: Privacy Considerations and Related Trade-offs",yes
53+bagdasaryan2018backdoor,arXiv:1807.00459,arxiv,"How To Backdoor Federated Learning",yes
54+guesmi2026reentrancy,10.4230/oasics.fmbc.2026.8,datacite,"Detecting Cross-Function Reentrancy from EVM Traces",yes
modified manuscript/tables/TABLES.json +241 −0
@@ -287,5 +287,246 @@
287 "./results/graph/field_flows_yearly.csv" 287 "./results/graph/field_flows_yearly.csv"
288 ], 288 ],
289 "rows": 9 289 "rows": 9
290+ },
291+ {
292+ "file": "tab_security_counts.tex",
293+ "caption": "Eligible works in the five security-and-privacy categories by publication year: works carrying the category (all) and works whose primary category it is (prim.).",
294+ "label": "tab:security_counts",
295+ "sources": [
296+ "./results/graph/category_counts_yearly.csv"
297+ ],
298+ "rows": 9
299+ },
300+ {
301+ "file": "tab_security_summary.tex",
302+ "caption": "Growth and structural-change summary for the five security-and-privacy categories (multi-label counts).",
303+ "label": "tab:security_summary",
304+ "sources": [
305+ "./results/graph/category_counts_yearly.csv",
306+ "./results/econometrics/breaks.csv"
307+ ],
308+ "rows": 5
309+ },
310+ {
311+ "file": "tab_security_llm_cooccurrence.tex",
312+ "caption": "Co-classification of security-and-privacy works with the LLM, AI-safety and AI-agents categories, by publication year: number of works carrying both categories and share of the security category's works.",
313+ "label": "tab:security_llm_cooccurrence",
314+ "sources": [
315+ "./results/graph/category_cooccurrence_yearly.csv",
316+ "./results/graph/category_counts_yearly.csv"
317+ ],
318+ "rows": 9
319+ },
320+ {
321+ "file": "tab_security_landmarks.tex",
322+ "caption": "Composition of the automatically selected landmark set of the security-and-privacy area, by category.",
323+ "label": "tab:security_landmarks",
324+ "sources": [
325+ "./literature_synthesis/landmarks/security_privacy_landmarks.csv"
326+ ],
327+ "rows": 5
328+ },
329+ {
330+ "file": "tab_systems_categories.tex",
331+ "caption": "The systems and infrastructure categories of Part VI: primary-category output, compound annual growth, and structural-change estimates of the monthly log-volume series.",
332+ "label": "tab:systems_categories",
333+ "sources": [
334+ "./results/bibliometrics/cagr_by_category.csv",
335+ "./results/econometrics/breaks.csv"
336+ ],
337+ "rows": 11
338+ },
339+ {
340+ "file": "tab_p9_periods.tex",
341+ "caption": "The eligible corpus by protocol period: completeness, source membership and document type.",
342+ "label": "tab:p9-periods",
343+ "sources": [
344+ "./results/bibliometrics/volume_monthly.csv"
345+ ],
346+ "rows": 13
347+ },
348+ {
349+ "file": "tab_p9_classification.tex",
350+ "caption": "Classification coverage by publication year: eligible works, works with a primary category, and multi-label assignments.",
351+ "label": "tab:p9-classification",
352+ "sources": [
353+ "./results/graph/category_counts_yearly.csv",
354+ "./data/metadata/prisma_counts.json"
355+ ],
356+ "rows": 9
357+ },
358+ {
359+ "file": "tab_p9_area_shares.tex",
360+ "caption": "Share of primary-category assignments by top-level area and year (per cent), with the two protocol periods.",
361+ "label": "tab:p9-area-shares",
362+ "sources": [
363+ "./results/graph/category_counts_yearly.csv"
364+ ],
365+ "rows": 11
366+ },
367+ {
368+ "file": "tab_p9_category_periods.tex",
369+ "caption": "The forty categories by protocol period: primary-category output, composition shares, multi-label growth and compound annual growth rates.",
370+ "label": "tab:p9-category-periods",
371+ "sources": [
372+ "./results/graph/category_counts_yearly.csv",
373+ "./results/bibliometrics/cagr_by_category.csv"
374+ ],
375+ "rows": 40
376+ },
377+ {
378+ "file": "tab_p9_ai_shares.tex",
379+ "caption": "Share of primary-category assignments held by the AI and machine-learning categories in 2018, 2022 and 2025 (per cent, bootstrap 95\\% intervals).",
380+ "label": "tab:p9-ai-shares",
381+ "sources": [
382+ "./results/bibliometrics/category_share_ci_yearly.csv"
383+ ],
384+ "rows": 10
385+ },
386+ {
387+ "file": "tab_p9_its_total_events.tex",
388+ "caption": "Interrupted time series of the total monthly series at the reference date and at four alternative event dates (log counts, month-of-year dummies, Newey--West standard errors).",
389+ "label": "tab:p9-its-total-events",
390+ "sources": [
391+ "./results/econometrics/its_results.json"
392+ ],
393+ "rows": 5
394+ },
395+ {
396+ "file": "tab_p9_total_robustness.tex",
397+ "caption": "Robustness diagnostics of the structural-change analysis of the total monthly series.",
398+ "label": "tab:p9-total-robustness",
399+ "sources": [
400+ "./results/econometrics/its_results.json"
401+ ],
402+ "rows": 12
403+ },
404+ {
405+ "file": "tab_p9_its_slopes.tex",
406+ "caption": "ITS slope change at December 2022 by category: pre-period slope, change, HAC and placebo $p$-values, and implied post-period slope (monthly log growth).",
407+ "label": "tab:p9-its-slopes",
408+ "sources": [
409+ "./results/econometrics/its_results.json"
410+ ],
411+ "rows": 40
412+ },
413+ {
414+ "file": "tab_p9_its_levels.tex",
415+ "caption": "ITS level change at December 2022 by category: full-sample and donut estimates, HAC and placebo $p$-values, and the BOCPD change probability over December 2022 to March 2023.",
416+ "label": "tab:p9-its-levels",
417+ "sources": [
418+ "./results/econometrics/its_results.json"
419+ ],
420+ "rows": 40
421+ },
422+ {
423+ "file": "tab_p9_unknown_breaks.tex",
424+ "caption": "Data-driven break and changepoint dates by category, and whether any falls in the H3 window (October 2022 to June 2023).",
425+ "label": "tab:p9-unknown-breaks",
426+ "sources": [
427+ "./results/econometrics/its_results.json"
428+ ],
429+ "rows": 40
430+ },
431+ {
432+ "file": "tab_p9_altdates.tex",
433+ "caption": "ITS level change (HAC $p$) by category at the five pre-specified event dates.",
434+ "label": "tab:p9-altdates",
435+ "sources": [
436+ "./results/econometrics/its_results.json"
437+ ],
438+ "rows": 40
439+ },
440+ {
441+ "file": "tab_p9_h3.tex",
442+ "caption": "Hypothesis H3: break dates and ITS estimates for the eight categories named in the protocol.",
443+ "label": "tab:p9-h3",
444+ "sources": [
445+ "./results/econometrics/its_results.json"
446+ ],
447+ "rows": 8
448+ },
449+ {
450+ "file": "tab_p9_cross_area.tex",
451+ "caption": "Cross-area share of internal references by citing area and citing year (per cent), with the aggregate share and its composition-adjusted counterpart.",
452+ "label": "tab:p9-cross-area",
453+ "sources": [
454+ "./results/graph/cross_area_share_yearly.csv"
455+ ],
456+ "rows": 11
457+ },
458+ {
459+ "file": "tab_p9_modularity.tex",
460+ "caption": "Modularity of the yearly category citation-flow network at the category and area partitions, and the within-category self-share of citations.",
461+ "label": "tab:p9-modularity",
462+ "sources": [
463+ "./results/graph/modularity_yearly.csv"
464+ ],
465+ "rows": 9
466+ },
467+ {
468+ "file": "tab_p9_area_flow_change.tex",
469+ "caption": "Change in the area citation-flow matrix between 2018--2022 and 2023--2026: post-period row share minus baseline row share (percentage points).",
470+ "label": "tab:p9-area-flow-change",
471+ "sources": [
472+ "./results/graph/area_flows_yearly.csv"
473+ ],
474+ "rows": 8
475+ },
476+ {
477+ "file": "tab_p9_cooccurrence.tex",
478+ "caption": "The fifteen most frequent co-classification pairs (works carrying both categories in their label set), 2018--2022 and 2023--2026.",
479+ "label": "tab:p9-cooccurrence",
480+ "sources": [
481+ "./results/graph/category_cooccurrence_yearly.csv"
482+ ],
483+ "rows": 15
484+ },
485+ {
486+ "file": "tab_p9_llm_partners.tex",
487+ "caption": "Categories most often co-assigned with Large Language Models, by span.",
488+ "label": "tab:p9-llm-partners",
489+ "sources": [
490+ "./results/graph/category_cooccurrence_yearly.csv"
491+ ],
492+ "rows": 15
493+ },
494+ {
495+ "file": "tab_p9_openness.tex",
496+ "caption": "Openness and metadata indicators by publication year: arXiv linkage, preprint share, open access, abstract availability and document type (per cent of eligible works).",
497+ "label": "tab:p9-openness",
498+ "sources": [
499+ "./results/bibliometrics/volume_yearly.csv"
500+ ],
501+ "rows": 9
502+ },
503+ {
504+ "file": "tab_p9_landmarks.tex",
505+ "caption": "Composition of the landmark set by area and by publication year.",
506+ "label": "tab:p9-landmarks",
507+ "sources": [
508+ "./literature_synthesis/landmarks/selection_report.json"
509+ ],
510+ "rows": 17
511+ },
512+ {
513+ "file": "tab_p9_emerging_setup.tex",
514+ "caption": "Out-of-time evaluation design of the emerging-topic detector: cutoffs, panel sizes and positive rates.",
515+ "label": "tab:p9-emerging-setup",
516+ "sources": [
517+ "./results/emerging/metrics_h12.json",
518+ "./results/emerging/metrics_h24.json"
519+ ],
520+ "rows": 2
521+ },
522+ {
523+ "file": "tab_p9_emerging_cutoffs.tex",
524+ "caption": "AUC of the three detectors at each test cutoff, by horizon.",
525+ "label": "tab:p9-emerging-cutoffs",
526+ "sources": [
527+ "./results/emerging/metrics_h12.json",
528+ "./results/emerging/metrics_h24.json"
529+ ],
530+ "rows": 8
290 } 531 }
291 ] 532 ]
\ No newline at end of file
modified manuscript/tables/numbers.tex +2302 −0
@@ -4809,3 +4809,2305 @@
4809 \newcommand{\KnownSetMedianLeadMonths}{7} 4809 \newcommand{\KnownSetMedianLeadMonths}{7}
4810 \newcommand{\KnownSetNDetected}{12} 4810 \newcommand{\KnownSetNDetected}{12}
4811 \newcommand{\KnownSetNPositiveLead}{11} 4811 \newcommand{\KnownSetNPositiveLead}{11}
4812+\newcommand{\CagrAiAcceleratorsNTwoZeroOneEight}{818}
4813+\newcommand{\CagrAiAcceleratorsNTwoZeroTwoTwo}{1{,}421}
4814+\newcommand{\CagrAiAcceleratorsNTwoZeroTwoThree}{1{,}599}
4815+\newcommand{\CagrAiAcceleratorsNTwoZeroTwoFive}{1{,}906}
4816+\newcommand{\CagrAiAcceleratorsCagrBaseline}{0.148}
4817+\newcommand{\CagrAiAcceleratorsCagrPost}{0.092}
4818+\newcommand{\CagrAiAcceleratorsShareTwoZeroOneEight}{0.002}
4819+\newcommand{\CagrAiAcceleratorsShareTwoZeroTwoTwo}{0.003}
4820+\newcommand{\CagrAiAcceleratorsShareTwoZeroTwoFive}{0.002}
4821+\newcommand{\CagrAiAgentsNTwoZeroOneEight}{5{,}436}
4822+\newcommand{\CagrAiAgentsNTwoZeroTwoTwo}{4{,}521}
4823+\newcommand{\CagrAiAgentsNTwoZeroTwoThree}{4{,}991}
4824+\newcommand{\CagrAiAgentsNTwoZeroTwoFive}{8{,}819}
4825+\newcommand{\CagrAiAgentsCagrBaseline}{-0.045}
4826+\newcommand{\CagrAiAgentsCagrPost}{0.329}
4827+\newcommand{\CagrAiAgentsShareTwoZeroOneEight}{0.011}
4828+\newcommand{\CagrAiAgentsShareTwoZeroTwoTwo}{0.008}
4829+\newcommand{\CagrAiAgentsShareTwoZeroTwoFive}{0.010}
4830+\newcommand{\CagrAiSafetyNTwoZeroOneEight}{1{,}052}
4831+\newcommand{\CagrAiSafetyNTwoZeroTwoTwo}{2{,}954}
4832+\newcommand{\CagrAiSafetyNTwoZeroTwoThree}{4{,}015}
4833+\newcommand{\CagrAiSafetyNTwoZeroTwoFive}{10{,}885}
4834+\newcommand{\CagrAiSafetyCagrBaseline}{0.294}
4835+\newcommand{\CagrAiSafetyCagrPost}{0.647}
4836+\newcommand{\CagrAiSafetyShareTwoZeroOneEight}{0.002}
4837+\newcommand{\CagrAiSafetyShareTwoZeroTwoTwo}{0.005}
4838+\newcommand{\CagrAiSafetyShareTwoZeroTwoFive}{0.012}
4839+\newcommand{\CagrAlgorithmsNTwoZeroOneEight}{29{,}148}
4840+\newcommand{\CagrAlgorithmsNTwoZeroTwoTwo}{28{,}202}
4841+\newcommand{\CagrAlgorithmsNTwoZeroTwoThree}{28{,}531}
4842+\newcommand{\CagrAlgorithmsNTwoZeroTwoFive}{33{,}795}
4843+\newcommand{\CagrAlgorithmsCagrBaseline}{-0.008}
4844+\newcommand{\CagrAlgorithmsCagrPost}{0.088}
4845+\newcommand{\CagrAlgorithmsShareTwoZeroOneEight}{0.058}
4846+\newcommand{\CagrAlgorithmsShareTwoZeroTwoTwo}{0.050}
4847+\newcommand{\CagrAlgorithmsShareTwoZeroTwoFive}{0.038}
4848+\newcommand{\CagrAppliedCryptographyNTwoZeroOneEight}{10{,}782}
4849+\newcommand{\CagrAppliedCryptographyNTwoZeroTwoTwo}{10{,}299}
4850+\newcommand{\CagrAppliedCryptographyNTwoZeroTwoThree}{11{,}219}
4851+\newcommand{\CagrAppliedCryptographyNTwoZeroTwoFive}{14{,}612}
4852+\newcommand{\CagrAppliedCryptographyCagrBaseline}{-0.011}
4853+\newcommand{\CagrAppliedCryptographyCagrPost}{0.141}
4854+\newcommand{\CagrAppliedCryptographyShareTwoZeroOneEight}{0.022}
4855+\newcommand{\CagrAppliedCryptographyShareTwoZeroTwoTwo}{0.018}
4856+\newcommand{\CagrAppliedCryptographyShareTwoZeroTwoFive}{0.016}
4857+\newcommand{\CagrAutonomousScientificDiscoveryNTwoZeroOneEight}{86}
4858+\newcommand{\CagrAutonomousScientificDiscoveryNTwoZeroTwoTwo}{236}
4859+\newcommand{\CagrAutonomousScientificDiscoveryNTwoZeroTwoThree}{400}
4860+\newcommand{\CagrAutonomousScientificDiscoveryNTwoZeroTwoFive}{764}
4861+\newcommand{\CagrAutonomousScientificDiscoveryCagrBaseline}{0.287}
4862+\newcommand{\CagrAutonomousScientificDiscoveryCagrPost}{0.382}
4863+\newcommand{\CagrAutonomousScientificDiscoveryShareTwoZeroOneEight}{0.000}
4864+\newcommand{\CagrAutonomousScientificDiscoveryShareTwoZeroTwoTwo}{0.000}
4865+\newcommand{\CagrAutonomousScientificDiscoveryShareTwoZeroTwoFive}{0.001}
4866+\newcommand{\CagrBlockchainNTwoZeroOneEight}{6{,}441}
4867+\newcommand{\CagrBlockchainNTwoZeroTwoTwo}{11{,}770}
4868+\newcommand{\CagrBlockchainNTwoZeroTwoThree}{14{,}349}
4869+\newcommand{\CagrBlockchainNTwoZeroTwoFive}{18{,}012}
4870+\newcommand{\CagrBlockchainCagrBaseline}{0.163}
4871+\newcommand{\CagrBlockchainCagrPost}{0.120}
4872+\newcommand{\CagrBlockchainShareTwoZeroOneEight}{0.013}
4873+\newcommand{\CagrBlockchainShareTwoZeroTwoTwo}{0.021}
4874+\newcommand{\CagrBlockchainShareTwoZeroTwoFive}{0.020}
4875+\newcommand{\CagrCloudComputingNTwoZeroOneEight}{9{,}396}
4876+\newcommand{\CagrCloudComputingNTwoZeroTwoTwo}{7{,}103}
4877+\newcommand{\CagrCloudComputingNTwoZeroTwoThree}{7{,}643}
4878+\newcommand{\CagrCloudComputingNTwoZeroTwoFive}{10{,}974}
4879+\newcommand{\CagrCloudComputingCagrBaseline}{-0.068}
4880+\newcommand{\CagrCloudComputingCagrPost}{0.198}
4881+\newcommand{\CagrCloudComputingShareTwoZeroOneEight}{0.019}
4882+\newcommand{\CagrCloudComputingShareTwoZeroTwoTwo}{0.013}
4883+\newcommand{\CagrCloudComputingShareTwoZeroTwoFive}{0.012}
4884+\newcommand{\CagrCompilersNTwoZeroOneEight}{725}
4885+\newcommand{\CagrCompilersNTwoZeroTwoTwo}{642}
4886+\newcommand{\CagrCompilersNTwoZeroTwoThree}{692}
4887+\newcommand{\CagrCompilersNTwoZeroTwoFive}{767}
4888+\newcommand{\CagrCompilersCagrBaseline}{-0.030}
4889+\newcommand{\CagrCompilersCagrPost}{0.053}
4890+\newcommand{\CagrCompilersShareTwoZeroOneEight}{0.001}
4891+\newcommand{\CagrCompilersShareTwoZeroTwoTwo}{0.001}
4892+\newcommand{\CagrCompilersShareTwoZeroTwoFive}{0.001}
4893+\newcommand{\CagrComputationalBiologyNTwoZeroOneEight}{12{,}968}
4894+\newcommand{\CagrComputationalBiologyNTwoZeroTwoTwo}{19{,}536}
4895+\newcommand{\CagrComputationalBiologyNTwoZeroTwoThree}{20{,}576}
4896+\newcommand{\CagrComputationalBiologyNTwoZeroTwoFive}{28{,}154}
4897+\newcommand{\CagrComputationalBiologyCagrBaseline}{0.108}
4898+\newcommand{\CagrComputationalBiologyCagrPost}{0.170}
4899+\newcommand{\CagrComputationalBiologyShareTwoZeroOneEight}{0.026}
4900+\newcommand{\CagrComputationalBiologyShareTwoZeroTwoTwo}{0.035}
4901+\newcommand{\CagrComputationalBiologyShareTwoZeroTwoFive}{0.032}
4902+\newcommand{\CagrComputationalComplexityNTwoZeroOneEight}{3{,}837}
4903+\newcommand{\CagrComputationalComplexityNTwoZeroTwoTwo}{3{,}541}
4904+\newcommand{\CagrComputationalComplexityNTwoZeroTwoThree}{3{,}499}
4905+\newcommand{\CagrComputationalComplexityNTwoZeroTwoFive}{4{,}209}
4906+\newcommand{\CagrComputationalComplexityCagrBaseline}{-0.020}
4907+\newcommand{\CagrComputationalComplexityCagrPost}{0.097}
4908+\newcommand{\CagrComputationalComplexityShareTwoZeroOneEight}{0.008}
4909+\newcommand{\CagrComputationalComplexityShareTwoZeroTwoTwo}{0.006}
4910+\newcommand{\CagrComputationalComplexityShareTwoZeroTwoFive}{0.005}
4911+\newcommand{\CagrComputerArchitectureNTwoZeroOneEight}{19{,}554}
4912+\newcommand{\CagrComputerArchitectureNTwoZeroTwoTwo}{15{,}174}
4913+\newcommand{\CagrComputerArchitectureNTwoZeroTwoThree}{16{,}369}
4914+\newcommand{\CagrComputerArchitectureNTwoZeroTwoFive}{20{,}209}
4915+\newcommand{\CagrComputerArchitectureCagrBaseline}{-0.061}
4916+\newcommand{\CagrComputerArchitectureCagrPost}{0.111}
4917+\newcommand{\CagrComputerArchitectureShareTwoZeroOneEight}{0.039}
4918+\newcommand{\CagrComputerArchitectureShareTwoZeroTwoTwo}{0.027}
4919+\newcommand{\CagrComputerArchitectureShareTwoZeroTwoFive}{0.023}
4920+\newcommand{\CagrComputerGraphicsNTwoZeroOneEight}{10{,}198}
4921+\newcommand{\CagrComputerGraphicsNTwoZeroTwoTwo}{8{,}287}
4922+\newcommand{\CagrComputerGraphicsNTwoZeroTwoThree}{9{,}297}
4923+\newcommand{\CagrComputerGraphicsNTwoZeroTwoFive}{12{,}440}
4924+\newcommand{\CagrComputerGraphicsCagrBaseline}{-0.051}
4925+\newcommand{\CagrComputerGraphicsCagrPost}{0.157}
4926+\newcommand{\CagrComputerGraphicsShareTwoZeroOneEight}{0.020}
4927+\newcommand{\CagrComputerGraphicsShareTwoZeroTwoTwo}{0.015}
4928+\newcommand{\CagrComputerGraphicsShareTwoZeroTwoFive}{0.014}
4929+\newcommand{\CagrComputerNetworksNTwoZeroOneEight}{58{,}744}
4930+\newcommand{\CagrComputerNetworksNTwoZeroTwoTwo}{49{,}141}
4931+\newcommand{\CagrComputerNetworksNTwoZeroTwoThree}{51{,}881}
4932+\newcommand{\CagrComputerNetworksNTwoZeroTwoFive}{59{,}555}
4933+\newcommand{\CagrComputerNetworksCagrBaseline}{-0.044}
4934+\newcommand{\CagrComputerNetworksCagrPost}{0.071}
4935+\newcommand{\CagrComputerNetworksShareTwoZeroOneEight}{0.118}
4936+\newcommand{\CagrComputerNetworksShareTwoZeroTwoTwo}{0.088}
4937+\newcommand{\CagrComputerNetworksShareTwoZeroTwoFive}{0.067}
4938+\newcommand{\CagrComputerVisionNTwoZeroOneEight}{69{,}738}
4939+\newcommand{\CagrComputerVisionNTwoZeroTwoTwo}{98{,}704}
4940+\newcommand{\CagrComputerVisionNTwoZeroTwoThree}{114{,}261}
4941+\newcommand{\CagrComputerVisionNTwoZeroTwoFive}{150{,}757}
4942+\newcommand{\CagrComputerVisionCagrBaseline}{0.091}
4943+\newcommand{\CagrComputerVisionCagrPost}{0.149}
4944+\newcommand{\CagrComputerVisionShareTwoZeroOneEight}{0.140}
4945+\newcommand{\CagrComputerVisionShareTwoZeroTwoTwo}{0.176}
4946+\newcommand{\CagrComputerVisionShareTwoZeroTwoFive}{0.169}
4947+\newcommand{\CagrCybersecurityNTwoZeroOneEight}{29{,}440}
4948+\newcommand{\CagrCybersecurityNTwoZeroTwoTwo}{31{,}554}
4949+\newcommand{\CagrCybersecurityNTwoZeroTwoThree}{36{,}481}
4950+\newcommand{\CagrCybersecurityNTwoZeroTwoFive}{59{,}044}
4951+\newcommand{\CagrCybersecurityCagrBaseline}{0.017}
4952+\newcommand{\CagrCybersecurityCagrPost}{0.272}
4953+\newcommand{\CagrCybersecurityShareTwoZeroOneEight}{0.059}
4954+\newcommand{\CagrCybersecurityShareTwoZeroTwoTwo}{0.056}
4955+\newcommand{\CagrCybersecurityShareTwoZeroTwoFive}{0.066}
4956+\newcommand{\CagrDatabaseSystemsNTwoZeroOneEight}{16{,}265}
4957+\newcommand{\CagrDatabaseSystemsNTwoZeroTwoTwo}{13{,}625}
4958+\newcommand{\CagrDatabaseSystemsNTwoZeroTwoThree}{14{,}487}
4959+\newcommand{\CagrDatabaseSystemsNTwoZeroTwoFive}{16{,}681}
4960+\newcommand{\CagrDatabaseSystemsCagrBaseline}{-0.043}
4961+\newcommand{\CagrDatabaseSystemsCagrPost}{0.073}
4962+\newcommand{\CagrDatabaseSystemsShareTwoZeroOneEight}{0.033}
4963+\newcommand{\CagrDatabaseSystemsShareTwoZeroTwoTwo}{0.024}
4964+\newcommand{\CagrDatabaseSystemsShareTwoZeroTwoFive}{0.019}
4965+\newcommand{\CagrDeepLearningArchitecturesNTwoZeroOneEight}{11{,}992}
4966+\newcommand{\CagrDeepLearningArchitecturesNTwoZeroTwoTwo}{22{,}270}
4967+\newcommand{\CagrDeepLearningArchitecturesNTwoZeroTwoThree}{25{,}607}
4968+\newcommand{\CagrDeepLearningArchitecturesNTwoZeroTwoFive}{34{,}141}
4969+\newcommand{\CagrDeepLearningArchitecturesCagrBaseline}{0.167}
4970+\newcommand{\CagrDeepLearningArchitecturesCagrPost}{0.155}
4971+\newcommand{\CagrDeepLearningArchitecturesShareTwoZeroOneEight}{0.024}
4972+\newcommand{\CagrDeepLearningArchitecturesShareTwoZeroTwoTwo}{0.040}
4973+\newcommand{\CagrDeepLearningArchitecturesShareTwoZeroTwoFive}{0.038}
4974+\newcommand{\CagrDistributedComputingNTwoZeroOneEight}{11{,}519}
4975+\newcommand{\CagrDistributedComputingNTwoZeroTwoTwo}{9{,}379}
4976+\newcommand{\CagrDistributedComputingNTwoZeroTwoThree}{9{,}798}
4977+\newcommand{\CagrDistributedComputingNTwoZeroTwoFive}{12{,}550}
4978+\newcommand{\CagrDistributedComputingCagrBaseline}{-0.050}
4979+\newcommand{\CagrDistributedComputingCagrPost}{0.132}
4980+\newcommand{\CagrDistributedComputingShareTwoZeroOneEight}{0.023}
4981+\newcommand{\CagrDistributedComputingShareTwoZeroTwoTwo}{0.017}
4982+\newcommand{\CagrDistributedComputingShareTwoZeroTwoFive}{0.014}
4983+\newcommand{\CagrEdgeComputingNTwoZeroOneEight}{5{,}185}
4984+\newcommand{\CagrEdgeComputingNTwoZeroTwoTwo}{8{,}554}
4985+\newcommand{\CagrEdgeComputingNTwoZeroTwoThree}{9{,}120}
4986+\newcommand{\CagrEdgeComputingNTwoZeroTwoFive}{11{,}554}
4987+\newcommand{\CagrEdgeComputingCagrBaseline}{0.133}
4988+\newcommand{\CagrEdgeComputingCagrPost}{0.126}
4989+\newcommand{\CagrEdgeComputingShareTwoZeroOneEight}{0.010}
4990+\newcommand{\CagrEdgeComputingShareTwoZeroTwoTwo}{0.015}
4991+\newcommand{\CagrEdgeComputingShareTwoZeroTwoFive}{0.013}
4992+\newcommand{\CagrFederatedLearningNTwoZeroOneEight}{143}
4993+\newcommand{\CagrFederatedLearningNTwoZeroTwoTwo}{974}
4994+\newcommand{\CagrFederatedLearningNTwoZeroTwoThree}{1{,}497}
4995+\newcommand{\CagrFederatedLearningNTwoZeroTwoFive}{2{,}544}
4996+\newcommand{\CagrFederatedLearningCagrBaseline}{0.615}
4997+\newcommand{\CagrFederatedLearningCagrPost}{0.304}
4998+\newcommand{\CagrFederatedLearningShareTwoZeroOneEight}{0.000}
4999+\newcommand{\CagrFederatedLearningShareTwoZeroTwoTwo}{0.002}
5000+\newcommand{\CagrFederatedLearningShareTwoZeroTwoFive}{0.003}
5001+\newcommand{\CagrFormalVerificationNTwoZeroOneEight}{5{,}307}
5002+\newcommand{\CagrFormalVerificationNTwoZeroTwoTwo}{3{,}995}
5003+\newcommand{\CagrFormalVerificationNTwoZeroTwoThree}{4{,}385}
5004+\newcommand{\CagrFormalVerificationNTwoZeroTwoFive}{4{,}623}
5005+\newcommand{\CagrFormalVerificationCagrBaseline}{-0.069}
5006+\newcommand{\CagrFormalVerificationCagrPost}{0.027}
5007+\newcommand{\CagrFormalVerificationShareTwoZeroOneEight}{0.011}
5008+\newcommand{\CagrFormalVerificationShareTwoZeroTwoTwo}{0.007}
5009+\newcommand{\CagrFormalVerificationShareTwoZeroTwoFive}{0.005}
5010+\newcommand{\CagrFoundationModelsNTwoZeroOneEight}{46}
5011+\newcommand{\CagrFoundationModelsNTwoZeroTwoTwo}{350}
5012+\newcommand{\CagrFoundationModelsNTwoZeroTwoThree}{571}
5013+\newcommand{\CagrFoundationModelsNTwoZeroTwoFive}{1{,}854}
5014+\newcommand{\CagrFoundationModelsCagrBaseline}{0.661}
5015+\newcommand{\CagrFoundationModelsCagrPost}{0.802}
5016+\newcommand{\CagrFoundationModelsShareTwoZeroOneEight}{0.000}
5017+\newcommand{\CagrFoundationModelsShareTwoZeroTwoTwo}{0.001}
5018+\newcommand{\CagrFoundationModelsShareTwoZeroTwoFive}{0.002}
5019+\newcommand{\CagrGenerativeAiNTwoZeroOneEight}{766}
5020+\newcommand{\CagrGenerativeAiNTwoZeroTwoTwo}{2{,}160}
5021+\newcommand{\CagrGenerativeAiNTwoZeroTwoThree}{3{,}646}
5022+\newcommand{\CagrGenerativeAiNTwoZeroTwoFive}{12{,}743}
5023+\newcommand{\CagrGenerativeAiCagrBaseline}{0.296}
5024+\newcommand{\CagrGenerativeAiCagrPost}{0.870}
5025+\newcommand{\CagrGenerativeAiShareTwoZeroOneEight}{0.002}
5026+\newcommand{\CagrGenerativeAiShareTwoZeroTwoTwo}{0.004}
5027+\newcommand{\CagrGenerativeAiShareTwoZeroTwoFive}{0.014}
5028+\newcommand{\CagrHciNTwoZeroOneEight}{34{,}225}
5029+\newcommand{\CagrHciNTwoZeroTwoTwo}{34{,}801}
5030+\newcommand{\CagrHciNTwoZeroTwoThree}{40{,}587}
5031+\newcommand{\CagrHciNTwoZeroTwoFive}{58{,}148}
5032+\newcommand{\CagrHciCagrBaseline}{0.004}
5033+\newcommand{\CagrHciCagrPost}{0.197}
5034+\newcommand{\CagrHciShareTwoZeroOneEight}{0.069}
5035+\newcommand{\CagrHciShareTwoZeroTwoTwo}{0.062}
5036+\newcommand{\CagrHciShareTwoZeroTwoFive}{0.065}
5037+\newcommand{\CagrInformationRetrievalNTwoZeroOneEight}{12{,}095}
5038+\newcommand{\CagrInformationRetrievalNTwoZeroTwoTwo}{11{,}243}
5039+\newcommand{\CagrInformationRetrievalNTwoZeroTwoThree}{12{,}568}
5040+\newcommand{\CagrInformationRetrievalNTwoZeroTwoFive}{17{,}379}
5041+\newcommand{\CagrInformationRetrievalCagrBaseline}{-0.018}
5042+\newcommand{\CagrInformationRetrievalCagrPost}{0.176}
5043+\newcommand{\CagrInformationRetrievalShareTwoZeroOneEight}{0.024}
5044+\newcommand{\CagrInformationRetrievalShareTwoZeroTwoTwo}{0.020}
5045+\newcommand{\CagrInformationRetrievalShareTwoZeroTwoFive}{0.019}
5046+\newcommand{\CagrLlmNTwoZeroOneEight}{1{,}635}
5047+\newcommand{\CagrLlmNTwoZeroTwoTwo}{3{,}266}
5048+\newcommand{\CagrLlmNTwoZeroTwoThree}{10{,}389}
5049+\newcommand{\CagrLlmNTwoZeroTwoFive}{45{,}158}
5050+\newcommand{\CagrLlmCagrBaseline}{0.189}
5051+\newcommand{\CagrLlmCagrPost}{1.085}
5052+\newcommand{\CagrLlmShareTwoZeroOneEight}{0.003}
5053+\newcommand{\CagrLlmShareTwoZeroTwoTwo}{0.006}
5054+\newcommand{\CagrLlmShareTwoZeroTwoFive}{0.051}
5055+\newcommand{\CagrMlTheoryNTwoZeroOneEight}{13{,}534}
5056+\newcommand{\CagrMlTheoryNTwoZeroTwoTwo}{20{,}131}
5057+\newcommand{\CagrMlTheoryNTwoZeroTwoThree}{22{,}893}
5058+\newcommand{\CagrMlTheoryNTwoZeroTwoFive}{32{,}798}
5059+\newcommand{\CagrMlTheoryCagrBaseline}{0.104}
5060+\newcommand{\CagrMlTheoryCagrPost}{0.197}
5061+\newcommand{\CagrMlTheoryShareTwoZeroOneEight}{0.027}
5062+\newcommand{\CagrMlTheoryShareTwoZeroTwoTwo}{0.036}
5063+\newcommand{\CagrMlTheoryShareTwoZeroTwoFive}{0.037}
5064+\newcommand{\CagrMultimodalAiNTwoZeroOneEight}{470}
5065+\newcommand{\CagrMultimodalAiNTwoZeroTwoTwo}{1{,}278}
5066+\newcommand{\CagrMultimodalAiNTwoZeroTwoThree}{1{,}932}
5067+\newcommand{\CagrMultimodalAiNTwoZeroTwoFive}{6{,}749}
5068+\newcommand{\CagrMultimodalAiCagrBaseline}{0.284}
5069+\newcommand{\CagrMultimodalAiCagrPost}{0.869}
5070+\newcommand{\CagrMultimodalAiShareTwoZeroOneEight}{0.001}
5071+\newcommand{\CagrMultimodalAiShareTwoZeroTwoTwo}{0.002}
5072+\newcommand{\CagrMultimodalAiShareTwoZeroTwoFive}{0.008}
5073+\newcommand{\CagrNlpNTwoZeroOneEight}{23{,}372}
5074+\newcommand{\CagrNlpNTwoZeroTwoTwo}{29{,}678}
5075+\newcommand{\CagrNlpNTwoZeroTwoThree}{33{,}965}
5076+\newcommand{\CagrNlpNTwoZeroTwoFive}{42{,}914}
5077+\newcommand{\CagrNlpCagrBaseline}{0.062}
5078+\newcommand{\CagrNlpCagrPost}{0.124}
5079+\newcommand{\CagrNlpShareTwoZeroOneEight}{0.047}
5080+\newcommand{\CagrNlpShareTwoZeroTwoTwo}{0.053}
5081+\newcommand{\CagrNlpShareTwoZeroTwoFive}{0.048}
5082+\newcommand{\CagrOperatingSystemsNTwoZeroOneEight}{1{,}556}
5083+\newcommand{\CagrOperatingSystemsNTwoZeroTwoTwo}{1{,}083}
5084+\newcommand{\CagrOperatingSystemsNTwoZeroTwoThree}{1{,}173}
5085+\newcommand{\CagrOperatingSystemsNTwoZeroTwoFive}{1{,}363}
5086+\newcommand{\CagrOperatingSystemsCagrBaseline}{-0.087}
5087+\newcommand{\CagrOperatingSystemsCagrPost}{0.078}
5088+\newcommand{\CagrOperatingSystemsShareTwoZeroOneEight}{0.003}
5089+\newcommand{\CagrOperatingSystemsShareTwoZeroTwoTwo}{0.002}
5090+\newcommand{\CagrOperatingSystemsShareTwoZeroTwoFive}{0.002}
5091+\newcommand{\CagrPrivacyPreservingComputationNTwoZeroOneEight}{3{,}105}
5092+\newcommand{\CagrPrivacyPreservingComputationNTwoZeroTwoTwo}{4{,}376}
5093+\newcommand{\CagrPrivacyPreservingComputationNTwoZeroTwoThree}{5{,}010}
5094+\newcommand{\CagrPrivacyPreservingComputationNTwoZeroTwoFive}{8{,}817}
5095+\newcommand{\CagrPrivacyPreservingComputationCagrBaseline}{0.090}
5096+\newcommand{\CagrPrivacyPreservingComputationCagrPost}{0.327}
5097+\newcommand{\CagrPrivacyPreservingComputationShareTwoZeroOneEight}{0.006}
5098+\newcommand{\CagrPrivacyPreservingComputationShareTwoZeroTwoTwo}{0.008}
5099+\newcommand{\CagrPrivacyPreservingComputationShareTwoZeroTwoFive}{0.010}
5100+\newcommand{\CagrProgrammingLanguagesNTwoZeroOneEight}{4{,}390}
5101+\newcommand{\CagrProgrammingLanguagesNTwoZeroTwoTwo}{3{,}664}
5102+\newcommand{\CagrProgrammingLanguagesNTwoZeroTwoThree}{3{,}876}
5103+\newcommand{\CagrProgrammingLanguagesNTwoZeroTwoFive}{4{,}193}
5104+\newcommand{\CagrProgrammingLanguagesCagrBaseline}{-0.044}
5105+\newcommand{\CagrProgrammingLanguagesCagrPost}{0.040}
5106+\newcommand{\CagrProgrammingLanguagesShareTwoZeroOneEight}{0.009}
5107+\newcommand{\CagrProgrammingLanguagesShareTwoZeroTwoTwo}{0.007}
5108+\newcommand{\CagrProgrammingLanguagesShareTwoZeroTwoFive}{0.005}
5109+\newcommand{\CagrQuantumComputingNTwoZeroOneEight}{6{,}224}
5110+\newcommand{\CagrQuantumComputingNTwoZeroTwoTwo}{9{,}216}
5111+\newcommand{\CagrQuantumComputingNTwoZeroTwoThree}{10{,}544}
5112+\newcommand{\CagrQuantumComputingNTwoZeroTwoFive}{17{,}471}
5113+\newcommand{\CagrQuantumComputingCagrBaseline}{0.103}
5114+\newcommand{\CagrQuantumComputingCagrPost}{0.287}
5115+\newcommand{\CagrQuantumComputingShareTwoZeroOneEight}{0.012}
5116+\newcommand{\CagrQuantumComputingShareTwoZeroTwoTwo}{0.016}
5117+\newcommand{\CagrQuantumComputingShareTwoZeroTwoFive}{0.020}
5118+\newcommand{\CagrReinforcementLearningNTwoZeroOneEight}{3{,}296}
5119+\newcommand{\CagrReinforcementLearningNTwoZeroTwoTwo}{7{,}017}
5120+\newcommand{\CagrReinforcementLearningNTwoZeroTwoThree}{8{,}422}
5121+\newcommand{\CagrReinforcementLearningNTwoZeroTwoFive}{13{,}210}
5122+\newcommand{\CagrReinforcementLearningCagrBaseline}{0.208}
5123+\newcommand{\CagrReinforcementLearningCagrPost}{0.252}
5124+\newcommand{\CagrReinforcementLearningShareTwoZeroOneEight}{0.007}
5125+\newcommand{\CagrReinforcementLearningShareTwoZeroTwoTwo}{0.013}
5126+\newcommand{\CagrReinforcementLearningShareTwoZeroTwoFive}{0.015}
5127+\newcommand{\CagrRoboticsNTwoZeroOneEight}{17{,}496}
5128+\newcommand{\CagrRoboticsNTwoZeroTwoTwo}{23{,}103}
5129+\newcommand{\CagrRoboticsNTwoZeroTwoThree}{25{,}759}
5130+\newcommand{\CagrRoboticsNTwoZeroTwoFive}{33{,}464}
5131+\newcommand{\CagrRoboticsCagrBaseline}{0.072}
5132+\newcommand{\CagrRoboticsCagrPost}{0.140}
5133+\newcommand{\CagrRoboticsShareTwoZeroOneEight}{0.035}
5134+\newcommand{\CagrRoboticsShareTwoZeroTwoTwo}{0.041}
5135+\newcommand{\CagrRoboticsShareTwoZeroTwoFive}{0.037}
5136+\newcommand{\CagrScientificComputingNTwoZeroOneEight}{16{,}882}
5137+\newcommand{\CagrScientificComputingNTwoZeroTwoTwo}{19{,}932}
5138+\newcommand{\CagrScientificComputingNTwoZeroTwoThree}{20{,}423}
5139+\newcommand{\CagrScientificComputingNTwoZeroTwoFive}{23{,}191}
5140+\newcommand{\CagrScientificComputingCagrBaseline}{0.042}
5141+\newcommand{\CagrScientificComputingCagrPost}{0.066}
5142+\newcommand{\CagrScientificComputingShareTwoZeroOneEight}{0.034}
5143+\newcommand{\CagrScientificComputingShareTwoZeroTwoTwo}{0.036}
5144+\newcommand{\CagrScientificComputingShareTwoZeroTwoFive}{0.026}
5145+\newcommand{\CagrSoftwareEngineeringNTwoZeroOneEight}{29{,}016}
5146+\newcommand{\CagrSoftwareEngineeringNTwoZeroTwoTwo}{23{,}725}
5147+\newcommand{\CagrSoftwareEngineeringNTwoZeroTwoThree}{26{,}245}
5148+\newcommand{\CagrSoftwareEngineeringNTwoZeroTwoFive}{36{,}381}
5149+\newcommand{\CagrSoftwareEngineeringCagrBaseline}{-0.049}
5150+\newcommand{\CagrSoftwareEngineeringCagrPost}{0.177}
5151+\newcommand{\CagrSoftwareEngineeringShareTwoZeroOneEight}{0.058}
5152+\newcommand{\CagrSoftwareEngineeringShareTwoZeroTwoTwo}{0.042}
5153+\newcommand{\CagrSoftwareEngineeringShareTwoZeroTwoFive}{0.041}
5154+\newcommand{\CagrSpeechAudioNTwoZeroOneEight}{12{,}096}
5155+\newcommand{\CagrSpeechAudioNTwoZeroTwoTwo}{12{,}698}
5156+\newcommand{\CagrSpeechAudioNTwoZeroTwoThree}{14{,}688}
5157+\newcommand{\CagrSpeechAudioNTwoZeroTwoFive}{18{,}373}
5158+\newcommand{\CagrSpeechAudioCagrBaseline}{0.012}
5159+\newcommand{\CagrSpeechAudioCagrPost}{0.118}
5160+\newcommand{\CagrSpeechAudioShareTwoZeroOneEight}{0.024}
5161+\newcommand{\CagrSpeechAudioShareTwoZeroTwoTwo}{0.023}
5162+\newcommand{\CagrSpeechAudioShareTwoZeroTwoFive}{0.021}
5163+\newcommand{\CagrSustainableComputingNTwoZeroOneEight}{583}
5164+\newcommand{\CagrSustainableComputingNTwoZeroTwoTwo}{830}
5165+\newcommand{\CagrSustainableComputingNTwoZeroTwoThree}{1{,}078}
5166+\newcommand{\CagrSustainableComputingNTwoZeroTwoFive}{1{,}639}
5167+\newcommand{\CagrSustainableComputingCagrBaseline}{0.092}
5168+\newcommand{\CagrSustainableComputingCagrPost}{0.233}
5169+\newcommand{\CagrSustainableComputingShareTwoZeroOneEight}{0.001}
5170+\newcommand{\CagrSustainableComputingShareTwoZeroTwoTwo}{0.001}
5171+\newcommand{\CagrSustainableComputingShareTwoZeroTwoFive}{0.002}
5172+\newcommand{\CatcountAiAcceleratorsTwoZeroOneEightPrimary}{818}
5173+\newcommand{\CatcountAiAcceleratorsTwoZeroOneNinePrimary}{992}
5174+\newcommand{\CatcountAiAcceleratorsTwoZeroTwoZeroPrimary}{1{,}085}
5175+\newcommand{\CatcountAiAcceleratorsTwoZeroTwoOnePrimary}{1{,}249}
5176+\newcommand{\CatcountAiAcceleratorsTwoZeroTwoTwoPrimary}{1{,}421}
5177+\newcommand{\CatcountAiAcceleratorsTwoZeroTwoThreePrimary}{1{,}599}
5178+\newcommand{\CatcountAiAcceleratorsTwoZeroTwoFourPrimary}{1{,}720}
5179+\newcommand{\CatcountAiAcceleratorsTwoZeroTwoFivePrimary}{1{,}906}
5180+\newcommand{\CatcountAiAcceleratorsTwoZeroTwoSixPrimary}{1{,}649}
5181+\newcommand{\CatcountAiAgentsTwoZeroOneEightPrimary}{5{,}436}
5182+\newcommand{\CatcountAiAgentsTwoZeroOneNinePrimary}{5{,}506}
5183+\newcommand{\CatcountAiAgentsTwoZeroTwoZeroPrimary}{5{,}129}
5184+\newcommand{\CatcountAiAgentsTwoZeroTwoOnePrimary}{4{,}892}
5185+\newcommand{\CatcountAiAgentsTwoZeroTwoTwoPrimary}{4{,}521}
5186+\newcommand{\CatcountAiAgentsTwoZeroTwoThreePrimary}{4{,}991}
5187+\newcommand{\CatcountAiAgentsTwoZeroTwoFourPrimary}{5{,}899}
5188+\newcommand{\CatcountAiAgentsTwoZeroTwoFivePrimary}{8{,}819}
5189+\newcommand{\CatcountAiAgentsTwoZeroTwoSixPrimary}{12{,}209}
5190+\newcommand{\CatcountAiSafetyTwoZeroOneEightPrimary}{1{,}052}
5191+\newcommand{\CatcountAiSafetyTwoZeroOneNinePrimary}{1{,}479}
5192+\newcommand{\CatcountAiSafetyTwoZeroTwoZeroPrimary}{2{,}154}
5193+\newcommand{\CatcountAiSafetyTwoZeroTwoOnePrimary}{2{,}494}
5194+\newcommand{\CatcountAiSafetyTwoZeroTwoTwoPrimary}{2{,}954}
5195+\newcommand{\CatcountAiSafetyTwoZeroTwoThreePrimary}{4{,}015}
5196+\newcommand{\CatcountAiSafetyTwoZeroTwoFourPrimary}{6{,}217}
5197+\newcommand{\CatcountAiSafetyTwoZeroTwoFivePrimary}{10{,}885}
5198+\newcommand{\CatcountAiSafetyTwoZeroTwoSixPrimary}{12{,}544}
5199+\newcommand{\CatcountAlgorithmsTwoZeroOneEightPrimary}{29{,}148}
5200+\newcommand{\CatcountAlgorithmsTwoZeroOneNinePrimary}{29{,}252}
5201+\newcommand{\CatcountAlgorithmsTwoZeroTwoZeroPrimary}{29{,}376}
5202+\newcommand{\CatcountAlgorithmsTwoZeroTwoOnePrimary}{29{,}023}
5203+\newcommand{\CatcountAlgorithmsTwoZeroTwoTwoPrimary}{28{,}202}
5204+\newcommand{\CatcountAlgorithmsTwoZeroTwoThreePrimary}{28{,}531}
5205+\newcommand{\CatcountAlgorithmsTwoZeroTwoFourPrimary}{29{,}957}
5206+\newcommand{\CatcountAlgorithmsTwoZeroTwoFivePrimary}{33{,}795}
5207+\newcommand{\CatcountAlgorithmsTwoZeroTwoSixPrimary}{28{,}776}
5208+\newcommand{\CatcountAppliedCryptographyTwoZeroOneEightPrimary}{10{,}782}
5209+\newcommand{\CatcountAppliedCryptographyTwoZeroOneNinePrimary}{11{,}355}
5210+\newcommand{\CatcountAppliedCryptographyTwoZeroTwoZeroPrimary}{11{,}279}
5211+\newcommand{\CatcountAppliedCryptographyTwoZeroTwoOnePrimary}{10{,}924}
5212+\newcommand{\CatcountAppliedCryptographyTwoZeroTwoTwoPrimary}{10{,}299}
5213+\newcommand{\CatcountAppliedCryptographyTwoZeroTwoThreePrimary}{11{,}219}
5214+\newcommand{\CatcountAppliedCryptographyTwoZeroTwoFourPrimary}{12{,}426}
5215+\newcommand{\CatcountAppliedCryptographyTwoZeroTwoFivePrimary}{14{,}612}
5216+\newcommand{\CatcountAppliedCryptographyTwoZeroTwoSixPrimary}{12{,}178}
5217+\newcommand{\CatcountAutonomousScientificDiscoveryTwoZeroOneEightPrimary}{86}
5218+\newcommand{\CatcountAutonomousScientificDiscoveryTwoZeroOneNinePrimary}{114}
5219+\newcommand{\CatcountAutonomousScientificDiscoveryTwoZeroTwoZeroPrimary}{193}
5220+\newcommand{\CatcountAutonomousScientificDiscoveryTwoZeroTwoOnePrimary}{221}
5221+\newcommand{\CatcountAutonomousScientificDiscoveryTwoZeroTwoTwoPrimary}{236}
5222+\newcommand{\CatcountAutonomousScientificDiscoveryTwoZeroTwoThreePrimary}{400}
5223+\newcommand{\CatcountAutonomousScientificDiscoveryTwoZeroTwoFourPrimary}{456}
5224+\newcommand{\CatcountAutonomousScientificDiscoveryTwoZeroTwoFivePrimary}{764}
5225+\newcommand{\CatcountAutonomousScientificDiscoveryTwoZeroTwoSixPrimary}{910}
5226+\newcommand{\CatcountBlockchainTwoZeroOneEightPrimary}{6{,}441}
5227+\newcommand{\CatcountBlockchainTwoZeroOneNinePrimary}{9{,}047}
5228+\newcommand{\CatcountBlockchainTwoZeroTwoZeroPrimary}{9{,}691}
5229+\newcommand{\CatcountBlockchainTwoZeroTwoOnePrimary}{10{,}647}
5230+\newcommand{\CatcountBlockchainTwoZeroTwoTwoPrimary}{11{,}770}
5231+\newcommand{\CatcountBlockchainTwoZeroTwoThreePrimary}{14{,}349}
5232+\newcommand{\CatcountBlockchainTwoZeroTwoFourPrimary}{15{,}376}
5233+\newcommand{\CatcountBlockchainTwoZeroTwoFivePrimary}{18{,}012}
5234+\newcommand{\CatcountBlockchainTwoZeroTwoSixPrimary}{13{,}276}
5235+\newcommand{\CatcountCloudComputingTwoZeroOneEightPrimary}{9{,}396}
5236+\newcommand{\CatcountCloudComputingTwoZeroOneNinePrimary}{8{,}756}
5237+\newcommand{\CatcountCloudComputingTwoZeroTwoZeroPrimary}{8{,}184}
5238+\newcommand{\CatcountCloudComputingTwoZeroTwoOnePrimary}{7{,}735}
5239+\newcommand{\CatcountCloudComputingTwoZeroTwoTwoPrimary}{7{,}103}
5240+\newcommand{\CatcountCloudComputingTwoZeroTwoThreePrimary}{7{,}643}
5241+\newcommand{\CatcountCloudComputingTwoZeroTwoFourPrimary}{8{,}609}
5242+\newcommand{\CatcountCloudComputingTwoZeroTwoFivePrimary}{10{,}974}
5243+\newcommand{\CatcountCloudComputingTwoZeroTwoSixPrimary}{8{,}774}
5244+\newcommand{\CatcountCompilersTwoZeroOneEightPrimary}{725}
5245+\newcommand{\CatcountCompilersTwoZeroOneNinePrimary}{689}
5246+\newcommand{\CatcountCompilersTwoZeroTwoZeroPrimary}{737}
5247+\newcommand{\CatcountCompilersTwoZeroTwoOnePrimary}{709}
5248+\newcommand{\CatcountCompilersTwoZeroTwoTwoPrimary}{642}
5249+\newcommand{\CatcountCompilersTwoZeroTwoThreePrimary}{692}
5250+\newcommand{\CatcountCompilersTwoZeroTwoFourPrimary}{768}
5251+\newcommand{\CatcountCompilersTwoZeroTwoFivePrimary}{767}
5252+\newcommand{\CatcountCompilersTwoZeroTwoSixPrimary}{621}
5253+\newcommand{\CatcountComputationalBiologyTwoZeroOneEightPrimary}{12{,}968}
5254+\newcommand{\CatcountComputationalBiologyTwoZeroOneNinePrimary}{14{,}876}
5255+\newcommand{\CatcountComputationalBiologyTwoZeroTwoZeroPrimary}{18{,}078}
5256+\newcommand{\CatcountComputationalBiologyTwoZeroTwoOnePrimary}{19{,}218}
5257+\newcommand{\CatcountComputationalBiologyTwoZeroTwoTwoPrimary}{19{,}536}
5258+\newcommand{\CatcountComputationalBiologyTwoZeroTwoThreePrimary}{20{,}576}
5259+\newcommand{\CatcountComputationalBiologyTwoZeroTwoFourPrimary}{23{,}522}
5260+\newcommand{\CatcountComputationalBiologyTwoZeroTwoFivePrimary}{28{,}154}
5261+\newcommand{\CatcountComputationalBiologyTwoZeroTwoSixPrimary}{23{,}787}
5262+\newcommand{\CatcountComputationalComplexityTwoZeroOneEightPrimary}{3{,}837}
5263+\newcommand{\CatcountComputationalComplexityTwoZeroOneNinePrimary}{3{,}952}
5264+\newcommand{\CatcountComputationalComplexityTwoZeroTwoZeroPrimary}{3{,}872}
5265+\newcommand{\CatcountComputationalComplexityTwoZeroTwoOnePrimary}{3{,}765}
5266+\newcommand{\CatcountComputationalComplexityTwoZeroTwoTwoPrimary}{3{,}541}
5267+\newcommand{\CatcountComputationalComplexityTwoZeroTwoThreePrimary}{3{,}499}
5268+\newcommand{\CatcountComputationalComplexityTwoZeroTwoFourPrimary}{3{,}824}
5269+\newcommand{\CatcountComputationalComplexityTwoZeroTwoFivePrimary}{4{,}209}
5270+\newcommand{\CatcountComputationalComplexityTwoZeroTwoSixPrimary}{4{,}067}
5271+\newcommand{\CatcountComputerArchitectureTwoZeroOneEightPrimary}{19{,}554}
5272+\newcommand{\CatcountComputerArchitectureTwoZeroOneNinePrimary}{18{,}846}
5273+\newcommand{\CatcountComputerArchitectureTwoZeroTwoZeroPrimary}{17{,}491}
5274+\newcommand{\CatcountComputerArchitectureTwoZeroTwoOnePrimary}{16{,}400}
5275+\newcommand{\CatcountComputerArchitectureTwoZeroTwoTwoPrimary}{15{,}174}
5276+\newcommand{\CatcountComputerArchitectureTwoZeroTwoThreePrimary}{16{,}369}
5277+\newcommand{\CatcountComputerArchitectureTwoZeroTwoFourPrimary}{17{,}159}
5278+\newcommand{\CatcountComputerArchitectureTwoZeroTwoFivePrimary}{20{,}209}
5279+\newcommand{\CatcountComputerArchitectureTwoZeroTwoSixPrimary}{15{,}794}
5280+\newcommand{\CatcountComputerGraphicsTwoZeroOneEightPrimary}{10{,}198}
5281+\newcommand{\CatcountComputerGraphicsTwoZeroOneNinePrimary}{9{,}798}
5282+\newcommand{\CatcountComputerGraphicsTwoZeroTwoZeroPrimary}{9{,}445}
5283+\newcommand{\CatcountComputerGraphicsTwoZeroTwoOnePrimary}{8{,}969}
5284+\newcommand{\CatcountComputerGraphicsTwoZeroTwoTwoPrimary}{8{,}287}
5285+\newcommand{\CatcountComputerGraphicsTwoZeroTwoThreePrimary}{9{,}297}
5286+\newcommand{\CatcountComputerGraphicsTwoZeroTwoFourPrimary}{10{,}611}
5287+\newcommand{\CatcountComputerGraphicsTwoZeroTwoFivePrimary}{12{,}440}
5288+\newcommand{\CatcountComputerGraphicsTwoZeroTwoSixPrimary}{10{,}260}
5289+\newcommand{\CatcountComputerNetworksTwoZeroOneEightPrimary}{58{,}744}
5290+\newcommand{\CatcountComputerNetworksTwoZeroOneNinePrimary}{58{,}299}
5291+\newcommand{\CatcountComputerNetworksTwoZeroTwoZeroPrimary}{54{,}389}
5292+\newcommand{\CatcountComputerNetworksTwoZeroTwoOnePrimary}{51{,}644}
5293+\newcommand{\CatcountComputerNetworksTwoZeroTwoTwoPrimary}{49{,}141}
5294+\newcommand{\CatcountComputerNetworksTwoZeroTwoThreePrimary}{51{,}881}
5295+\newcommand{\CatcountComputerNetworksTwoZeroTwoFourPrimary}{54{,}342}
5296+\newcommand{\CatcountComputerNetworksTwoZeroTwoFivePrimary}{59{,}555}
5297+\newcommand{\CatcountComputerNetworksTwoZeroTwoSixPrimary}{42{,}416}
5298+\newcommand{\CatcountComputerVisionTwoZeroOneEightPrimary}{69{,}738}
5299+\newcommand{\CatcountComputerVisionTwoZeroOneNinePrimary}{81{,}670}
5300+\newcommand{\CatcountComputerVisionTwoZeroTwoZeroPrimary}{84{,}046}
5301+\newcommand{\CatcountComputerVisionTwoZeroTwoOnePrimary}{93{,}096}
5302+\newcommand{\CatcountComputerVisionTwoZeroTwoTwoPrimary}{98{,}704}
5303+\newcommand{\CatcountComputerVisionTwoZeroTwoThreePrimary}{114{,}261}
5304+\newcommand{\CatcountComputerVisionTwoZeroTwoFourPrimary}{133{,}317}
5305+\newcommand{\CatcountComputerVisionTwoZeroTwoFivePrimary}{150{,}757}
5306+\newcommand{\CatcountComputerVisionTwoZeroTwoSixPrimary}{117{,}911}
5307+\newcommand{\CatcountCybersecurityTwoZeroOneEightPrimary}{29{,}440}
5308+\newcommand{\CatcountCybersecurityTwoZeroOneNinePrimary}{31{,}561}
5309+\newcommand{\CatcountCybersecurityTwoZeroTwoZeroPrimary}{32{,}705}
5310+\newcommand{\CatcountCybersecurityTwoZeroTwoOnePrimary}{31{,}779}
5311+\newcommand{\CatcountCybersecurityTwoZeroTwoTwoPrimary}{31{,}554}
5312+\newcommand{\CatcountCybersecurityTwoZeroTwoThreePrimary}{36{,}481}
5313+\newcommand{\CatcountCybersecurityTwoZeroTwoFourPrimary}{45{,}548}
5314+\newcommand{\CatcountCybersecurityTwoZeroTwoFivePrimary}{59{,}044}
5315+\newcommand{\CatcountCybersecurityTwoZeroTwoSixPrimary}{51{,}618}
5316+\newcommand{\CatcountDatabaseSystemsTwoZeroOneEightPrimary}{16{,}265}
5317+\newcommand{\CatcountDatabaseSystemsTwoZeroOneNinePrimary}{16{,}579}
5318+\newcommand{\CatcountDatabaseSystemsTwoZeroTwoZeroPrimary}{15{,}061}
5319+\newcommand{\CatcountDatabaseSystemsTwoZeroTwoOnePrimary}{14{,}332}
5320+\newcommand{\CatcountDatabaseSystemsTwoZeroTwoTwoPrimary}{13{,}625}
5321+\newcommand{\CatcountDatabaseSystemsTwoZeroTwoThreePrimary}{14{,}487}
5322+\newcommand{\CatcountDatabaseSystemsTwoZeroTwoFourPrimary}{15{,}024}
5323+\newcommand{\CatcountDatabaseSystemsTwoZeroTwoFivePrimary}{16{,}681}
5324+\newcommand{\CatcountDatabaseSystemsTwoZeroTwoSixPrimary}{14{,}095}
5325+\newcommand{\CatcountDeepLearningArchitecturesTwoZeroOneEightPrimary}{11{,}992}
5326+\newcommand{\CatcountDeepLearningArchitecturesTwoZeroOneNinePrimary}{15{,}363}
5327+\newcommand{\CatcountDeepLearningArchitecturesTwoZeroTwoZeroPrimary}{17{,}909}
5328+\newcommand{\CatcountDeepLearningArchitecturesTwoZeroTwoOnePrimary}{21{,}224}
5329+\newcommand{\CatcountDeepLearningArchitecturesTwoZeroTwoTwoPrimary}{22{,}270}
5330+\newcommand{\CatcountDeepLearningArchitecturesTwoZeroTwoThreePrimary}{25{,}607}
5331+\newcommand{\CatcountDeepLearningArchitecturesTwoZeroTwoFourPrimary}{28{,}985}
5332+\newcommand{\CatcountDeepLearningArchitecturesTwoZeroTwoFivePrimary}{34{,}141}
5333+\newcommand{\CatcountDeepLearningArchitecturesTwoZeroTwoSixPrimary}{26{,}153}
5334+\newcommand{\CatcountDistributedComputingTwoZeroOneEightPrimary}{11{,}519}
5335+\newcommand{\CatcountDistributedComputingTwoZeroOneNinePrimary}{11{,}495}
5336+\newcommand{\CatcountDistributedComputingTwoZeroTwoZeroPrimary}{10{,}901}
5337+\newcommand{\CatcountDistributedComputingTwoZeroTwoOnePrimary}{10{,}096}
5338+\newcommand{\CatcountDistributedComputingTwoZeroTwoTwoPrimary}{9{,}379}
5339+\newcommand{\CatcountDistributedComputingTwoZeroTwoThreePrimary}{9{,}798}
5340+\newcommand{\CatcountDistributedComputingTwoZeroTwoFourPrimary}{10{,}601}
5341+\newcommand{\CatcountDistributedComputingTwoZeroTwoFivePrimary}{12{,}550}
5342+\newcommand{\CatcountDistributedComputingTwoZeroTwoSixPrimary}{10{,}002}
5343+\newcommand{\CatcountEdgeComputingTwoZeroOneEightPrimary}{5{,}185}
5344+\newcommand{\CatcountEdgeComputingTwoZeroOneNinePrimary}{6{,}650}
5345+\newcommand{\CatcountEdgeComputingTwoZeroTwoZeroPrimary}{7{,}115}
5346+\newcommand{\CatcountEdgeComputingTwoZeroTwoOnePrimary}{7{,}910}
5347+\newcommand{\CatcountEdgeComputingTwoZeroTwoTwoPrimary}{8{,}554}
5348+\newcommand{\CatcountEdgeComputingTwoZeroTwoThreePrimary}{9{,}120}
5349+\newcommand{\CatcountEdgeComputingTwoZeroTwoFourPrimary}{10{,}210}
5350+\newcommand{\CatcountEdgeComputingTwoZeroTwoFivePrimary}{11{,}554}
5351+\newcommand{\CatcountEdgeComputingTwoZeroTwoSixPrimary}{8{,}718}
5352+\newcommand{\CatcountFederatedLearningTwoZeroOneEightPrimary}{143}
5353+\newcommand{\CatcountFederatedLearningTwoZeroOneNinePrimary}{216}
5354+\newcommand{\CatcountFederatedLearningTwoZeroTwoZeroPrimary}{304}
5355+\newcommand{\CatcountFederatedLearningTwoZeroTwoOnePrimary}{651}
5356+\newcommand{\CatcountFederatedLearningTwoZeroTwoTwoPrimary}{974}
5357+\newcommand{\CatcountFederatedLearningTwoZeroTwoThreePrimary}{1{,}497}
5358+\newcommand{\CatcountFederatedLearningTwoZeroTwoFourPrimary}{2{,}070}
5359+\newcommand{\CatcountFederatedLearningTwoZeroTwoFivePrimary}{2{,}544}
5360+\newcommand{\CatcountFederatedLearningTwoZeroTwoSixPrimary}{2{,}841}
5361+\newcommand{\CatcountFormalVerificationTwoZeroOneEightPrimary}{5{,}307}
5362+\newcommand{\CatcountFormalVerificationTwoZeroOneNinePrimary}{4{,}802}
5363+\newcommand{\CatcountFormalVerificationTwoZeroTwoZeroPrimary}{4{,}535}
5364+\newcommand{\CatcountFormalVerificationTwoZeroTwoOnePrimary}{4{,}511}
5365+\newcommand{\CatcountFormalVerificationTwoZeroTwoTwoPrimary}{3{,}995}
5366+\newcommand{\CatcountFormalVerificationTwoZeroTwoThreePrimary}{4{,}385}
5367+\newcommand{\CatcountFormalVerificationTwoZeroTwoFourPrimary}{4{,}233}
5368+\newcommand{\CatcountFormalVerificationTwoZeroTwoFivePrimary}{4{,}623}
5369+\newcommand{\CatcountFormalVerificationTwoZeroTwoSixPrimary}{4{,}866}
5370+\newcommand{\CatcountFoundationModelsTwoZeroOneEightPrimary}{46}
5371+\newcommand{\CatcountFoundationModelsTwoZeroOneNinePrimary}{83}
5372+\newcommand{\CatcountFoundationModelsTwoZeroTwoZeroPrimary}{137}
5373+\newcommand{\CatcountFoundationModelsTwoZeroTwoOnePrimary}{249}
5374+\newcommand{\CatcountFoundationModelsTwoZeroTwoTwoPrimary}{350}
5375+\newcommand{\CatcountFoundationModelsTwoZeroTwoThreePrimary}{571}
5376+\newcommand{\CatcountFoundationModelsTwoZeroTwoFourPrimary}{1{,}046}
5377+\newcommand{\CatcountFoundationModelsTwoZeroTwoFivePrimary}{1{,}854}
5378+\newcommand{\CatcountFoundationModelsTwoZeroTwoSixPrimary}{2{,}688}
5379+\newcommand{\CatcountGenerativeAiTwoZeroOneEightPrimary}{766}
5380+\newcommand{\CatcountGenerativeAiTwoZeroOneNinePrimary}{1{,}266}
5381+\newcommand{\CatcountGenerativeAiTwoZeroTwoZeroPrimary}{1{,}498}
5382+\newcommand{\CatcountGenerativeAiTwoZeroTwoOnePrimary}{1{,}960}
5383+\newcommand{\CatcountGenerativeAiTwoZeroTwoTwoPrimary}{2{,}160}
5384+\newcommand{\CatcountGenerativeAiTwoZeroTwoThreePrimary}{3{,}646}
5385+\newcommand{\CatcountGenerativeAiTwoZeroTwoFourPrimary}{7{,}147}
5386+\newcommand{\CatcountGenerativeAiTwoZeroTwoFivePrimary}{12{,}743}
5387+\newcommand{\CatcountGenerativeAiTwoZeroTwoSixPrimary}{14{,}511}
5388+\newcommand{\CatcountHciTwoZeroOneEightPrimary}{34{,}225}
5389+\newcommand{\CatcountHciTwoZeroOneNinePrimary}{37{,}212}
5390+\newcommand{\CatcountHciTwoZeroTwoZeroPrimary}{36{,}251}
5391+\newcommand{\CatcountHciTwoZeroTwoOnePrimary}{34{,}429}
5392+\newcommand{\CatcountHciTwoZeroTwoTwoPrimary}{34{,}801}
5393+\newcommand{\CatcountHciTwoZeroTwoThreePrimary}{40{,}587}
5394+\newcommand{\CatcountHciTwoZeroTwoFourPrimary}{48{,}074}
5395+\newcommand{\CatcountHciTwoZeroTwoFivePrimary}{58{,}148}
5396+\newcommand{\CatcountHciTwoZeroTwoSixPrimary}{47{,}219}
5397+\newcommand{\CatcountInformationRetrievalTwoZeroOneEightPrimary}{12{,}095}
5398+\newcommand{\CatcountInformationRetrievalTwoZeroOneNinePrimary}{12{,}304}
5399+\newcommand{\CatcountInformationRetrievalTwoZeroTwoZeroPrimary}{11{,}777}
5400+\newcommand{\CatcountInformationRetrievalTwoZeroTwoOnePrimary}{11{,}543}
5401+\newcommand{\CatcountInformationRetrievalTwoZeroTwoTwoPrimary}{11{,}243}
5402+\newcommand{\CatcountInformationRetrievalTwoZeroTwoThreePrimary}{12{,}568}
5403+\newcommand{\CatcountInformationRetrievalTwoZeroTwoFourPrimary}{14{,}146}
5404+\newcommand{\CatcountInformationRetrievalTwoZeroTwoFivePrimary}{17{,}379}
5405+\newcommand{\CatcountInformationRetrievalTwoZeroTwoSixPrimary}{16{,}901}
5406+\newcommand{\CatcountLlmTwoZeroOneEightPrimary}{1{,}635}
5407+\newcommand{\CatcountLlmTwoZeroOneNinePrimary}{1{,}857}
5408+\newcommand{\CatcountLlmTwoZeroTwoZeroPrimary}{2{,}461}
5409+\newcommand{\CatcountLlmTwoZeroTwoOnePrimary}{2{,}849}
5410+\newcommand{\CatcountLlmTwoZeroTwoTwoPrimary}{3{,}266}
5411+\newcommand{\CatcountLlmTwoZeroTwoThreePrimary}{10{,}389}
5412+\newcommand{\CatcountLlmTwoZeroTwoFourPrimary}{24{,}873}
5413+\newcommand{\CatcountLlmTwoZeroTwoFivePrimary}{45{,}158}
5414+\newcommand{\CatcountLlmTwoZeroTwoSixPrimary}{51{,}454}
5415+\newcommand{\CatcountMlTheoryTwoZeroOneEightPrimary}{13{,}534}
5416+\newcommand{\CatcountMlTheoryTwoZeroOneNinePrimary}{17{,}056}
5417+\newcommand{\CatcountMlTheoryTwoZeroTwoZeroPrimary}{19{,}309}
5418+\newcommand{\CatcountMlTheoryTwoZeroTwoOnePrimary}{19{,}956}
5419+\newcommand{\CatcountMlTheoryTwoZeroTwoTwoPrimary}{20{,}131}
5420+\newcommand{\CatcountMlTheoryTwoZeroTwoThreePrimary}{22{,}893}
5421+\newcommand{\CatcountMlTheoryTwoZeroTwoFourPrimary}{26{,}710}
5422+\newcommand{\CatcountMlTheoryTwoZeroTwoFivePrimary}{32{,}798}
5423+\newcommand{\CatcountMlTheoryTwoZeroTwoSixPrimary}{29{,}457}
5424+\newcommand{\CatcountMultimodalAiTwoZeroOneEightPrimary}{470}
5425+\newcommand{\CatcountMultimodalAiTwoZeroOneNinePrimary}{632}
5426+\newcommand{\CatcountMultimodalAiTwoZeroTwoZeroPrimary}{755}
5427+\newcommand{\CatcountMultimodalAiTwoZeroTwoOnePrimary}{948}
5428+\newcommand{\CatcountMultimodalAiTwoZeroTwoTwoPrimary}{1{,}278}
5429+\newcommand{\CatcountMultimodalAiTwoZeroTwoThreePrimary}{1{,}932}
5430+\newcommand{\CatcountMultimodalAiTwoZeroTwoFourPrimary}{3{,}408}
5431+\newcommand{\CatcountMultimodalAiTwoZeroTwoFivePrimary}{6{,}749}
5432+\newcommand{\CatcountMultimodalAiTwoZeroTwoSixPrimary}{8{,}806}
5433+\newcommand{\CatcountNlpTwoZeroOneEightPrimary}{23{,}372}
5434+\newcommand{\CatcountNlpTwoZeroOneNinePrimary}{25{,}738}
5435+\newcommand{\CatcountNlpTwoZeroTwoZeroPrimary}{28{,}747}
5436+\newcommand{\CatcountNlpTwoZeroTwoOnePrimary}{29{,}593}
5437+\newcommand{\CatcountNlpTwoZeroTwoTwoPrimary}{29{,}678}
5438+\newcommand{\CatcountNlpTwoZeroTwoThreePrimary}{33{,}965}
5439+\newcommand{\CatcountNlpTwoZeroTwoFourPrimary}{38{,}831}
5440+\newcommand{\CatcountNlpTwoZeroTwoFivePrimary}{42{,}914}
5441+\newcommand{\CatcountNlpTwoZeroTwoSixPrimary}{34{,}785}
5442+\newcommand{\CatcountOperatingSystemsTwoZeroOneEightPrimary}{1{,}556}
5443+\newcommand{\CatcountOperatingSystemsTwoZeroOneNinePrimary}{1{,}579}
5444+\newcommand{\CatcountOperatingSystemsTwoZeroTwoZeroPrimary}{1{,}454}
5445+\newcommand{\CatcountOperatingSystemsTwoZeroTwoOnePrimary}{1{,}323}
5446+\newcommand{\CatcountOperatingSystemsTwoZeroTwoTwoPrimary}{1{,}083}
5447+\newcommand{\CatcountOperatingSystemsTwoZeroTwoThreePrimary}{1{,}173}
5448+\newcommand{\CatcountOperatingSystemsTwoZeroTwoFourPrimary}{1{,}260}
5449+\newcommand{\CatcountOperatingSystemsTwoZeroTwoFivePrimary}{1{,}363}
5450+\newcommand{\CatcountOperatingSystemsTwoZeroTwoSixPrimary}{1{,}247}
5451+\newcommand{\CatcountPrivacyPreservingComputationTwoZeroOneEightPrimary}{3{,}105}
5452+\newcommand{\CatcountPrivacyPreservingComputationTwoZeroOneNinePrimary}{3{,}256}
5453+\newcommand{\CatcountPrivacyPreservingComputationTwoZeroTwoZeroPrimary}{4{,}157}
5454+\newcommand{\CatcountPrivacyPreservingComputationTwoZeroTwoOnePrimary}{4{,}241}
5455+\newcommand{\CatcountPrivacyPreservingComputationTwoZeroTwoTwoPrimary}{4{,}376}
5456+\newcommand{\CatcountPrivacyPreservingComputationTwoZeroTwoThreePrimary}{5{,}010}
5457+\newcommand{\CatcountPrivacyPreservingComputationTwoZeroTwoFourPrimary}{6{,}686}
5458+\newcommand{\CatcountPrivacyPreservingComputationTwoZeroTwoFivePrimary}{8{,}817}
5459+\newcommand{\CatcountPrivacyPreservingComputationTwoZeroTwoSixPrimary}{7{,}625}
5460+\newcommand{\CatcountProgrammingLanguagesTwoZeroOneEightPrimary}{4{,}390}
5461+\newcommand{\CatcountProgrammingLanguagesTwoZeroOneNinePrimary}{4{,}474}
5462+\newcommand{\CatcountProgrammingLanguagesTwoZeroTwoZeroPrimary}{4{,}126}
5463+\newcommand{\CatcountProgrammingLanguagesTwoZeroTwoOnePrimary}{3{,}978}
5464+\newcommand{\CatcountProgrammingLanguagesTwoZeroTwoTwoPrimary}{3{,}664}
5465+\newcommand{\CatcountProgrammingLanguagesTwoZeroTwoThreePrimary}{3{,}876}
5466+\newcommand{\CatcountProgrammingLanguagesTwoZeroTwoFourPrimary}{3{,}846}
5467+\newcommand{\CatcountProgrammingLanguagesTwoZeroTwoFivePrimary}{4{,}193}
5468+\newcommand{\CatcountProgrammingLanguagesTwoZeroTwoSixPrimary}{3{,}691}
5469+\newcommand{\CatcountQuantumComputingTwoZeroOneEightPrimary}{6{,}224}
5470+\newcommand{\CatcountQuantumComputingTwoZeroOneNinePrimary}{7{,}482}
5471+\newcommand{\CatcountQuantumComputingTwoZeroTwoZeroPrimary}{8{,}099}
5472+\newcommand{\CatcountQuantumComputingTwoZeroTwoOnePrimary}{9{,}280}
5473+\newcommand{\CatcountQuantumComputingTwoZeroTwoTwoPrimary}{9{,}216}
5474+\newcommand{\CatcountQuantumComputingTwoZeroTwoThreePrimary}{10{,}544}
5475+\newcommand{\CatcountQuantumComputingTwoZeroTwoFourPrimary}{12{,}722}
5476+\newcommand{\CatcountQuantumComputingTwoZeroTwoFivePrimary}{17{,}471}
5477+\newcommand{\CatcountQuantumComputingTwoZeroTwoSixPrimary}{19{,}272}
5478+\newcommand{\CatcountReinforcementLearningTwoZeroOneEightPrimary}{3{,}296}
5479+\newcommand{\CatcountReinforcementLearningTwoZeroOneNinePrimary}{4{,}741}
5480+\newcommand{\CatcountReinforcementLearningTwoZeroTwoZeroPrimary}{5{,}675}
5481+\newcommand{\CatcountReinforcementLearningTwoZeroTwoOnePrimary}{6{,}592}
5482+\newcommand{\CatcountReinforcementLearningTwoZeroTwoTwoPrimary}{7{,}017}
5483+\newcommand{\CatcountReinforcementLearningTwoZeroTwoThreePrimary}{8{,}422}
5484+\newcommand{\CatcountReinforcementLearningTwoZeroTwoFourPrimary}{9{,}892}
5485+\newcommand{\CatcountReinforcementLearningTwoZeroTwoFivePrimary}{13{,}210}
5486+\newcommand{\CatcountReinforcementLearningTwoZeroTwoSixPrimary}{12{,}555}
5487+\newcommand{\CatcountRoboticsTwoZeroOneEightPrimary}{17{,}496}
5488+\newcommand{\CatcountRoboticsTwoZeroOneNinePrimary}{20{,}055}
5489+\newcommand{\CatcountRoboticsTwoZeroTwoZeroPrimary}{20{,}353}
5490+\newcommand{\CatcountRoboticsTwoZeroTwoOnePrimary}{22{,}066}
5491+\newcommand{\CatcountRoboticsTwoZeroTwoTwoPrimary}{23{,}103}
5492+\newcommand{\CatcountRoboticsTwoZeroTwoThreePrimary}{25{,}759}
5493+\newcommand{\CatcountRoboticsTwoZeroTwoFourPrimary}{29{,}986}
5494+\newcommand{\CatcountRoboticsTwoZeroTwoFivePrimary}{33{,}464}
5495+\newcommand{\CatcountRoboticsTwoZeroTwoSixPrimary}{27{,}674}
5496+\newcommand{\CatcountScientificComputingTwoZeroOneEightPrimary}{16{,}882}
5497+\newcommand{\CatcountScientificComputingTwoZeroOneNinePrimary}{18{,}324}
5498+\newcommand{\CatcountScientificComputingTwoZeroTwoZeroPrimary}{19{,}740}
5499+\newcommand{\CatcountScientificComputingTwoZeroTwoOnePrimary}{20{,}299}
5500+\newcommand{\CatcountScientificComputingTwoZeroTwoTwoPrimary}{19{,}932}
5501+\newcommand{\CatcountScientificComputingTwoZeroTwoThreePrimary}{20{,}423}
5502+\newcommand{\CatcountScientificComputingTwoZeroTwoFourPrimary}{21{,}887}
5503+\newcommand{\CatcountScientificComputingTwoZeroTwoFivePrimary}{23{,}191}
5504+\newcommand{\CatcountScientificComputingTwoZeroTwoSixPrimary}{18{,}738}
5505+\newcommand{\CatcountSoftwareEngineeringTwoZeroOneEightPrimary}{29{,}016}
5506+\newcommand{\CatcountSoftwareEngineeringTwoZeroOneNinePrimary}{28{,}935}
5507+\newcommand{\CatcountSoftwareEngineeringTwoZeroTwoZeroPrimary}{27{,}756}
5508+\newcommand{\CatcountSoftwareEngineeringTwoZeroTwoOnePrimary}{24{,}671}
5509+\newcommand{\CatcountSoftwareEngineeringTwoZeroTwoTwoPrimary}{23{,}725}
5510+\newcommand{\CatcountSoftwareEngineeringTwoZeroTwoThreePrimary}{26{,}245}
5511+\newcommand{\CatcountSoftwareEngineeringTwoZeroTwoFourPrimary}{30{,}533}
5512+\newcommand{\CatcountSoftwareEngineeringTwoZeroTwoFivePrimary}{36{,}381}
5513+\newcommand{\CatcountSoftwareEngineeringTwoZeroTwoSixPrimary}{34{,}261}
5514+\newcommand{\CatcountSpeechAudioTwoZeroOneEightPrimary}{12{,}096}
5515+\newcommand{\CatcountSpeechAudioTwoZeroOneNinePrimary}{12{,}783}
5516+\newcommand{\CatcountSpeechAudioTwoZeroTwoZeroPrimary}{12{,}493}
5517+\newcommand{\CatcountSpeechAudioTwoZeroTwoOnePrimary}{12{,}787}
5518+\newcommand{\CatcountSpeechAudioTwoZeroTwoTwoPrimary}{12{,}698}
5519+\newcommand{\CatcountSpeechAudioTwoZeroTwoThreePrimary}{14{,}688}
5520+\newcommand{\CatcountSpeechAudioTwoZeroTwoFourPrimary}{16{,}427}
5521+\newcommand{\CatcountSpeechAudioTwoZeroTwoFivePrimary}{18{,}373}
5522+\newcommand{\CatcountSpeechAudioTwoZeroTwoSixPrimary}{15{,}624}
5523+\newcommand{\CatcountSustainableComputingTwoZeroOneEightPrimary}{583}
5524+\newcommand{\CatcountSustainableComputingTwoZeroOneNinePrimary}{667}
5525+\newcommand{\CatcountSustainableComputingTwoZeroTwoZeroPrimary}{634}
5526+\newcommand{\CatcountSustainableComputingTwoZeroTwoOnePrimary}{717}
5527+\newcommand{\CatcountSustainableComputingTwoZeroTwoTwoPrimary}{830}
5528+\newcommand{\CatcountSustainableComputingTwoZeroTwoThreePrimary}{1{,}078}
5529+\newcommand{\CatcountSustainableComputingTwoZeroTwoFourPrimary}{1{,}271}
5530+\newcommand{\CatcountSustainableComputingTwoZeroTwoFivePrimary}{1{,}639}
5531+\newcommand{\CatcountSustainableComputingTwoZeroTwoSixPrimary}{1{,}328}
5532+\newcommand{\CatcountAiAcceleratorsBaselinePrimary}{5{,}565}
5533+\newcommand{\CatcountAiAcceleratorsPostPrimary}{6{,}874}
5534+\newcommand{\CatcountAiAgentsBaselinePrimary}{25{,}484}
5535+\newcommand{\CatcountAiAgentsPostPrimary}{31{,}918}
5536+\newcommand{\CatcountAiSafetyBaselinePrimary}{10{,}133}
5537+\newcommand{\CatcountAiSafetyPostPrimary}{33{,}661}
5538+\newcommand{\CatcountAlgorithmsBaselinePrimary}{145{,}001}
5539+\newcommand{\CatcountAlgorithmsPostPrimary}{121{,}059}
5540+\newcommand{\CatcountAppliedCryptographyBaselinePrimary}{54{,}639}
5541+\newcommand{\CatcountAppliedCryptographyPostPrimary}{50{,}435}
5542+\newcommand{\CatcountAutonomousScientificDiscoveryBaselinePrimary}{850}
5543+\newcommand{\CatcountAutonomousScientificDiscoveryPostPrimary}{2{,}530}
5544+\newcommand{\CatcountBlockchainBaselinePrimary}{47{,}596}
5545+\newcommand{\CatcountBlockchainPostPrimary}{61{,}013}
5546+\newcommand{\CatcountCloudComputingBaselinePrimary}{41{,}174}
5547+\newcommand{\CatcountCloudComputingPostPrimary}{36{,}000}
5548+\newcommand{\CatcountCompilersBaselinePrimary}{3{,}502}
5549+\newcommand{\CatcountCompilersPostPrimary}{2{,}848}
5550+\newcommand{\CatcountComputationalBiologyBaselinePrimary}{84{,}676}
5551+\newcommand{\CatcountComputationalBiologyPostPrimary}{96{,}039}
5552+\newcommand{\CatcountComputationalComplexityBaselinePrimary}{18{,}967}
5553+\newcommand{\CatcountComputationalComplexityPostPrimary}{15{,}599}
5554+\newcommand{\CatcountComputerArchitectureBaselinePrimary}{87{,}465}
5555+\newcommand{\CatcountComputerArchitecturePostPrimary}{69{,}531}
5556+\newcommand{\CatcountComputerGraphicsBaselinePrimary}{46{,}697}
5557+\newcommand{\CatcountComputerGraphicsPostPrimary}{42{,}608}
5558+\newcommand{\CatcountComputerNetworksBaselinePrimary}{272{,}217}
5559+\newcommand{\CatcountComputerNetworksPostPrimary}{208{,}194}
5560+\newcommand{\CatcountComputerVisionBaselinePrimary}{427{,}254}
5561+\newcommand{\CatcountComputerVisionPostPrimary}{516{,}246}
5562+\newcommand{\CatcountCybersecurityBaselinePrimary}{157{,}039}
5563+\newcommand{\CatcountCybersecurityPostPrimary}{192{,}691}
5564+\newcommand{\CatcountDatabaseSystemsBaselinePrimary}{75{,}862}
5565+\newcommand{\CatcountDatabaseSystemsPostPrimary}{60{,}287}
5566+\newcommand{\CatcountDeepLearningArchitecturesBaselinePrimary}{88{,}758}
5567+\newcommand{\CatcountDeepLearningArchitecturesPostPrimary}{114{,}886}
5568+\newcommand{\CatcountDistributedComputingBaselinePrimary}{53{,}390}
5569+\newcommand{\CatcountDistributedComputingPostPrimary}{42{,}951}
5570+\newcommand{\CatcountEdgeComputingBaselinePrimary}{35{,}414}
5571+\newcommand{\CatcountEdgeComputingPostPrimary}{39{,}602}
5572+\newcommand{\CatcountFederatedLearningBaselinePrimary}{2{,}288}
5573+\newcommand{\CatcountFederatedLearningPostPrimary}{8{,}952}
5574+\newcommand{\CatcountFormalVerificationBaselinePrimary}{23{,}150}
5575+\newcommand{\CatcountFormalVerificationPostPrimary}{18{,}107}
5576+\newcommand{\CatcountFoundationModelsBaselinePrimary}{865}
5577+\newcommand{\CatcountFoundationModelsPostPrimary}{6{,}159}
5578+\newcommand{\CatcountGenerativeAiBaselinePrimary}{7{,}650}
5579+\newcommand{\CatcountGenerativeAiPostPrimary}{38{,}047}
5580+\newcommand{\CatcountHciBaselinePrimary}{176{,}918}
5581+\newcommand{\CatcountHciPostPrimary}{194{,}028}
5582+\newcommand{\CatcountInformationRetrievalBaselinePrimary}{58{,}962}
5583+\newcommand{\CatcountInformationRetrievalPostPrimary}{60{,}994}
5584+\newcommand{\CatcountLlmBaselinePrimary}{12{,}068}
5585+\newcommand{\CatcountLlmPostPrimary}{131{,}874}
5586+\newcommand{\CatcountMlTheoryBaselinePrimary}{89{,}986}
5587+\newcommand{\CatcountMlTheoryPostPrimary}{111{,}858}
5588+\newcommand{\CatcountMultimodalAiBaselinePrimary}{4{,}083}
5589+\newcommand{\CatcountMultimodalAiPostPrimary}{20{,}895}
5590+\newcommand{\CatcountNlpBaselinePrimary}{137{,}128}
5591+\newcommand{\CatcountNlpPostPrimary}{150{,}495}
5592+\newcommand{\CatcountOperatingSystemsBaselinePrimary}{6{,}995}
5593+\newcommand{\CatcountOperatingSystemsPostPrimary}{5{,}043}
5594+\newcommand{\CatcountPrivacyPreservingComputationBaselinePrimary}{19{,}135}
5595+\newcommand{\CatcountPrivacyPreservingComputationPostPrimary}{28{,}138}
5596+\newcommand{\CatcountProgrammingLanguagesBaselinePrimary}{20{,}632}
5597+\newcommand{\CatcountProgrammingLanguagesPostPrimary}{15{,}606}
5598+\newcommand{\CatcountQuantumComputingBaselinePrimary}{40{,}301}
5599+\newcommand{\CatcountQuantumComputingPostPrimary}{60{,}009}
5600+\newcommand{\CatcountReinforcementLearningBaselinePrimary}{27{,}321}
5601+\newcommand{\CatcountReinforcementLearningPostPrimary}{44{,}079}
5602+\newcommand{\CatcountRoboticsBaselinePrimary}{103{,}073}
5603+\newcommand{\CatcountRoboticsPostPrimary}{116{,}883}
5604+\newcommand{\CatcountScientificComputingBaselinePrimary}{95{,}177}
5605+\newcommand{\CatcountScientificComputingPostPrimary}{84{,}239}
5606+\newcommand{\CatcountSoftwareEngineeringBaselinePrimary}{134{,}103}
5607+\newcommand{\CatcountSoftwareEngineeringPostPrimary}{127{,}420}
5608+\newcommand{\CatcountSpeechAudioBaselinePrimary}{62{,}857}
5609+\newcommand{\CatcountSpeechAudioPostPrimary}{65{,}112}
5610+\newcommand{\CatcountSustainableComputingBaselinePrimary}{3{,}431}
5611+\newcommand{\CatcountSustainableComputingPostPrimary}{5{,}316}
5612+% ---- Part VII-b security macros (Agent H) ----
5613+\newcommand{\SecCountAppliedCryptographyTwoZeroOneEightMultilabel}{29{,}699}
5614+\newcommand{\SecCountAppliedCryptographyTwoZeroOneEightPrimary}{10{,}782}
5615+\newcommand{\SecSharePrimaryAppliedCryptographyTwoZeroOneEightPct}{2.16}
5616+\newcommand{\SecCountAppliedCryptographyTwoZeroOneNineMultilabel}{31{,}450}
5617+\newcommand{\SecCountAppliedCryptographyTwoZeroOneNinePrimary}{11{,}355}
5618+\newcommand{\SecSharePrimaryAppliedCryptographyTwoZeroOneNinePct}{2.10}
5619+\newcommand{\SecCountAppliedCryptographyTwoZeroTwoZeroMultilabel}{31{,}778}
5620+\newcommand{\SecCountAppliedCryptographyTwoZeroTwoZeroPrimary}{11{,}279}
5621+\newcommand{\SecSharePrimaryAppliedCryptographyTwoZeroTwoZeroPct}{2.05}
5622+\newcommand{\SecCountAppliedCryptographyTwoZeroTwoOneMultilabel}{32{,}097}
5623+\newcommand{\SecCountAppliedCryptographyTwoZeroTwoOnePrimary}{10{,}924}
5624+\newcommand{\SecSharePrimaryAppliedCryptographyTwoZeroTwoOnePct}{1.95}
5625+\newcommand{\SecCountAppliedCryptographyTwoZeroTwoTwoMultilabel}{32{,}075}
5626+\newcommand{\SecCountAppliedCryptographyTwoZeroTwoTwoPrimary}{10{,}299}
5627+\newcommand{\SecSharePrimaryAppliedCryptographyTwoZeroTwoTwoPct}{1.84}
5628+\newcommand{\SecCountAppliedCryptographyTwoZeroTwoThreeMultilabel}{35{,}724}
5629+\newcommand{\SecCountAppliedCryptographyTwoZeroTwoThreePrimary}{11{,}219}
5630+\newcommand{\SecSharePrimaryAppliedCryptographyTwoZeroTwoThreePct}{1.77}
5631+\newcommand{\SecCountAppliedCryptographyTwoZeroTwoFourMultilabel}{41{,}181}
5632+\newcommand{\SecCountAppliedCryptographyTwoZeroTwoFourPrimary}{12{,}426}
5633+\newcommand{\SecSharePrimaryAppliedCryptographyTwoZeroTwoFourPct}{1.68}
5634+\newcommand{\SecCountAppliedCryptographyTwoZeroTwoFiveMultilabel}{51{,}759}
5635+\newcommand{\SecCountAppliedCryptographyTwoZeroTwoFivePrimary}{14{,}612}
5636+\newcommand{\SecSharePrimaryAppliedCryptographyTwoZeroTwoFivePct}{1.64}
5637+\newcommand{\SecCountAppliedCryptographyTwoZeroTwoSixMultilabel}{43{,}945}
5638+\newcommand{\SecCountAppliedCryptographyTwoZeroTwoSixPrimary}{12{,}178}
5639+\newcommand{\SecSharePrimaryAppliedCryptographyTwoZeroTwoSixPct}{1.58}
5640+\newcommand{\SecCountBlockchainTwoZeroOneEightMultilabel}{10{,}872}
5641+\newcommand{\SecCountBlockchainTwoZeroOneEightPrimary}{6{,}441}
5642+\newcommand{\SecSharePrimaryBlockchainTwoZeroOneEightPct}{1.29}
5643+\newcommand{\SecCountBlockchainTwoZeroOneNineMultilabel}{15{,}314}
5644+\newcommand{\SecCountBlockchainTwoZeroOneNinePrimary}{9{,}047}
5645+\newcommand{\SecSharePrimaryBlockchainTwoZeroOneNinePct}{1.68}
5646+\newcommand{\SecCountBlockchainTwoZeroTwoZeroMultilabel}{16{,}707}
5647+\newcommand{\SecCountBlockchainTwoZeroTwoZeroPrimary}{9{,}691}
5648+\newcommand{\SecSharePrimaryBlockchainTwoZeroTwoZeroPct}{1.76}
5649+\newcommand{\SecCountBlockchainTwoZeroTwoOneMultilabel}{18{,}128}
5650+\newcommand{\SecCountBlockchainTwoZeroTwoOnePrimary}{10{,}647}
5651+\newcommand{\SecSharePrimaryBlockchainTwoZeroTwoOnePct}{1.90}
5652+\newcommand{\SecCountBlockchainTwoZeroTwoTwoMultilabel}{20{,}159}
5653+\newcommand{\SecCountBlockchainTwoZeroTwoTwoPrimary}{11{,}770}
5654+\newcommand{\SecSharePrimaryBlockchainTwoZeroTwoTwoPct}{2.10}
5655+\newcommand{\SecCountBlockchainTwoZeroTwoThreeMultilabel}{24{,}314}
5656+\newcommand{\SecCountBlockchainTwoZeroTwoThreePrimary}{14{,}349}
5657+\newcommand{\SecSharePrimaryBlockchainTwoZeroTwoThreePct}{2.26}
5658+\newcommand{\SecCountBlockchainTwoZeroTwoFourMultilabel}{26{,}919}
5659+\newcommand{\SecCountBlockchainTwoZeroTwoFourPrimary}{15{,}376}
5660+\newcommand{\SecSharePrimaryBlockchainTwoZeroTwoFourPct}{2.08}
5661+\newcommand{\SecCountBlockchainTwoZeroTwoFiveMultilabel}{33{,}426}
5662+\newcommand{\SecCountBlockchainTwoZeroTwoFivePrimary}{18{,}012}
5663+\newcommand{\SecSharePrimaryBlockchainTwoZeroTwoFivePct}{2.02}
5664+\newcommand{\SecCountBlockchainTwoZeroTwoSixMultilabel}{24{,}814}
5665+\newcommand{\SecCountBlockchainTwoZeroTwoSixPrimary}{13{,}276}
5666+\newcommand{\SecSharePrimaryBlockchainTwoZeroTwoSixPct}{1.72}
5667+\newcommand{\SecCountCybersecurityTwoZeroOneEightMultilabel}{67{,}575}
5668+\newcommand{\SecCountCybersecurityTwoZeroOneEightPrimary}{29{,}440}
5669+\newcommand{\SecSharePrimaryCybersecurityTwoZeroOneEightPct}{5.89}
5670+\newcommand{\SecCountCybersecurityTwoZeroOneNineMultilabel}{73{,}397}
5671+\newcommand{\SecCountCybersecurityTwoZeroOneNinePrimary}{31{,}561}
5672+\newcommand{\SecSharePrimaryCybersecurityTwoZeroOneNinePct}{5.85}
5673+\newcommand{\SecCountCybersecurityTwoZeroTwoZeroMultilabel}{76{,}566}
5674+\newcommand{\SecCountCybersecurityTwoZeroTwoZeroPrimary}{32{,}705}
5675+\newcommand{\SecSharePrimaryCybersecurityTwoZeroTwoZeroPct}{5.96}
5676+\newcommand{\SecCountCybersecurityTwoZeroTwoOneMultilabel}{77{,}276}
5677+\newcommand{\SecCountCybersecurityTwoZeroTwoOnePrimary}{31{,}779}
5678+\newcommand{\SecSharePrimaryCybersecurityTwoZeroTwoOnePct}{5.69}
5679+\newcommand{\SecCountCybersecurityTwoZeroTwoTwoMultilabel}{78{,}082}
5680+\newcommand{\SecCountCybersecurityTwoZeroTwoTwoPrimary}{31{,}554}
5681+\newcommand{\SecSharePrimaryCybersecurityTwoZeroTwoTwoPct}{5.63}
5682+\newcommand{\SecCountCybersecurityTwoZeroTwoThreeMultilabel}{89{,}483}
5683+\newcommand{\SecCountCybersecurityTwoZeroTwoThreePrimary}{36{,}481}
5684+\newcommand{\SecSharePrimaryCybersecurityTwoZeroTwoThreePct}{5.75}
5685+\newcommand{\SecCountCybersecurityTwoZeroTwoFourMultilabel}{107{,}542}
5686+\newcommand{\SecCountCybersecurityTwoZeroTwoFourPrimary}{45{,}548}
5687+\newcommand{\SecSharePrimaryCybersecurityTwoZeroTwoFourPct}{6.16}
5688+\newcommand{\SecCountCybersecurityTwoZeroTwoFiveMultilabel}{139{,}134}
5689+\newcommand{\SecCountCybersecurityTwoZeroTwoFivePrimary}{59{,}044}
5690+\newcommand{\SecSharePrimaryCybersecurityTwoZeroTwoFivePct}{6.61}
5691+\newcommand{\SecCountCybersecurityTwoZeroTwoSixMultilabel}{120{,}483}
5692+\newcommand{\SecCountCybersecurityTwoZeroTwoSixPrimary}{51{,}618}
5693+\newcommand{\SecSharePrimaryCybersecurityTwoZeroTwoSixPct}{6.69}
5694+\newcommand{\SecCountFederatedLearningTwoZeroOneEightMultilabel}{1{,}040}
5695+\newcommand{\SecCountFederatedLearningTwoZeroOneEightPrimary}{143}
5696+\newcommand{\SecSharePrimaryFederatedLearningTwoZeroOneEightPct}{0.03}
5697+\newcommand{\SecCountFederatedLearningTwoZeroOneNineMultilabel}{1{,}578}
5698+\newcommand{\SecCountFederatedLearningTwoZeroOneNinePrimary}{216}
5699+\newcommand{\SecSharePrimaryFederatedLearningTwoZeroOneNinePct}{0.04}
5700+\newcommand{\SecCountFederatedLearningTwoZeroTwoZeroMultilabel}{2{,}684}
5701+\newcommand{\SecCountFederatedLearningTwoZeroTwoZeroPrimary}{304}
5702+\newcommand{\SecSharePrimaryFederatedLearningTwoZeroTwoZeroPct}{0.06}
5703+\newcommand{\SecCountFederatedLearningTwoZeroTwoOneMultilabel}{4{,}246}
5704+\newcommand{\SecCountFederatedLearningTwoZeroTwoOnePrimary}{651}
5705+\newcommand{\SecSharePrimaryFederatedLearningTwoZeroTwoOnePct}{0.12}
5706+\newcommand{\SecCountFederatedLearningTwoZeroTwoTwoMultilabel}{5{,}888}
5707+\newcommand{\SecCountFederatedLearningTwoZeroTwoTwoPrimary}{974}
5708+\newcommand{\SecSharePrimaryFederatedLearningTwoZeroTwoTwoPct}{0.17}
5709+\newcommand{\SecCountFederatedLearningTwoZeroTwoThreeMultilabel}{8{,}515}
5710+\newcommand{\SecCountFederatedLearningTwoZeroTwoThreePrimary}{1{,}497}
5711+\newcommand{\SecSharePrimaryFederatedLearningTwoZeroTwoThreePct}{0.24}
5712+\newcommand{\SecCountFederatedLearningTwoZeroTwoFourMultilabel}{11{,}923}
5713+\newcommand{\SecCountFederatedLearningTwoZeroTwoFourPrimary}{2{,}070}
5714+\newcommand{\SecSharePrimaryFederatedLearningTwoZeroTwoFourPct}{0.28}
5715+\newcommand{\SecCountFederatedLearningTwoZeroTwoFiveMultilabel}{18{,}093}
5716+\newcommand{\SecCountFederatedLearningTwoZeroTwoFivePrimary}{2{,}544}
5717+\newcommand{\SecSharePrimaryFederatedLearningTwoZeroTwoFivePct}{0.28}
5718+\newcommand{\SecCountFederatedLearningTwoZeroTwoSixMultilabel}{17{,}070}
5719+\newcommand{\SecCountFederatedLearningTwoZeroTwoSixPrimary}{2{,}841}
5720+\newcommand{\SecSharePrimaryFederatedLearningTwoZeroTwoSixPct}{0.37}
5721+\newcommand{\SecCountPrivacyPreservingComputationTwoZeroOneEightMultilabel}{11{,}378}
5722+\newcommand{\SecCountPrivacyPreservingComputationTwoZeroOneEightPrimary}{3{,}105}
5723+\newcommand{\SecSharePrimaryPrivacyPreservingComputationTwoZeroOneEightPct}{0.62}
5724+\newcommand{\SecCountPrivacyPreservingComputationTwoZeroOneNineMultilabel}{13{,}062}
5725+\newcommand{\SecCountPrivacyPreservingComputationTwoZeroOneNinePrimary}{3{,}256}
5726+\newcommand{\SecSharePrimaryPrivacyPreservingComputationTwoZeroOneNinePct}{0.60}
5727+\newcommand{\SecCountPrivacyPreservingComputationTwoZeroTwoZeroMultilabel}{15{,}779}
5728+\newcommand{\SecCountPrivacyPreservingComputationTwoZeroTwoZeroPrimary}{4{,}157}
5729+\newcommand{\SecSharePrimaryPrivacyPreservingComputationTwoZeroTwoZeroPct}{0.76}
5730+\newcommand{\SecCountPrivacyPreservingComputationTwoZeroTwoOneMultilabel}{17{,}795}
5731+\newcommand{\SecCountPrivacyPreservingComputationTwoZeroTwoOnePrimary}{4{,}241}
5732+\newcommand{\SecSharePrimaryPrivacyPreservingComputationTwoZeroTwoOnePct}{0.76}
5733+\newcommand{\SecCountPrivacyPreservingComputationTwoZeroTwoTwoMultilabel}{20{,}053}
5734+\newcommand{\SecCountPrivacyPreservingComputationTwoZeroTwoTwoPrimary}{4{,}376}
5735+\newcommand{\SecSharePrimaryPrivacyPreservingComputationTwoZeroTwoTwoPct}{0.78}
5736+\newcommand{\SecCountPrivacyPreservingComputationTwoZeroTwoThreeMultilabel}{25{,}211}
5737+\newcommand{\SecCountPrivacyPreservingComputationTwoZeroTwoThreePrimary}{5{,}010}
5738+\newcommand{\SecSharePrimaryPrivacyPreservingComputationTwoZeroTwoThreePct}{0.79}
5739+\newcommand{\SecCountPrivacyPreservingComputationTwoZeroTwoFourMultilabel}{34{,}579}
5740+\newcommand{\SecCountPrivacyPreservingComputationTwoZeroTwoFourPrimary}{6{,}686}
5741+\newcommand{\SecSharePrimaryPrivacyPreservingComputationTwoZeroTwoFourPct}{0.90}
5742+\newcommand{\SecCountPrivacyPreservingComputationTwoZeroTwoFiveMultilabel}{49{,}266}
5743+\newcommand{\SecCountPrivacyPreservingComputationTwoZeroTwoFivePrimary}{8{,}817}
5744+\newcommand{\SecSharePrimaryPrivacyPreservingComputationTwoZeroTwoFivePct}{0.99}
5745+\newcommand{\SecCountPrivacyPreservingComputationTwoZeroTwoSixMultilabel}{41{,}713}
5746+\newcommand{\SecCountPrivacyPreservingComputationTwoZeroTwoSixPrimary}{7{,}625}
5747+\newcommand{\SecSharePrimaryPrivacyPreservingComputationTwoZeroTwoSixPct}{0.99}
5748+\newcommand{\SecCagrMultilabelCybersecurityBaselinePct}{3.7}
5749+\newcommand{\SecRatioMultilabelCybersecurityBaseline}{1.16}
5750+\newcommand{\SecCagrPrimaryCybersecurityBaselinePct}{1.7}
5751+\newcommand{\SecCagrMultilabelCybersecurityPostPct}{24.7}
5752+\newcommand{\SecRatioMultilabelCybersecurityPost}{1.55}
5753+\newcommand{\SecCagrPrimaryCybersecurityPostPct}{27.2}
5754+\newcommand{\SecCagrMultilabelCybersecuritySpanPct}{21.2}
5755+\newcommand{\SecRatioMultilabelCybersecuritySpan}{1.78}
5756+\newcommand{\SecCagrPrimaryCybersecuritySpanPct}{23.2}
5757+\newcommand{\SecCountCybersecurityAllMultilabel}{829{,}538}
5758+\newcommand{\SecCountCybersecurityAllPrimary}{349{,}730}
5759+\newcommand{\SecCagrMultilabelAppliedCryptographyBaselinePct}{1.9}
5760+\newcommand{\SecRatioMultilabelAppliedCryptographyBaseline}{1.08}
5761+\newcommand{\SecCagrPrimaryAppliedCryptographyBaselinePct}{-1.1}
5762+\newcommand{\SecCagrMultilabelAppliedCryptographyPostPct}{20.4}
5763+\newcommand{\SecRatioMultilabelAppliedCryptographyPost}{1.45}
5764+\newcommand{\SecCagrPrimaryAppliedCryptographyPostPct}{14.1}
5765+\newcommand{\SecCagrMultilabelAppliedCryptographySpanPct}{17.3}
5766+\newcommand{\SecRatioMultilabelAppliedCryptographySpan}{1.61}
5767+\newcommand{\SecCagrPrimaryAppliedCryptographySpanPct}{12.4}
5768+\newcommand{\SecCountAppliedCryptographyAllMultilabel}{329{,}708}
5769+\newcommand{\SecCountAppliedCryptographyAllPrimary}{105{,}074}
5770+\newcommand{\SecCagrMultilabelBlockchainBaselinePct}{16.7}
5771+\newcommand{\SecRatioMultilabelBlockchainBaseline}{1.85}
5772+\newcommand{\SecCagrPrimaryBlockchainBaselinePct}{16.3}
5773+\newcommand{\SecCagrMultilabelBlockchainPostPct}{17.3}
5774+\newcommand{\SecRatioMultilabelBlockchainPost}{1.37}
5775+\newcommand{\SecCagrPrimaryBlockchainPostPct}{12.0}
5776+\newcommand{\SecCagrMultilabelBlockchainSpanPct}{18.4}
5777+\newcommand{\SecRatioMultilabelBlockchainSpan}{1.66}
5778+\newcommand{\SecCagrPrimaryBlockchainSpanPct}{15.2}
5779+\newcommand{\SecCountBlockchainAllMultilabel}{190{,}653}
5780+\newcommand{\SecCountBlockchainAllPrimary}{108{,}609}
5781+\newcommand{\SecCagrMultilabelPrivacyPreservingComputationBaselinePct}{15.2}
5782+\newcommand{\SecRatioMultilabelPrivacyPreservingComputationBaseline}{1.76}
5783+\newcommand{\SecCagrPrimaryPrivacyPreservingComputationBaselinePct}{9.0}
5784+\newcommand{\SecCagrMultilabelPrivacyPreservingComputationPostPct}{39.8}
5785+\newcommand{\SecRatioMultilabelPrivacyPreservingComputationPost}{1.95}
5786+\newcommand{\SecCagrPrimaryPrivacyPreservingComputationPostPct}{32.7}
5787+\newcommand{\SecCagrMultilabelPrivacyPreservingComputationSpanPct}{34.9}
5788+\newcommand{\SecRatioMultilabelPrivacyPreservingComputationSpan}{2.46}
5789+\newcommand{\SecCagrPrimaryPrivacyPreservingComputationSpanPct}{26.3}
5790+\newcommand{\SecCountPrivacyPreservingComputationAllMultilabel}{228{,}836}
5791+\newcommand{\SecCountPrivacyPreservingComputationAllPrimary}{47{,}273}
5792+\newcommand{\SecCagrMultilabelFederatedLearningBaselinePct}{54.3}
5793+\newcommand{\SecRatioMultilabelFederatedLearningBaseline}{5.66}
5794+\newcommand{\SecCagrPrimaryFederatedLearningBaselinePct}{61.5}
5795+\newcommand{\SecCagrMultilabelFederatedLearningPostPct}{45.8}
5796+\newcommand{\SecRatioMultilabelFederatedLearningPost}{2.12}
5797+\newcommand{\SecCagrPrimaryFederatedLearningPostPct}{30.4}
5798+\newcommand{\SecCagrMultilabelFederatedLearningSpanPct}{45.4}
5799+\newcommand{\SecRatioMultilabelFederatedLearningSpan}{3.07}
5800+\newcommand{\SecCagrPrimaryFederatedLearningSpanPct}{37.7}
5801+\newcommand{\SecCountFederatedLearningAllMultilabel}{71{,}037}
5802+\newcommand{\SecCountFederatedLearningAllPrimary}{11{,}240}
5803+\newcommand{\SecCagrAppliedCryptographyNTwoZeroOneEight}{10{,}782}
5804+\newcommand{\SecCagrAppliedCryptographyNTwoZeroTwoTwo}{10{,}299}
5805+\newcommand{\SecCagrAppliedCryptographyNTwoZeroTwoThree}{11{,}219}
5806+\newcommand{\SecCagrAppliedCryptographyNTwoZeroTwoFive}{14{,}612}
5807+\newcommand{\SecCagrAppliedCryptographyCagrBaselinePct}{-1.1}
5808+\newcommand{\SecCagrAppliedCryptographyCagrPostPct}{14.1}
5809+\newcommand{\SecCagrAppliedCryptographyShareTwoZeroOneEightPct}{2.16}
5810+\newcommand{\SecCagrAppliedCryptographyShareTwoZeroTwoTwoPct}{1.84}
5811+\newcommand{\SecCagrAppliedCryptographyShareTwoZeroTwoFivePct}{1.64}
5812+\newcommand{\SecCagrBlockchainNTwoZeroOneEight}{6{,}441}
5813+\newcommand{\SecCagrBlockchainNTwoZeroTwoTwo}{11{,}770}
5814+\newcommand{\SecCagrBlockchainNTwoZeroTwoThree}{14{,}349}
5815+\newcommand{\SecCagrBlockchainNTwoZeroTwoFive}{18{,}012}
5816+\newcommand{\SecCagrBlockchainCagrBaselinePct}{16.3}
5817+\newcommand{\SecCagrBlockchainCagrPostPct}{12.0}
5818+\newcommand{\SecCagrBlockchainShareTwoZeroOneEightPct}{1.29}
5819+\newcommand{\SecCagrBlockchainShareTwoZeroTwoTwoPct}{2.10}
5820+\newcommand{\SecCagrBlockchainShareTwoZeroTwoFivePct}{2.02}
5821+\newcommand{\SecCagrCybersecurityNTwoZeroOneEight}{29{,}440}
5822+\newcommand{\SecCagrCybersecurityNTwoZeroTwoTwo}{31{,}554}
5823+\newcommand{\SecCagrCybersecurityNTwoZeroTwoThree}{36{,}481}
5824+\newcommand{\SecCagrCybersecurityNTwoZeroTwoFive}{59{,}044}
5825+\newcommand{\SecCagrCybersecurityCagrBaselinePct}{1.7}
5826+\newcommand{\SecCagrCybersecurityCagrPostPct}{27.2}
5827+\newcommand{\SecCagrCybersecurityShareTwoZeroOneEightPct}{5.89}
5828+\newcommand{\SecCagrCybersecurityShareTwoZeroTwoTwoPct}{5.63}
5829+\newcommand{\SecCagrCybersecurityShareTwoZeroTwoFivePct}{6.61}
5830+\newcommand{\SecCagrFederatedLearningNTwoZeroOneEight}{143}
5831+\newcommand{\SecCagrFederatedLearningNTwoZeroTwoTwo}{974}
5832+\newcommand{\SecCagrFederatedLearningNTwoZeroTwoThree}{1{,}497}
5833+\newcommand{\SecCagrFederatedLearningNTwoZeroTwoFive}{2{,}544}
5834+\newcommand{\SecCagrFederatedLearningCagrBaselinePct}{61.5}
5835+\newcommand{\SecCagrFederatedLearningCagrPostPct}{30.4}
5836+\newcommand{\SecCagrFederatedLearningShareTwoZeroOneEightPct}{0.03}
5837+\newcommand{\SecCagrFederatedLearningShareTwoZeroTwoTwoPct}{0.17}
5838+\newcommand{\SecCagrFederatedLearningShareTwoZeroTwoFivePct}{0.28}
5839+\newcommand{\SecCagrPrivacyPreservingComputationNTwoZeroOneEight}{3{,}105}
5840+\newcommand{\SecCagrPrivacyPreservingComputationNTwoZeroTwoTwo}{4{,}376}
5841+\newcommand{\SecCagrPrivacyPreservingComputationNTwoZeroTwoThree}{5{,}010}
5842+\newcommand{\SecCagrPrivacyPreservingComputationNTwoZeroTwoFive}{8{,}817}
5843+\newcommand{\SecCagrPrivacyPreservingComputationCagrBaselinePct}{9.0}
5844+\newcommand{\SecCagrPrivacyPreservingComputationCagrPostPct}{32.7}
5845+\newcommand{\SecCagrPrivacyPreservingComputationShareTwoZeroOneEightPct}{0.62}
5846+\newcommand{\SecCagrPrivacyPreservingComputationShareTwoZeroTwoTwoPct}{0.78}
5847+\newcommand{\SecCagrPrivacyPreservingComputationShareTwoZeroTwoFivePct}{0.99}
5848+\newcommand{\SecBreakAppliedCryptographyBpBreaksBic}{2022-07-01}
5849+\newcommand{\SecBreakAppliedCryptographyPelt}{2022-11, 2024-12}
5850+\newcommand{\SecBreakAppliedCryptographyBocpdResets}{none}
5851+\newcommand{\SecBreakAppliedCryptographyChatgptLevel}{0.002}
5852+\newcommand{\SecBreakAppliedCryptographyChatgptLevelP}{0.922}
5853+\newcommand{\SecBreakAppliedCryptographyGptThreeLevel}{-0.101}
5854+\newcommand{\SecBreakAppliedCryptographyGptThreeLevelP}{0.132}
5855+\newcommand{\SecBreakAppliedCryptographyDalleTwoLevel}{-0.052}
5856+\newcommand{\SecBreakAppliedCryptographyDalleTwoLevelP}{0.052}
5857+\newcommand{\SecBreakAppliedCryptographyGptFourLevel}{0.035}
5858+\newcommand{\SecBreakAppliedCryptographyGptFourLevelP}{0.082}
5859+\newcommand{\SecBreakAppliedCryptographyOOneLevel}{0.142}
5860+\newcommand{\SecBreakAppliedCryptographyOOneLevelP}{0.002}
5861+\newcommand{\SecBreakBlockchainBpBreaksBic}{2020-04-01}
5862+\newcommand{\SecBreakBlockchainPelt}{2019-02, 2022-12}
5863+\newcommand{\SecBreakBlockchainBocpdResets}{2019-03}
5864+\newcommand{\SecBreakBlockchainChatgptLevel}{-0.008}
5865+\newcommand{\SecBreakBlockchainChatgptLevelP}{0.858}
5866+\newcommand{\SecBreakBlockchainGptThreeLevel}{-0.065}
5867+\newcommand{\SecBreakBlockchainGptThreeLevelP}{0.362}
5868+\newcommand{\SecBreakBlockchainDalleTwoLevel}{-0.023}
5869+\newcommand{\SecBreakBlockchainDalleTwoLevelP}{0.654}
5870+\newcommand{\SecBreakBlockchainGptFourLevel}{-0.024}
5871+\newcommand{\SecBreakBlockchainGptFourLevelP}{0.571}
5872+\newcommand{\SecBreakBlockchainOOneLevel}{0.007}
5873+\newcommand{\SecBreakBlockchainOOneLevelP}{0.842}
5874+\newcommand{\SecBreakCybersecurityBpBreaksBic}{2022-07-01}
5875+\newcommand{\SecBreakCybersecurityPelt}{2022-11, 2024-12}
5876+\newcommand{\SecBreakCybersecurityBocpdResets}{none}
5877+\newcommand{\SecBreakCybersecurityChatgptLevel}{-0.021}
5878+\newcommand{\SecBreakCybersecurityChatgptLevelP}{0.408}
5879+\newcommand{\SecBreakCybersecurityGptThreeLevel}{-0.113}
5880+\newcommand{\SecBreakCybersecurityGptThreeLevelP}{0.121}
5881+\newcommand{\SecBreakCybersecurityDalleTwoLevel}{-0.083}
5882+\newcommand{\SecBreakCybersecurityDalleTwoLevelP}{0.015}
5883+\newcommand{\SecBreakCybersecurityGptFourLevel}{0.010}
5884+\newcommand{\SecBreakCybersecurityGptFourLevelP}{0.713}
5885+\newcommand{\SecBreakCybersecurityOOneLevel}{0.149}
5886+\newcommand{\SecBreakCybersecurityOOneLevelP}{0.001}
5887+\newcommand{\SecBreakFederatedLearningBpBreaksBic}{2020-02-01}
5888+\newcommand{\SecBreakFederatedLearningPelt}{2020-02, 2023-05}
5889+\newcommand{\SecBreakFederatedLearningBocpdResets}{none}
5890+\newcommand{\SecBreakFederatedLearningChatgptLevel}{-0.126}
5891+\newcommand{\SecBreakFederatedLearningChatgptLevelP}{0.007}
5892+\newcommand{\SecBreakFederatedLearningGptThreeLevel}{0.166}
5893+\newcommand{\SecBreakFederatedLearningGptThreeLevelP}{0.002}
5894+\newcommand{\SecBreakFederatedLearningDalleTwoLevel}{-0.101}
5895+\newcommand{\SecBreakFederatedLearningDalleTwoLevelP}{0.067}
5896+\newcommand{\SecBreakFederatedLearningGptFourLevel}{-0.092}
5897+\newcommand{\SecBreakFederatedLearningGptFourLevelP}{0.075}
5898+\newcommand{\SecBreakFederatedLearningOOneLevel}{-0.115}
5899+\newcommand{\SecBreakFederatedLearningOOneLevelP}{0.010}
5900+\newcommand{\SecBreakPrivacyPreservingComputationBpBreaksBic}{2022-07-01}
5901+\newcommand{\SecBreakPrivacyPreservingComputationPelt}{2020-12, 2023-12}
5902+\newcommand{\SecBreakPrivacyPreservingComputationBocpdResets}{none}
5903+\newcommand{\SecBreakPrivacyPreservingComputationChatgptLevel}{-0.001}
5904+\newcommand{\SecBreakPrivacyPreservingComputationChatgptLevelP}{0.965}
5905+\newcommand{\SecBreakPrivacyPreservingComputationGptThreeLevel}{-0.044}
5906+\newcommand{\SecBreakPrivacyPreservingComputationGptThreeLevelP}{0.469}
5907+\newcommand{\SecBreakPrivacyPreservingComputationDalleTwoLevel}{-0.073}
5908+\newcommand{\SecBreakPrivacyPreservingComputationDalleTwoLevelP}{0.037}
5909+\newcommand{\SecBreakPrivacyPreservingComputationGptFourLevel}{0.042}
5910+\newcommand{\SecBreakPrivacyPreservingComputationGptFourLevelP}{0.200}
5911+\newcommand{\SecBreakPrivacyPreservingComputationOOneLevel}{0.198}
5912+\newcommand{\SecBreakPrivacyPreservingComputationOOneLevelP}{0.001}
5913+\newcommand{\SecFlowAiAgentsToCybersecurityTwoZeroOneNinePct}{1.6}
5914+\newcommand{\SecFlowAiAgentsToBlockchainTwoZeroOneNinePct}{0.7}
5915+\newcommand{\SecFlowAiAgentsToPrivacyPreservingComputationTwoZeroOneNinePct}{0.2}
5916+\newcommand{\SecFlowAiAgentsToAppliedCryptographyTwoZeroOneNinePct}{0.2}
5917+\newcommand{\SecFlowAiAgentsToFederatedLearningTwoZeroOneNinePct}{0.0}
5918+\newcommand{\SecFlowAiSafetyToCybersecurityTwoZeroOneNinePct}{6.2}
5919+\newcommand{\SecFlowAiSafetyToPrivacyPreservingComputationTwoZeroOneNinePct}{1.8}
5920+\newcommand{\SecFlowAiSafetyToAppliedCryptographyTwoZeroOneNinePct}{0.4}
5921+\newcommand{\SecFlowAiSafetyToBlockchainTwoZeroOneNinePct}{0.4}
5922+\newcommand{\SecFlowAppliedCryptographyToAppliedCryptographyTwoZeroOneNinePct}{54.3}
5923+\newcommand{\SecFlowAppliedCryptographyToCybersecurityTwoZeroOneNinePct}{13.8}
5924+\newcommand{\SecFlowAppliedCryptographyToComputerNetworksTwoZeroOneNinePct}{6.0}
5925+\newcommand{\SecFlowAppliedCryptographyToComputerVisionTwoZeroOneNinePct}{5.4}
5926+\newcommand{\SecFlowAppliedCryptographyToBlockchainTwoZeroOneNinePct}{3.9}
5927+\newcommand{\SecFlowAppliedCryptographyToQuantumComputingTwoZeroOneNinePct}{2.8}
5928+\newcommand{\SecFlowAppliedCryptographyToPrivacyPreservingComputationTwoZeroOneNinePct}{2.0}
5929+\newcommand{\SecFlowAppliedCryptographyToEdgeComputingTwoZeroOneNinePct}{1.3}
5930+\newcommand{\SecFlowAppliedCryptographyToDistributedComputingTwoZeroOneNinePct}{1.2}
5931+\newcommand{\SecFlowAppliedCryptographyToDatabaseSystemsTwoZeroOneNinePct}{0.5}
5932+\newcommand{\SecFlowAppliedCryptographyToMlTheoryTwoZeroOneNinePct}{0.5}
5933+\newcommand{\SecFlowAppliedCryptographyToSoftwareEngineeringTwoZeroOneNinePct}{0.3}
5934+\newcommand{\SecFlowAppliedCryptographyToHciTwoZeroOneNinePct}{0.3}
5935+\newcommand{\SecFlowAppliedCryptographyToAiAgentsTwoZeroOneNinePct}{0.1}
5936+\newcommand{\SecFlowAppliedCryptographyToGenerativeAiTwoZeroOneNinePct}{0.0}
5937+\newcommand{\SecFlowAppliedCryptographyToAiSafetyTwoZeroOneNinePct}{0.0}
5938+\newcommand{\SecFlowAppliedCryptographyToLlmTwoZeroOneNinePct}{0.0}
5939+\newcommand{\SecFlowBlockchainToBlockchainTwoZeroOneNinePct}{61.4}
5940+\newcommand{\SecFlowBlockchainToCybersecurityTwoZeroOneNinePct}{8.5}
5941+\newcommand{\SecFlowBlockchainToComputerNetworksTwoZeroOneNinePct}{5.8}
5942+\newcommand{\SecFlowBlockchainToDistributedComputingTwoZeroOneNinePct}{5.8}
5943+\newcommand{\SecFlowBlockchainToEdgeComputingTwoZeroOneNinePct}{5.1}
5944+\newcommand{\SecFlowBlockchainToAppliedCryptographyTwoZeroOneNinePct}{4.1}
5945+\newcommand{\SecFlowBlockchainToDatabaseSystemsTwoZeroOneNinePct}{1.1}
5946+\newcommand{\SecFlowBlockchainToSoftwareEngineeringTwoZeroOneNinePct}{1.1}
5947+\newcommand{\SecFlowBlockchainToPrivacyPreservingComputationTwoZeroOneNinePct}{1.0}
5948+\newcommand{\SecFlowBlockchainToHciTwoZeroOneNinePct}{0.5}
5949+\newcommand{\SecFlowBlockchainToComputerVisionTwoZeroOneNinePct}{0.5}
5950+\newcommand{\SecFlowBlockchainToMlTheoryTwoZeroOneNinePct}{0.3}
5951+\newcommand{\SecFlowBlockchainToAiAgentsTwoZeroOneNinePct}{0.3}
5952+\newcommand{\SecFlowBlockchainToAiSafetyTwoZeroOneNinePct}{0.2}
5953+\newcommand{\SecFlowBlockchainToQuantumComputingTwoZeroOneNinePct}{0.0}
5954+\newcommand{\SecFlowBlockchainToGenerativeAiTwoZeroOneNinePct}{0.0}
5955+\newcommand{\SecFlowBlockchainToLlmTwoZeroOneNinePct}{0.0}
5956+\newcommand{\SecFlowBlockchainToFederatedLearningTwoZeroOneNinePct}{0.0}
5957+\newcommand{\SecFlowComputerNetworksToCybersecurityTwoZeroOneNinePct}{3.9}
5958+\newcommand{\SecFlowComputerNetworksToAppliedCryptographyTwoZeroOneNinePct}{0.7}
5959+\newcommand{\SecFlowComputerNetworksToBlockchainTwoZeroOneNinePct}{0.6}
5960+\newcommand{\SecFlowComputerNetworksToPrivacyPreservingComputationTwoZeroOneNinePct}{0.4}
5961+\newcommand{\SecFlowComputerNetworksToFederatedLearningTwoZeroOneNinePct}{0.0}
5962+\newcommand{\SecFlowCybersecurityToCybersecurityTwoZeroOneNinePct}{49.2}
5963+\newcommand{\SecFlowCybersecurityToComputerNetworksTwoZeroOneNinePct}{10.3}
5964+\newcommand{\SecFlowCybersecurityToComputerVisionTwoZeroOneNinePct}{7.7}
5965+\newcommand{\SecFlowCybersecurityToAppliedCryptographyTwoZeroOneNinePct}{6.3}
5966+\newcommand{\SecFlowCybersecurityToBlockchainTwoZeroOneNinePct}{2.7}
5967+\newcommand{\SecFlowCybersecurityToSoftwareEngineeringTwoZeroOneNinePct}{2.5}
5968+\newcommand{\SecFlowCybersecurityToEdgeComputingTwoZeroOneNinePct}{2.1}
5969+\newcommand{\SecFlowCybersecurityToMlTheoryTwoZeroOneNinePct}{2.0}
5970+\newcommand{\SecFlowCybersecurityToPrivacyPreservingComputationTwoZeroOneNinePct}{1.8}
5971+\newcommand{\SecFlowCybersecurityToHciTwoZeroOneNinePct}{1.0}
5972+\newcommand{\SecFlowCybersecurityToDistributedComputingTwoZeroOneNinePct}{0.8}
5973+\newcommand{\SecFlowCybersecurityToDatabaseSystemsTwoZeroOneNinePct}{0.8}
5974+\newcommand{\SecFlowCybersecurityToQuantumComputingTwoZeroOneNinePct}{0.3}
5975+\newcommand{\SecFlowCybersecurityToAiAgentsTwoZeroOneNinePct}{0.3}
5976+\newcommand{\SecFlowCybersecurityToAiSafetyTwoZeroOneNinePct}{0.1}
5977+\newcommand{\SecFlowCybersecurityToGenerativeAiTwoZeroOneNinePct}{0.1}
5978+\newcommand{\SecFlowCybersecurityToLlmTwoZeroOneNinePct}{0.1}
5979+\newcommand{\SecFlowCybersecurityToFederatedLearningTwoZeroOneNinePct}{0.0}
5980+\newcommand{\SecFlowDistributedComputingToBlockchainTwoZeroOneNinePct}{4.8}
5981+\newcommand{\SecFlowDistributedComputingToCybersecurityTwoZeroOneNinePct}{2.3}
5982+\newcommand{\SecFlowDistributedComputingToAppliedCryptographyTwoZeroOneNinePct}{1.4}
5983+\newcommand{\SecFlowDistributedComputingToPrivacyPreservingComputationTwoZeroOneNinePct}{0.3}
5984+\newcommand{\SecFlowDistributedComputingToFederatedLearningTwoZeroOneNinePct}{0.0}
5985+\newcommand{\SecFlowEdgeComputingToCybersecurityTwoZeroOneNinePct}{7.0}
5986+\newcommand{\SecFlowEdgeComputingToBlockchainTwoZeroOneNinePct}{4.3}
5987+\newcommand{\SecFlowEdgeComputingToAppliedCryptographyTwoZeroOneNinePct}{1.3}
5988+\newcommand{\SecFlowEdgeComputingToPrivacyPreservingComputationTwoZeroOneNinePct}{1.2}
5989+\newcommand{\SecFlowEdgeComputingToFederatedLearningTwoZeroOneNinePct}{0.0}
5990+\newcommand{\SecFlowFederatedLearningToMlTheoryTwoZeroOneNinePct}{17.6}
5991+\newcommand{\SecFlowFederatedLearningToFederatedLearningTwoZeroOneNinePct}{10.8}
5992+\newcommand{\SecFlowFederatedLearningToPrivacyPreservingComputationTwoZeroOneNinePct}{9.0}
5993+\newcommand{\SecFlowFederatedLearningToComputerNetworksTwoZeroOneNinePct}{7.5}
5994+\newcommand{\SecFlowFederatedLearningToDistributedComputingTwoZeroOneNinePct}{7.2}
5995+\newcommand{\SecFlowFederatedLearningToCybersecurityTwoZeroOneNinePct}{6.5}
5996+\newcommand{\SecFlowFederatedLearningToComputerVisionTwoZeroOneNinePct}{5.7}
5997+\newcommand{\SecFlowFederatedLearningToEdgeComputingTwoZeroOneNinePct}{3.2}
5998+\newcommand{\SecFlowFederatedLearningToAppliedCryptographyTwoZeroOneNinePct}{2.5}
5999+\newcommand{\SecFlowFederatedLearningToHciTwoZeroOneNinePct}{2.5}
6000+\newcommand{\SecFlowFederatedLearningToBlockchainTwoZeroOneNinePct}{1.1}
6001+\newcommand{\SecFlowFederatedLearningToAiAgentsTwoZeroOneNinePct}{1.1}
6002+\newcommand{\SecFlowFederatedLearningToAiSafetyTwoZeroOneNinePct}{0.7}
6003+\newcommand{\SecFlowFederatedLearningToDatabaseSystemsTwoZeroOneNinePct}{0.7}
6004+\newcommand{\SecFlowFederatedLearningToGenerativeAiTwoZeroOneNinePct}{0.4}
6005+\newcommand{\SecFlowFederatedLearningToSoftwareEngineeringTwoZeroOneNinePct}{0.4}
6006+\newcommand{\SecFlowGenerativeAiToCybersecurityTwoZeroOneNinePct}{2.2}
6007+\newcommand{\SecFlowGenerativeAiToPrivacyPreservingComputationTwoZeroOneNinePct}{1.3}
6008+\newcommand{\SecFlowGenerativeAiToAppliedCryptographyTwoZeroOneNinePct}{0.0}
6009+\newcommand{\SecFlowGenerativeAiToFederatedLearningTwoZeroOneNinePct}{0.0}
6010+\newcommand{\SecFlowLlmToCybersecurityTwoZeroOneNinePct}{0.9}
6011+\newcommand{\SecFlowLlmToPrivacyPreservingComputationTwoZeroOneNinePct}{0.1}
6012+\newcommand{\SecFlowLlmToAppliedCryptographyTwoZeroOneNinePct}{0.0}
6013+\newcommand{\SecFlowLlmToBlockchainTwoZeroOneNinePct}{0.0}
6014+\newcommand{\SecFlowLlmToFederatedLearningTwoZeroOneNinePct}{0.0}
6015+\newcommand{\SecFlowPrivacyPreservingComputationToPrivacyPreservingComputationTwoZeroOneNinePct}{32.5}
6016+\newcommand{\SecFlowPrivacyPreservingComputationToCybersecurityTwoZeroOneNinePct}{15.3}
6017+\newcommand{\SecFlowPrivacyPreservingComputationToComputerNetworksTwoZeroOneNinePct}{9.3}
6018+\newcommand{\SecFlowPrivacyPreservingComputationToAppliedCryptographyTwoZeroOneNinePct}{7.0}
6019+\newcommand{\SecFlowPrivacyPreservingComputationToMlTheoryTwoZeroOneNinePct}{4.9}
6020+\newcommand{\SecFlowPrivacyPreservingComputationToComputerVisionTwoZeroOneNinePct}{4.0}
6021+\newcommand{\SecFlowPrivacyPreservingComputationToDatabaseSystemsTwoZeroOneNinePct}{3.4}
6022+\newcommand{\SecFlowPrivacyPreservingComputationToEdgeComputingTwoZeroOneNinePct}{3.0}
6023+\newcommand{\SecFlowPrivacyPreservingComputationToHciTwoZeroOneNinePct}{2.8}
6024+\newcommand{\SecFlowPrivacyPreservingComputationToBlockchainTwoZeroOneNinePct}{1.8}
6025+\newcommand{\SecFlowPrivacyPreservingComputationToDistributedComputingTwoZeroOneNinePct}{1.0}
6026+\newcommand{\SecFlowPrivacyPreservingComputationToSoftwareEngineeringTwoZeroOneNinePct}{0.9}
6027+\newcommand{\SecFlowPrivacyPreservingComputationToAiSafetyTwoZeroOneNinePct}{0.7}
6028+\newcommand{\SecFlowPrivacyPreservingComputationToGenerativeAiTwoZeroOneNinePct}{0.4}
6029+\newcommand{\SecFlowPrivacyPreservingComputationToFederatedLearningTwoZeroOneNinePct}{0.3}
6030+\newcommand{\SecFlowPrivacyPreservingComputationToAiAgentsTwoZeroOneNinePct}{0.2}
6031+\newcommand{\SecFlowPrivacyPreservingComputationToLlmTwoZeroOneNinePct}{0.1}
6032+\newcommand{\SecFlowPrivacyPreservingComputationToQuantumComputingTwoZeroOneNinePct}{0.1}
6033+\newcommand{\SecFlowAiAgentsToCybersecurityTwoZeroTwoTwoPct}{2.5}
6034+\newcommand{\SecFlowAiAgentsToBlockchainTwoZeroTwoTwoPct}{0.6}
6035+\newcommand{\SecFlowAiAgentsToPrivacyPreservingComputationTwoZeroTwoTwoPct}{0.3}
6036+\newcommand{\SecFlowAiAgentsToAppliedCryptographyTwoZeroTwoTwoPct}{0.3}
6037+\newcommand{\SecFlowAiAgentsToFederatedLearningTwoZeroTwoTwoPct}{0.0}
6038+\newcommand{\SecFlowAiSafetyToCybersecurityTwoZeroTwoTwoPct}{4.1}
6039+\newcommand{\SecFlowAiSafetyToPrivacyPreservingComputationTwoZeroTwoTwoPct}{2.1}
6040+\newcommand{\SecFlowAiSafetyToBlockchainTwoZeroTwoTwoPct}{0.5}
6041+\newcommand{\SecFlowAiSafetyToAppliedCryptographyTwoZeroTwoTwoPct}{0.2}
6042+\newcommand{\SecFlowAiSafetyToFederatedLearningTwoZeroTwoTwoPct}{0.1}
6043+\newcommand{\SecFlowAppliedCryptographyToAppliedCryptographyTwoZeroTwoTwoPct}{52.0}
6044+\newcommand{\SecFlowAppliedCryptographyToCybersecurityTwoZeroTwoTwoPct}{13.5}
6045+\newcommand{\SecFlowAppliedCryptographyToComputerVisionTwoZeroTwoTwoPct}{6.4}
6046+\newcommand{\SecFlowAppliedCryptographyToComputerNetworksTwoZeroTwoTwoPct}{5.4}
6047+\newcommand{\SecFlowAppliedCryptographyToBlockchainTwoZeroTwoTwoPct}{4.7}
6048+\newcommand{\SecFlowAppliedCryptographyToQuantumComputingTwoZeroTwoTwoPct}{3.0}
6049+\newcommand{\SecFlowAppliedCryptographyToPrivacyPreservingComputationTwoZeroTwoTwoPct}{2.1}
6050+\newcommand{\SecFlowAppliedCryptographyToEdgeComputingTwoZeroTwoTwoPct}{1.8}
6051+\newcommand{\SecFlowAppliedCryptographyToDistributedComputingTwoZeroTwoTwoPct}{1.0}
6052+\newcommand{\SecFlowAppliedCryptographyToMlTheoryTwoZeroTwoTwoPct}{0.5}
6053+\newcommand{\SecFlowAppliedCryptographyToDatabaseSystemsTwoZeroTwoTwoPct}{0.5}
6054+\newcommand{\SecFlowAppliedCryptographyToSoftwareEngineeringTwoZeroTwoTwoPct}{0.3}
6055+\newcommand{\SecFlowAppliedCryptographyToHciTwoZeroTwoTwoPct}{0.2}
6056+\newcommand{\SecFlowAppliedCryptographyToFederatedLearningTwoZeroTwoTwoPct}{0.1}
6057+\newcommand{\SecFlowAppliedCryptographyToAiAgentsTwoZeroTwoTwoPct}{0.1}
6058+\newcommand{\SecFlowAppliedCryptographyToGenerativeAiTwoZeroTwoTwoPct}{0.0}
6059+\newcommand{\SecFlowAppliedCryptographyToAiSafetyTwoZeroTwoTwoPct}{0.0}
6060+\newcommand{\SecFlowAppliedCryptographyToLlmTwoZeroTwoTwoPct}{0.0}
6061+\newcommand{\SecFlowBlockchainToBlockchainTwoZeroTwoTwoPct}{61.7}
6062+\newcommand{\SecFlowBlockchainToCybersecurityTwoZeroTwoTwoPct}{7.4}
6063+\newcommand{\SecFlowBlockchainToComputerNetworksTwoZeroTwoTwoPct}{7.0}
6064+\newcommand{\SecFlowBlockchainToEdgeComputingTwoZeroTwoTwoPct}{5.5}
6065+\newcommand{\SecFlowBlockchainToDistributedComputingTwoZeroTwoTwoPct}{3.8}
6066+\newcommand{\SecFlowBlockchainToAppliedCryptographyTwoZeroTwoTwoPct}{3.6}
6067+\newcommand{\SecFlowBlockchainToSoftwareEngineeringTwoZeroTwoTwoPct}{1.3}
6068+\newcommand{\SecFlowBlockchainToPrivacyPreservingComputationTwoZeroTwoTwoPct}{1.1}
6069+\newcommand{\SecFlowBlockchainToDatabaseSystemsTwoZeroTwoTwoPct}{1.0}
6070+\newcommand{\SecFlowBlockchainToComputerVisionTwoZeroTwoTwoPct}{1.0}
6071+\newcommand{\SecFlowBlockchainToHciTwoZeroTwoTwoPct}{0.7}
6072+\newcommand{\SecFlowBlockchainToMlTheoryTwoZeroTwoTwoPct}{0.6}
6073+\newcommand{\SecFlowBlockchainToAiAgentsTwoZeroTwoTwoPct}{0.2}
6074+\newcommand{\SecFlowBlockchainToFederatedLearningTwoZeroTwoTwoPct}{0.2}
6075+\newcommand{\SecFlowBlockchainToQuantumComputingTwoZeroTwoTwoPct}{0.2}
6076+\newcommand{\SecFlowBlockchainToAiSafetyTwoZeroTwoTwoPct}{0.1}
6077+\newcommand{\SecFlowBlockchainToGenerativeAiTwoZeroTwoTwoPct}{0.0}
6078+\newcommand{\SecFlowBlockchainToLlmTwoZeroTwoTwoPct}{0.0}
6079+\newcommand{\SecFlowComputerNetworksToCybersecurityTwoZeroTwoTwoPct}{5.4}
6080+\newcommand{\SecFlowComputerNetworksToBlockchainTwoZeroTwoTwoPct}{1.1}
6081+\newcommand{\SecFlowComputerNetworksToAppliedCryptographyTwoZeroTwoTwoPct}{0.7}
6082+\newcommand{\SecFlowComputerNetworksToPrivacyPreservingComputationTwoZeroTwoTwoPct}{0.5}
6083+\newcommand{\SecFlowComputerNetworksToFederatedLearningTwoZeroTwoTwoPct}{0.1}
6084+\newcommand{\SecFlowCybersecurityToCybersecurityTwoZeroTwoTwoPct}{47.8}
6085+\newcommand{\SecFlowCybersecurityToComputerNetworksTwoZeroTwoTwoPct}{12.5}
6086+\newcommand{\SecFlowCybersecurityToComputerVisionTwoZeroTwoTwoPct}{8.0}
6087+\newcommand{\SecFlowCybersecurityToAppliedCryptographyTwoZeroTwoTwoPct}{5.2}
6088+\newcommand{\SecFlowCybersecurityToBlockchainTwoZeroTwoTwoPct}{3.0}
6089+\newcommand{\SecFlowCybersecurityToEdgeComputingTwoZeroTwoTwoPct}{2.4}
6090+\newcommand{\SecFlowCybersecurityToSoftwareEngineeringTwoZeroTwoTwoPct}{2.3}
6091+\newcommand{\SecFlowCybersecurityToMlTheoryTwoZeroTwoTwoPct}{2.2}
6092+\newcommand{\SecFlowCybersecurityToPrivacyPreservingComputationTwoZeroTwoTwoPct}{1.8}
6093+\newcommand{\SecFlowCybersecurityToHciTwoZeroTwoTwoPct}{1.2}
6094+\newcommand{\SecFlowCybersecurityToDistributedComputingTwoZeroTwoTwoPct}{0.7}
6095+\newcommand{\SecFlowCybersecurityToDatabaseSystemsTwoZeroTwoTwoPct}{0.6}
6096+\newcommand{\SecFlowCybersecurityToQuantumComputingTwoZeroTwoTwoPct}{0.3}
6097+\newcommand{\SecFlowCybersecurityToAiSafetyTwoZeroTwoTwoPct}{0.3}
6098+\newcommand{\SecFlowCybersecurityToAiAgentsTwoZeroTwoTwoPct}{0.3}
6099+\newcommand{\SecFlowCybersecurityToGenerativeAiTwoZeroTwoTwoPct}{0.2}
6100+\newcommand{\SecFlowCybersecurityToLlmTwoZeroTwoTwoPct}{0.1}
6101+\newcommand{\SecFlowCybersecurityToFederatedLearningTwoZeroTwoTwoPct}{0.1}
6102+\newcommand{\SecFlowDistributedComputingToBlockchainTwoZeroTwoTwoPct}{5.4}
6103+\newcommand{\SecFlowDistributedComputingToCybersecurityTwoZeroTwoTwoPct}{2.6}
6104+\newcommand{\SecFlowDistributedComputingToAppliedCryptographyTwoZeroTwoTwoPct}{1.3}
6105+\newcommand{\SecFlowDistributedComputingToPrivacyPreservingComputationTwoZeroTwoTwoPct}{0.5}
6106+\newcommand{\SecFlowDistributedComputingToFederatedLearningTwoZeroTwoTwoPct}{0.2}
6107+\newcommand{\SecFlowEdgeComputingToCybersecurityTwoZeroTwoTwoPct}{6.5}
6108+\newcommand{\SecFlowEdgeComputingToBlockchainTwoZeroTwoTwoPct}{5.3}
6109+\newcommand{\SecFlowEdgeComputingToAppliedCryptographyTwoZeroTwoTwoPct}{1.5}
6110+\newcommand{\SecFlowEdgeComputingToPrivacyPreservingComputationTwoZeroTwoTwoPct}{1.5}
6111+\newcommand{\SecFlowEdgeComputingToFederatedLearningTwoZeroTwoTwoPct}{0.5}
6112+\newcommand{\SecFlowFederatedLearningToMlTheoryTwoZeroTwoTwoPct}{17.9}
6113+\newcommand{\SecFlowFederatedLearningToComputerNetworksTwoZeroTwoTwoPct}{13.7}
6114+\newcommand{\SecFlowFederatedLearningToPrivacyPreservingComputationTwoZeroTwoTwoPct}{11.6}
6115+\newcommand{\SecFlowFederatedLearningToComputerVisionTwoZeroTwoTwoPct}{10.3}
6116+\newcommand{\SecFlowFederatedLearningToFederatedLearningTwoZeroTwoTwoPct}{10.3}
6117+\newcommand{\SecFlowFederatedLearningToEdgeComputingTwoZeroTwoTwoPct}{6.5}
6118+\newcommand{\SecFlowFederatedLearningToCybersecurityTwoZeroTwoTwoPct}{6.1}
6119+\newcommand{\SecFlowFederatedLearningToDistributedComputingTwoZeroTwoTwoPct}{2.8}
6120+\newcommand{\SecFlowFederatedLearningToAppliedCryptographyTwoZeroTwoTwoPct}{2.3}
6121+\newcommand{\SecFlowFederatedLearningToBlockchainTwoZeroTwoTwoPct}{1.8}
6122+\newcommand{\SecFlowFederatedLearningToHciTwoZeroTwoTwoPct}{0.9}
6123+\newcommand{\SecFlowFederatedLearningToAiSafetyTwoZeroTwoTwoPct}{0.7}
6124+\newcommand{\SecFlowFederatedLearningToSoftwareEngineeringTwoZeroTwoTwoPct}{0.7}
6125+\newcommand{\SecFlowFederatedLearningToDatabaseSystemsTwoZeroTwoTwoPct}{0.6}
6126+\newcommand{\SecFlowFederatedLearningToGenerativeAiTwoZeroTwoTwoPct}{0.3}
6127+\newcommand{\SecFlowFederatedLearningToLlmTwoZeroTwoTwoPct}{0.3}
6128+\newcommand{\SecFlowFederatedLearningToQuantumComputingTwoZeroTwoTwoPct}{0.2}
6129+\newcommand{\SecFlowFederatedLearningToAiAgentsTwoZeroTwoTwoPct}{0.2}
6130+\newcommand{\SecFlowGenerativeAiToCybersecurityTwoZeroTwoTwoPct}{1.8}
6131+\newcommand{\SecFlowGenerativeAiToPrivacyPreservingComputationTwoZeroTwoTwoPct}{0.9}
6132+\newcommand{\SecFlowGenerativeAiToBlockchainTwoZeroTwoTwoPct}{0.2}
6133+\newcommand{\SecFlowGenerativeAiToAppliedCryptographyTwoZeroTwoTwoPct}{0.1}
6134+\newcommand{\SecFlowGenerativeAiToFederatedLearningTwoZeroTwoTwoPct}{0.0}
6135+\newcommand{\SecFlowLlmToCybersecurityTwoZeroTwoTwoPct}{0.9}
6136+\newcommand{\SecFlowLlmToPrivacyPreservingComputationTwoZeroTwoTwoPct}{0.2}
6137+\newcommand{\SecFlowLlmToBlockchainTwoZeroTwoTwoPct}{0.1}
6138+\newcommand{\SecFlowLlmToAppliedCryptographyTwoZeroTwoTwoPct}{0.0}
6139+\newcommand{\SecFlowLlmToFederatedLearningTwoZeroTwoTwoPct}{0.0}
6140+\newcommand{\SecFlowPrivacyPreservingComputationToPrivacyPreservingComputationTwoZeroTwoTwoPct}{31.4}
6141+\newcommand{\SecFlowPrivacyPreservingComputationToCybersecurityTwoZeroTwoTwoPct}{13.1}
6142+\newcommand{\SecFlowPrivacyPreservingComputationToComputerNetworksTwoZeroTwoTwoPct}{7.4}
6143+\newcommand{\SecFlowPrivacyPreservingComputationToMlTheoryTwoZeroTwoTwoPct}{6.5}
6144+\newcommand{\SecFlowPrivacyPreservingComputationToComputerVisionTwoZeroTwoTwoPct}{6.4}
6145+\newcommand{\SecFlowPrivacyPreservingComputationToAppliedCryptographyTwoZeroTwoTwoPct}{5.4}
6146+\newcommand{\SecFlowPrivacyPreservingComputationToHciTwoZeroTwoTwoPct}{3.5}
6147+\newcommand{\SecFlowPrivacyPreservingComputationToEdgeComputingTwoZeroTwoTwoPct}{3.4}
6148+\newcommand{\SecFlowPrivacyPreservingComputationToFederatedLearningTwoZeroTwoTwoPct}{2.6}
6149+\newcommand{\SecFlowPrivacyPreservingComputationToBlockchainTwoZeroTwoTwoPct}{2.1}
6150+\newcommand{\SecFlowPrivacyPreservingComputationToDatabaseSystemsTwoZeroTwoTwoPct}{2.0}
6151+\newcommand{\SecFlowPrivacyPreservingComputationToAiSafetyTwoZeroTwoTwoPct}{1.5}
6152+\newcommand{\SecFlowPrivacyPreservingComputationToDistributedComputingTwoZeroTwoTwoPct}{1.1}
6153+\newcommand{\SecFlowPrivacyPreservingComputationToSoftwareEngineeringTwoZeroTwoTwoPct}{0.9}
6154+\newcommand{\SecFlowPrivacyPreservingComputationToGenerativeAiTwoZeroTwoTwoPct}{0.5}
6155+\newcommand{\SecFlowPrivacyPreservingComputationToAiAgentsTwoZeroTwoTwoPct}{0.4}
6156+\newcommand{\SecFlowPrivacyPreservingComputationToLlmTwoZeroTwoTwoPct}{0.2}
6157+\newcommand{\SecFlowPrivacyPreservingComputationToQuantumComputingTwoZeroTwoTwoPct}{0.1}
6158+\newcommand{\SecFlowAiAgentsToCybersecurityTwoZeroTwoFivePct}{2.5}
6159+\newcommand{\SecFlowAiAgentsToPrivacyPreservingComputationTwoZeroTwoFivePct}{0.6}
6160+\newcommand{\SecFlowAiAgentsToBlockchainTwoZeroTwoFivePct}{0.5}
6161+\newcommand{\SecFlowAiAgentsToAppliedCryptographyTwoZeroTwoFivePct}{0.4}
6162+\newcommand{\SecFlowAiAgentsToFederatedLearningTwoZeroTwoFivePct}{0.1}
6163+\newcommand{\SecFlowAiSafetyToCybersecurityTwoZeroTwoFivePct}{4.0}
6164+\newcommand{\SecFlowAiSafetyToPrivacyPreservingComputationTwoZeroTwoFivePct}{2.0}
6165+\newcommand{\SecFlowAiSafetyToBlockchainTwoZeroTwoFivePct}{0.5}
6166+\newcommand{\SecFlowAiSafetyToFederatedLearningTwoZeroTwoFivePct}{0.3}
6167+\newcommand{\SecFlowAiSafetyToAppliedCryptographyTwoZeroTwoFivePct}{0.2}
6168+\newcommand{\SecFlowAppliedCryptographyToAppliedCryptographyTwoZeroTwoFivePct}{49.5}
6169+\newcommand{\SecFlowAppliedCryptographyToCybersecurityTwoZeroTwoFivePct}{14.1}
6170+\newcommand{\SecFlowAppliedCryptographyToComputerVisionTwoZeroTwoFivePct}{5.8}
6171+\newcommand{\SecFlowAppliedCryptographyToComputerNetworksTwoZeroTwoFivePct}{5.5}
6172+\newcommand{\SecFlowAppliedCryptographyToBlockchainTwoZeroTwoFivePct}{5.1}
6173+\newcommand{\SecFlowAppliedCryptographyToQuantumComputingTwoZeroTwoFivePct}{4.1}
6174+\newcommand{\SecFlowAppliedCryptographyToPrivacyPreservingComputationTwoZeroTwoFivePct}{2.4}
6175+\newcommand{\SecFlowAppliedCryptographyToEdgeComputingTwoZeroTwoFivePct}{2.0}
6176+\newcommand{\SecFlowAppliedCryptographyToDistributedComputingTwoZeroTwoFivePct}{1.1}
6177+\newcommand{\SecFlowAppliedCryptographyToMlTheoryTwoZeroTwoFivePct}{0.6}
6178+\newcommand{\SecFlowAppliedCryptographyToFederatedLearningTwoZeroTwoFivePct}{0.4}
6179+\newcommand{\SecFlowAppliedCryptographyToDatabaseSystemsTwoZeroTwoFivePct}{0.4}
6180+\newcommand{\SecFlowAppliedCryptographyToHciTwoZeroTwoFivePct}{0.2}
6181+\newcommand{\SecFlowAppliedCryptographyToSoftwareEngineeringTwoZeroTwoFivePct}{0.2}
6182+\newcommand{\SecFlowAppliedCryptographyToAiAgentsTwoZeroTwoFivePct}{0.1}
6183+\newcommand{\SecFlowAppliedCryptographyToGenerativeAiTwoZeroTwoFivePct}{0.1}
6184+\newcommand{\SecFlowAppliedCryptographyToAiSafetyTwoZeroTwoFivePct}{0.1}
6185+\newcommand{\SecFlowAppliedCryptographyToLlmTwoZeroTwoFivePct}{0.1}
6186+\newcommand{\SecFlowBlockchainToBlockchainTwoZeroTwoFivePct}{57.2}
6187+\newcommand{\SecFlowBlockchainToCybersecurityTwoZeroTwoFivePct}{9.2}
6188+\newcommand{\SecFlowBlockchainToComputerNetworksTwoZeroTwoFivePct}{6.3}
6189+\newcommand{\SecFlowBlockchainToEdgeComputingTwoZeroTwoFivePct}{5.1}
6190+\newcommand{\SecFlowBlockchainToAppliedCryptographyTwoZeroTwoFivePct}{4.1}
6191+\newcommand{\SecFlowBlockchainToDistributedComputingTwoZeroTwoFivePct}{3.6}
6192+\newcommand{\SecFlowBlockchainToSoftwareEngineeringTwoZeroTwoFivePct}{1.8}
6193+\newcommand{\SecFlowBlockchainToPrivacyPreservingComputationTwoZeroTwoFivePct}{1.4}
6194+\newcommand{\SecFlowBlockchainToComputerVisionTwoZeroTwoFivePct}{1.3}
6195+\newcommand{\SecFlowBlockchainToHciTwoZeroTwoFivePct}{1.1}
6196+\newcommand{\SecFlowBlockchainToMlTheoryTwoZeroTwoFivePct}{1.0}
6197+\newcommand{\SecFlowBlockchainToDatabaseSystemsTwoZeroTwoFivePct}{0.9}
6198+\newcommand{\SecFlowBlockchainToFederatedLearningTwoZeroTwoFivePct}{0.4}
6199+\newcommand{\SecFlowBlockchainToQuantumComputingTwoZeroTwoFivePct}{0.3}
6200+\newcommand{\SecFlowBlockchainToAiSafetyTwoZeroTwoFivePct}{0.3}
6201+\newcommand{\SecFlowBlockchainToLlmTwoZeroTwoFivePct}{0.3}
6202+\newcommand{\SecFlowBlockchainToAiAgentsTwoZeroTwoFivePct}{0.2}
6203+\newcommand{\SecFlowBlockchainToGenerativeAiTwoZeroTwoFivePct}{0.1}
6204+\newcommand{\SecFlowComputerNetworksToCybersecurityTwoZeroTwoFivePct}{6.8}
6205+\newcommand{\SecFlowComputerNetworksToBlockchainTwoZeroTwoFivePct}{0.8}
6206+\newcommand{\SecFlowComputerNetworksToAppliedCryptographyTwoZeroTwoFivePct}{0.7}
6207+\newcommand{\SecFlowComputerNetworksToPrivacyPreservingComputationTwoZeroTwoFivePct}{0.5}
6208+\newcommand{\SecFlowComputerNetworksToFederatedLearningTwoZeroTwoFivePct}{0.2}
6209+\newcommand{\SecFlowCybersecurityToCybersecurityTwoZeroTwoFivePct}{47.5}
6210+\newcommand{\SecFlowCybersecurityToComputerNetworksTwoZeroTwoFivePct}{13.6}
6211+\newcommand{\SecFlowCybersecurityToComputerVisionTwoZeroTwoFivePct}{7.6}
6212+\newcommand{\SecFlowCybersecurityToAppliedCryptographyTwoZeroTwoFivePct}{4.5}
6213+\newcommand{\SecFlowCybersecurityToBlockchainTwoZeroTwoFivePct}{2.7}
6214+\newcommand{\SecFlowCybersecurityToSoftwareEngineeringTwoZeroTwoFivePct}{2.6}
6215+\newcommand{\SecFlowCybersecurityToMlTheoryTwoZeroTwoFivePct}{2.4}
6216+\newcommand{\SecFlowCybersecurityToEdgeComputingTwoZeroTwoFivePct}{2.3}
6217+\newcommand{\SecFlowCybersecurityToPrivacyPreservingComputationTwoZeroTwoFivePct}{1.8}
6218+\newcommand{\SecFlowCybersecurityToHciTwoZeroTwoFivePct}{1.1}
6219+\newcommand{\SecFlowCybersecurityToLlmTwoZeroTwoFivePct}{0.7}
6220+\newcommand{\SecFlowCybersecurityToDistributedComputingTwoZeroTwoFivePct}{0.6}
6221+\newcommand{\SecFlowCybersecurityToDatabaseSystemsTwoZeroTwoFivePct}{0.5}
6222+\newcommand{\SecFlowCybersecurityToAiSafetyTwoZeroTwoFivePct}{0.5}
6223+\newcommand{\SecFlowCybersecurityToQuantumComputingTwoZeroTwoFivePct}{0.5}
6224+\newcommand{\SecFlowCybersecurityToFederatedLearningTwoZeroTwoFivePct}{0.3}
6225+\newcommand{\SecFlowCybersecurityToGenerativeAiTwoZeroTwoFivePct}{0.3}
6226+\newcommand{\SecFlowCybersecurityToAiAgentsTwoZeroTwoFivePct}{0.3}
6227+\newcommand{\SecFlowDistributedComputingToBlockchainTwoZeroTwoFivePct}{3.7}
6228+\newcommand{\SecFlowDistributedComputingToCybersecurityTwoZeroTwoFivePct}{2.9}
6229+\newcommand{\SecFlowDistributedComputingToAppliedCryptographyTwoZeroTwoFivePct}{1.2}
6230+\newcommand{\SecFlowDistributedComputingToPrivacyPreservingComputationTwoZeroTwoFivePct}{0.6}
6231+\newcommand{\SecFlowDistributedComputingToFederatedLearningTwoZeroTwoFivePct}{0.3}
6232+\newcommand{\SecFlowEdgeComputingToCybersecurityTwoZeroTwoFivePct}{6.3}
6233+\newcommand{\SecFlowEdgeComputingToBlockchainTwoZeroTwoFivePct}{3.9}
6234+\newcommand{\SecFlowEdgeComputingToPrivacyPreservingComputationTwoZeroTwoFivePct}{2.0}
6235+\newcommand{\SecFlowEdgeComputingToAppliedCryptographyTwoZeroTwoFivePct}{1.6}
6236+\newcommand{\SecFlowEdgeComputingToFederatedLearningTwoZeroTwoFivePct}{1.2}
6237+\newcommand{\SecFlowFederatedLearningToFederatedLearningTwoZeroTwoFivePct}{16.0}
6238+\newcommand{\SecFlowFederatedLearningToPrivacyPreservingComputationTwoZeroTwoFivePct}{15.5}
6239+\newcommand{\SecFlowFederatedLearningToMlTheoryTwoZeroTwoFivePct}{13.0}
6240+\newcommand{\SecFlowFederatedLearningToComputerNetworksTwoZeroTwoFivePct}{10.4}
6241+\newcommand{\SecFlowFederatedLearningToComputerVisionTwoZeroTwoFivePct}{10.0}
6242+\newcommand{\SecFlowFederatedLearningToCybersecurityTwoZeroTwoFivePct}{6.8}
6243+\newcommand{\SecFlowFederatedLearningToEdgeComputingTwoZeroTwoFivePct}{6.7}
6244+\newcommand{\SecFlowFederatedLearningToAppliedCryptographyTwoZeroTwoFivePct}{2.1}
6245+\newcommand{\SecFlowFederatedLearningToBlockchainTwoZeroTwoFivePct}{2.0}
6246+\newcommand{\SecFlowFederatedLearningToDistributedComputingTwoZeroTwoFivePct}{1.8}
6247+\newcommand{\SecFlowFederatedLearningToAiSafetyTwoZeroTwoFivePct}{0.9}
6248+\newcommand{\SecFlowFederatedLearningToDatabaseSystemsTwoZeroTwoFivePct}{0.6}
6249+\newcommand{\SecFlowFederatedLearningToHciTwoZeroTwoFivePct}{0.5}
6250+\newcommand{\SecFlowFederatedLearningToLlmTwoZeroTwoFivePct}{0.5}
6251+\newcommand{\SecFlowFederatedLearningToGenerativeAiTwoZeroTwoFivePct}{0.4}
6252+\newcommand{\SecFlowFederatedLearningToSoftwareEngineeringTwoZeroTwoFivePct}{0.3}
6253+\newcommand{\SecFlowFederatedLearningToAiAgentsTwoZeroTwoFivePct}{0.1}
6254+\newcommand{\SecFlowFederatedLearningToQuantumComputingTwoZeroTwoFivePct}{0.1}
6255+\newcommand{\SecFlowGenerativeAiToCybersecurityTwoZeroTwoFivePct}{1.8}
6256+\newcommand{\SecFlowGenerativeAiToPrivacyPreservingComputationTwoZeroTwoFivePct}{1.0}
6257+\newcommand{\SecFlowGenerativeAiToBlockchainTwoZeroTwoFivePct}{0.3}
6258+\newcommand{\SecFlowGenerativeAiToAppliedCryptographyTwoZeroTwoFivePct}{0.2}
6259+\newcommand{\SecFlowGenerativeAiToFederatedLearningTwoZeroTwoFivePct}{0.1}
6260+\newcommand{\SecFlowLlmToCybersecurityTwoZeroTwoFivePct}{1.6}
6261+\newcommand{\SecFlowLlmToPrivacyPreservingComputationTwoZeroTwoFivePct}{0.5}
6262+\newcommand{\SecFlowLlmToBlockchainTwoZeroTwoFivePct}{0.2}
6263+\newcommand{\SecFlowLlmToFederatedLearningTwoZeroTwoFivePct}{0.1}
6264+\newcommand{\SecFlowLlmToAppliedCryptographyTwoZeroTwoFivePct}{0.1}
6265+\newcommand{\SecFlowPrivacyPreservingComputationToPrivacyPreservingComputationTwoZeroTwoFivePct}{24.4}
6266+\newcommand{\SecFlowPrivacyPreservingComputationToCybersecurityTwoZeroTwoFivePct}{11.2}
6267+\newcommand{\SecFlowPrivacyPreservingComputationToComputerVisionTwoZeroTwoFivePct}{8.4}
6268+\newcommand{\SecFlowPrivacyPreservingComputationToMlTheoryTwoZeroTwoFivePct}{7.2}
6269+\newcommand{\SecFlowPrivacyPreservingComputationToComputerNetworksTwoZeroTwoFivePct}{7.1}
6270+\newcommand{\SecFlowPrivacyPreservingComputationToFederatedLearningTwoZeroTwoFivePct}{6.1}
6271+\newcommand{\SecFlowPrivacyPreservingComputationToAppliedCryptographyTwoZeroTwoFivePct}{4.1}
6272+\newcommand{\SecFlowPrivacyPreservingComputationToEdgeComputingTwoZeroTwoFivePct}{3.8}
6273+\newcommand{\SecFlowPrivacyPreservingComputationToHciTwoZeroTwoFivePct}{2.8}
6274+\newcommand{\SecFlowPrivacyPreservingComputationToBlockchainTwoZeroTwoFivePct}{2.3}
6275+\newcommand{\SecFlowPrivacyPreservingComputationToDatabaseSystemsTwoZeroTwoFivePct}{1.9}
6276+\newcommand{\SecFlowPrivacyPreservingComputationToLlmTwoZeroTwoFivePct}{1.6}
6277+\newcommand{\SecFlowPrivacyPreservingComputationToAiSafetyTwoZeroTwoFivePct}{1.6}
6278+\newcommand{\SecFlowPrivacyPreservingComputationToDistributedComputingTwoZeroTwoFivePct}{1.2}
6279+\newcommand{\SecFlowPrivacyPreservingComputationToGenerativeAiTwoZeroTwoFivePct}{1.0}
6280+\newcommand{\SecFlowPrivacyPreservingComputationToSoftwareEngineeringTwoZeroTwoFivePct}{0.7}
6281+\newcommand{\SecFlowPrivacyPreservingComputationToAiAgentsTwoZeroTwoFivePct}{0.4}
6282+\newcommand{\SecFlowPrivacyPreservingComputationToQuantumComputingTwoZeroTwoFivePct}{0.2}
6283+\newcommand{\SecFlowAiAgentsToCybersecurityTwoZeroTwoSixPct}{2.7}
6284+\newcommand{\SecFlowAiAgentsToPrivacyPreservingComputationTwoZeroTwoSixPct}{0.7}
6285+\newcommand{\SecFlowAiAgentsToBlockchainTwoZeroTwoSixPct}{0.4}
6286+\newcommand{\SecFlowAiAgentsToAppliedCryptographyTwoZeroTwoSixPct}{0.3}
6287+\newcommand{\SecFlowAiAgentsToFederatedLearningTwoZeroTwoSixPct}{0.1}
6288+\newcommand{\SecFlowAiSafetyToCybersecurityTwoZeroTwoSixPct}{4.0}
6289+\newcommand{\SecFlowAiSafetyToPrivacyPreservingComputationTwoZeroTwoSixPct}{1.9}
6290+\newcommand{\SecFlowAiSafetyToBlockchainTwoZeroTwoSixPct}{0.6}
6291+\newcommand{\SecFlowAiSafetyToFederatedLearningTwoZeroTwoSixPct}{0.3}
6292+\newcommand{\SecFlowAiSafetyToAppliedCryptographyTwoZeroTwoSixPct}{0.2}
6293+\newcommand{\SecFlowAppliedCryptographyToAppliedCryptographyTwoZeroTwoSixPct}{50.7}
6294+\newcommand{\SecFlowAppliedCryptographyToCybersecurityTwoZeroTwoSixPct}{14.7}
6295+\newcommand{\SecFlowAppliedCryptographyToComputerVisionTwoZeroTwoSixPct}{6.2}
6296+\newcommand{\SecFlowAppliedCryptographyToComputerNetworksTwoZeroTwoSixPct}{5.4}
6297+\newcommand{\SecFlowAppliedCryptographyToBlockchainTwoZeroTwoSixPct}{3.8}
6298+\newcommand{\SecFlowAppliedCryptographyToQuantumComputingTwoZeroTwoSixPct}{3.2}
6299+\newcommand{\SecFlowAppliedCryptographyToPrivacyPreservingComputationTwoZeroTwoSixPct}{2.3}
6300+\newcommand{\SecFlowAppliedCryptographyToEdgeComputingTwoZeroTwoSixPct}{1.9}
6301+\newcommand{\SecFlowAppliedCryptographyToDistributedComputingTwoZeroTwoSixPct}{1.0}
6302+\newcommand{\SecFlowAppliedCryptographyToMlTheoryTwoZeroTwoSixPct}{0.7}
6303+\newcommand{\SecFlowAppliedCryptographyToFederatedLearningTwoZeroTwoSixPct}{0.4}
6304+\newcommand{\SecFlowAppliedCryptographyToDatabaseSystemsTwoZeroTwoSixPct}{0.4}
6305+\newcommand{\SecFlowAppliedCryptographyToSoftwareEngineeringTwoZeroTwoSixPct}{0.3}
6306+\newcommand{\SecFlowAppliedCryptographyToHciTwoZeroTwoSixPct}{0.2}
6307+\newcommand{\SecFlowAppliedCryptographyToLlmTwoZeroTwoSixPct}{0.1}
6308+\newcommand{\SecFlowAppliedCryptographyToAiSafetyTwoZeroTwoSixPct}{0.1}
6309+\newcommand{\SecFlowAppliedCryptographyToAiAgentsTwoZeroTwoSixPct}{0.1}
6310+\newcommand{\SecFlowAppliedCryptographyToGenerativeAiTwoZeroTwoSixPct}{0.1}
6311+\newcommand{\SecFlowBlockchainToBlockchainTwoZeroTwoSixPct}{54.1}
6312+\newcommand{\SecFlowBlockchainToCybersecurityTwoZeroTwoSixPct}{9.8}
6313+\newcommand{\SecFlowBlockchainToComputerNetworksTwoZeroTwoSixPct}{6.7}
6314+\newcommand{\SecFlowBlockchainToEdgeComputingTwoZeroTwoSixPct}{5.0}
6315+\newcommand{\SecFlowBlockchainToAppliedCryptographyTwoZeroTwoSixPct}{4.3}
6316+\newcommand{\SecFlowBlockchainToDistributedComputingTwoZeroTwoSixPct}{3.7}
6317+\newcommand{\SecFlowBlockchainToSoftwareEngineeringTwoZeroTwoSixPct}{2.5}
6318+\newcommand{\SecFlowBlockchainToPrivacyPreservingComputationTwoZeroTwoSixPct}{1.8}
6319+\newcommand{\SecFlowBlockchainToComputerVisionTwoZeroTwoSixPct}{1.3}
6320+\newcommand{\SecFlowBlockchainToMlTheoryTwoZeroTwoSixPct}{1.2}
6321+\newcommand{\SecFlowBlockchainToDatabaseSystemsTwoZeroTwoSixPct}{1.0}
6322+\newcommand{\SecFlowBlockchainToHciTwoZeroTwoSixPct}{0.9}
6323+\newcommand{\SecFlowBlockchainToFederatedLearningTwoZeroTwoSixPct}{0.7}
6324+\newcommand{\SecFlowBlockchainToQuantumComputingTwoZeroTwoSixPct}{0.4}
6325+\newcommand{\SecFlowBlockchainToAiSafetyTwoZeroTwoSixPct}{0.4}
6326+\newcommand{\SecFlowBlockchainToLlmTwoZeroTwoSixPct}{0.3}
6327+\newcommand{\SecFlowBlockchainToAiAgentsTwoZeroTwoSixPct}{0.2}
6328+\newcommand{\SecFlowBlockchainToGenerativeAiTwoZeroTwoSixPct}{0.1}
6329+\newcommand{\SecFlowComputerNetworksToCybersecurityTwoZeroTwoSixPct}{7.3}
6330+\newcommand{\SecFlowComputerNetworksToAppliedCryptographyTwoZeroTwoSixPct}{0.7}
6331+\newcommand{\SecFlowComputerNetworksToBlockchainTwoZeroTwoSixPct}{0.7}
6332+\newcommand{\SecFlowComputerNetworksToPrivacyPreservingComputationTwoZeroTwoSixPct}{0.6}
6333+\newcommand{\SecFlowComputerNetworksToFederatedLearningTwoZeroTwoSixPct}{0.3}
6334+\newcommand{\SecFlowCybersecurityToCybersecurityTwoZeroTwoSixPct}{49.0}
6335+\newcommand{\SecFlowCybersecurityToComputerNetworksTwoZeroTwoSixPct}{13.6}
6336+\newcommand{\SecFlowCybersecurityToComputerVisionTwoZeroTwoSixPct}{7.4}
6337+\newcommand{\SecFlowCybersecurityToAppliedCryptographyTwoZeroTwoSixPct}{3.9}
6338+\newcommand{\SecFlowCybersecurityToSoftwareEngineeringTwoZeroTwoSixPct}{2.8}
6339+\newcommand{\SecFlowCybersecurityToMlTheoryTwoZeroTwoSixPct}{2.4}
6340+\newcommand{\SecFlowCybersecurityToBlockchainTwoZeroTwoSixPct}{2.3}
6341+\newcommand{\SecFlowCybersecurityToEdgeComputingTwoZeroTwoSixPct}{2.1}
6342+\newcommand{\SecFlowCybersecurityToPrivacyPreservingComputationTwoZeroTwoSixPct}{1.6}
6343+\newcommand{\SecFlowCybersecurityToLlmTwoZeroTwoSixPct}{1.1}
6344+\newcommand{\SecFlowCybersecurityToHciTwoZeroTwoSixPct}{1.0}
6345+\newcommand{\SecFlowCybersecurityToQuantumComputingTwoZeroTwoSixPct}{0.6}
6346+\newcommand{\SecFlowCybersecurityToAiSafetyTwoZeroTwoSixPct}{0.6}
6347+\newcommand{\SecFlowCybersecurityToDistributedComputingTwoZeroTwoSixPct}{0.5}
6348+\newcommand{\SecFlowCybersecurityToDatabaseSystemsTwoZeroTwoSixPct}{0.5}
6349+\newcommand{\SecFlowCybersecurityToGenerativeAiTwoZeroTwoSixPct}{0.4}
6350+\newcommand{\SecFlowCybersecurityToFederatedLearningTwoZeroTwoSixPct}{0.3}
6351+\newcommand{\SecFlowCybersecurityToAiAgentsTwoZeroTwoSixPct}{0.3}
6352+\newcommand{\SecFlowDistributedComputingToBlockchainTwoZeroTwoSixPct}{3.5}
6353+\newcommand{\SecFlowDistributedComputingToCybersecurityTwoZeroTwoSixPct}{3.0}
6354+\newcommand{\SecFlowDistributedComputingToAppliedCryptographyTwoZeroTwoSixPct}{1.2}
6355+\newcommand{\SecFlowDistributedComputingToPrivacyPreservingComputationTwoZeroTwoSixPct}{0.5}
6356+\newcommand{\SecFlowDistributedComputingToFederatedLearningTwoZeroTwoSixPct}{0.3}
6357+\newcommand{\SecFlowEdgeComputingToCybersecurityTwoZeroTwoSixPct}{6.6}
6358+\newcommand{\SecFlowEdgeComputingToBlockchainTwoZeroTwoSixPct}{3.0}
6359+\newcommand{\SecFlowEdgeComputingToPrivacyPreservingComputationTwoZeroTwoSixPct}{2.3}
6360+\newcommand{\SecFlowEdgeComputingToAppliedCryptographyTwoZeroTwoSixPct}{1.7}
6361+\newcommand{\SecFlowEdgeComputingToFederatedLearningTwoZeroTwoSixPct}{1.3}
6362+\newcommand{\SecFlowFederatedLearningToPrivacyPreservingComputationTwoZeroTwoSixPct}{15.8}
6363+\newcommand{\SecFlowFederatedLearningToFederatedLearningTwoZeroTwoSixPct}{15.4}
6364+\newcommand{\SecFlowFederatedLearningToComputerNetworksTwoZeroTwoSixPct}{12.1}
6365+\newcommand{\SecFlowFederatedLearningToMlTheoryTwoZeroTwoSixPct}{10.9}
6366+\newcommand{\SecFlowFederatedLearningToComputerVisionTwoZeroTwoSixPct}{9.0}
6367+\newcommand{\SecFlowFederatedLearningToEdgeComputingTwoZeroTwoSixPct}{7.6}
6368+\newcommand{\SecFlowFederatedLearningToCybersecurityTwoZeroTwoSixPct}{7.2}
6369+\newcommand{\SecFlowFederatedLearningToAppliedCryptographyTwoZeroTwoSixPct}{2.1}
6370+\newcommand{\SecFlowFederatedLearningToBlockchainTwoZeroTwoSixPct}{2.1}
6371+\newcommand{\SecFlowFederatedLearningToDistributedComputingTwoZeroTwoSixPct}{1.7}
6372+\newcommand{\SecFlowFederatedLearningToLlmTwoZeroTwoSixPct}{1.1}
6373+\newcommand{\SecFlowFederatedLearningToAiSafetyTwoZeroTwoSixPct}{0.7}
6374+\newcommand{\SecFlowFederatedLearningToHciTwoZeroTwoSixPct}{0.6}
6375+\newcommand{\SecFlowFederatedLearningToDatabaseSystemsTwoZeroTwoSixPct}{0.5}
6376+\newcommand{\SecFlowFederatedLearningToSoftwareEngineeringTwoZeroTwoSixPct}{0.4}
6377+\newcommand{\SecFlowFederatedLearningToGenerativeAiTwoZeroTwoSixPct}{0.3}
6378+\newcommand{\SecFlowFederatedLearningToAiAgentsTwoZeroTwoSixPct}{0.1}
6379+\newcommand{\SecFlowFederatedLearningToQuantumComputingTwoZeroTwoSixPct}{0.1}
6380+\newcommand{\SecFlowGenerativeAiToCybersecurityTwoZeroTwoSixPct}{1.7}
6381+\newcommand{\SecFlowGenerativeAiToPrivacyPreservingComputationTwoZeroTwoSixPct}{0.9}
6382+\newcommand{\SecFlowGenerativeAiToBlockchainTwoZeroTwoSixPct}{0.2}
6383+\newcommand{\SecFlowGenerativeAiToFederatedLearningTwoZeroTwoSixPct}{0.1}
6384+\newcommand{\SecFlowGenerativeAiToAppliedCryptographyTwoZeroTwoSixPct}{0.1}
6385+\newcommand{\SecFlowLlmToCybersecurityTwoZeroTwoSixPct}{2.0}
6386+\newcommand{\SecFlowLlmToPrivacyPreservingComputationTwoZeroTwoSixPct}{0.6}
6387+\newcommand{\SecFlowLlmToBlockchainTwoZeroTwoSixPct}{0.2}
6388+\newcommand{\SecFlowLlmToFederatedLearningTwoZeroTwoSixPct}{0.1}
6389+\newcommand{\SecFlowLlmToAppliedCryptographyTwoZeroTwoSixPct}{0.1}
6390+\newcommand{\SecFlowPrivacyPreservingComputationToPrivacyPreservingComputationTwoZeroTwoSixPct}{25.7}
6391+\newcommand{\SecFlowPrivacyPreservingComputationToCybersecurityTwoZeroTwoSixPct}{12.0}
6392+\newcommand{\SecFlowPrivacyPreservingComputationToComputerVisionTwoZeroTwoSixPct}{7.5}
6393+\newcommand{\SecFlowPrivacyPreservingComputationToMlTheoryTwoZeroTwoSixPct}{6.5}
6394+\newcommand{\SecFlowPrivacyPreservingComputationToComputerNetworksTwoZeroTwoSixPct}{6.2}
6395+\newcommand{\SecFlowPrivacyPreservingComputationToFederatedLearningTwoZeroTwoSixPct}{5.7}
6396+\newcommand{\SecFlowPrivacyPreservingComputationToAppliedCryptographyTwoZeroTwoSixPct}{4.7}
6397+\newcommand{\SecFlowPrivacyPreservingComputationToEdgeComputingTwoZeroTwoSixPct}{3.6}
6398+\newcommand{\SecFlowPrivacyPreservingComputationToHciTwoZeroTwoSixPct}{2.5}
6399+\newcommand{\SecFlowPrivacyPreservingComputationToLlmTwoZeroTwoSixPct}{2.2}
6400+\newcommand{\SecFlowPrivacyPreservingComputationToBlockchainTwoZeroTwoSixPct}{2.1}
6401+\newcommand{\SecFlowPrivacyPreservingComputationToDatabaseSystemsTwoZeroTwoSixPct}{1.8}
6402+\newcommand{\SecFlowPrivacyPreservingComputationToAiSafetyTwoZeroTwoSixPct}{1.6}
6403+\newcommand{\SecFlowPrivacyPreservingComputationToDistributedComputingTwoZeroTwoSixPct}{1.0}
6404+\newcommand{\SecFlowPrivacyPreservingComputationToGenerativeAiTwoZeroTwoSixPct}{1.0}
6405+\newcommand{\SecFlowPrivacyPreservingComputationToSoftwareEngineeringTwoZeroTwoSixPct}{0.9}
6406+\newcommand{\SecFlowPrivacyPreservingComputationToAiAgentsTwoZeroTwoSixPct}{0.6}
6407+\newcommand{\SecFlowPrivacyPreservingComputationToQuantumComputingTwoZeroTwoSixPct}{0.2}
6408+\newcommand{\SecFlowAppliedCryptographyTotalTwoZeroOneNine}{17{,}783}
6409+\newcommand{\SecFlowAppliedCryptographyTotalTwoZeroTwoTwo}{71{,}899}
6410+\newcommand{\SecFlowAppliedCryptographyTotalTwoZeroTwoFive}{124{,}901}
6411+\newcommand{\SecFlowAppliedCryptographyTotalTwoZeroTwoSix}{83{,}865}
6412+\newcommand{\SecFlowBlockchainTotalTwoZeroOneNine}{19{,}877}
6413+\newcommand{\SecFlowBlockchainTotalTwoZeroTwoTwo}{110{,}937}
6414+\newcommand{\SecFlowBlockchainTotalTwoZeroTwoFive}{154{,}305}
6415+\newcommand{\SecFlowBlockchainTotalTwoZeroTwoSix}{86{,}875}
6416+\newcommand{\SecFlowCybersecurityTotalTwoZeroOneNine}{47{,}051}
6417+\newcommand{\SecFlowCybersecurityTotalTwoZeroTwoTwo}{226{,}033}
6418+\newcommand{\SecFlowCybersecurityTotalTwoZeroTwoFive}{458{,}928}
6419+\newcommand{\SecFlowCybersecurityTotalTwoZeroTwoSix}{289{,}376}
6420+\newcommand{\SecFlowFederatedLearningTotalTwoZeroOneNine}{279}
6421+\newcommand{\SecFlowFederatedLearningTotalTwoZeroTwoTwo}{10{,}464}
6422+\newcommand{\SecFlowFederatedLearningTotalTwoZeroTwoFive}{32{,}933}
6423+\newcommand{\SecFlowFederatedLearningTotalTwoZeroTwoSix}{29{,}919}
6424+\newcommand{\SecFlowPrivacyPreservingComputationTotalTwoZeroOneNine}{6{,}686}
6425+\newcommand{\SecFlowPrivacyPreservingComputationTotalTwoZeroTwoTwo}{37{,}649}
6426+\newcommand{\SecFlowPrivacyPreservingComputationTotalTwoZeroTwoFive}{82{,}044}
6427+\newcommand{\SecFlowPrivacyPreservingComputationTotalTwoZeroTwoSix}{52{,}767}
6428+\newcommand{\SecCoocAppliedCryptographyCybersecurityTwoZeroOneEight}{16{,}451}
6429+\newcommand{\SecCoocAppliedCryptographyCybersecurityTwoZeroOneEightPct}{55.4}
6430+\newcommand{\SecCoocCybersecurityPrivacyPreservingComputationTwoZeroOneEight}{6{,}716}
6431+\newcommand{\SecCoocCybersecurityPrivacyPreservingComputationTwoZeroOneEightPct}{9.9}
6432+\newcommand{\SecCoocAppliedCryptographyPrivacyPreservingComputationTwoZeroOneEight}{4{,}182}
6433+\newcommand{\SecCoocAppliedCryptographyPrivacyPreservingComputationTwoZeroOneEightPct}{14.1}
6434+\newcommand{\SecCoocBlockchainCybersecurityTwoZeroOneEight}{3{,}298}
6435+\newcommand{\SecCoocBlockchainCybersecurityTwoZeroOneEightPct}{30.3}
6436+\newcommand{\SecCoocAppliedCryptographyBlockchainTwoZeroOneEight}{2{,}668}
6437+\newcommand{\SecCoocAppliedCryptographyBlockchainTwoZeroOneEightPct}{9.0}
6438+\newcommand{\SecCoocBlockchainPrivacyPreservingComputationTwoZeroOneEight}{1{,}015}
6439+\newcommand{\SecCoocBlockchainPrivacyPreservingComputationTwoZeroOneEightPct}{9.3}
6440+\newcommand{\SecCoocCybersecurityAiSafetyTwoZeroOneEight}{929}
6441+\newcommand{\SecCoocCybersecurityAiSafetyTwoZeroOneEightPct}{1.4}
6442+\newcommand{\SecCoocCybersecurityAiAgentsTwoZeroOneEight}{571}
6443+\newcommand{\SecCoocCybersecurityAiAgentsTwoZeroOneEightPct}{0.8}
6444+\newcommand{\SecCoocCybersecurityLlmTwoZeroOneEight}{536}
6445+\newcommand{\SecCoocCybersecurityLlmTwoZeroOneEightPct}{0.8}
6446+\newcommand{\SecCoocCybersecurityGenerativeAiTwoZeroOneEight}{492}
6447+\newcommand{\SecCoocCybersecurityGenerativeAiTwoZeroOneEightPct}{0.7}
6448+\newcommand{\SecCoocFederatedLearningPrivacyPreservingComputationTwoZeroOneEight}{416}
6449+\newcommand{\SecCoocFederatedLearningPrivacyPreservingComputationTwoZeroOneEightPct}{40.0}
6450+\newcommand{\SecCoocPrivacyPreservingComputationAiSafetyTwoZeroOneEight}{368}
6451+\newcommand{\SecCoocPrivacyPreservingComputationAiSafetyTwoZeroOneEightPct}{3.2}
6452+\newcommand{\SecCoocCybersecurityFederatedLearningTwoZeroOneEight}{320}
6453+\newcommand{\SecCoocCybersecurityFederatedLearningTwoZeroOneEightPct}{0.5}
6454+\newcommand{\SecCoocBlockchainLlmTwoZeroOneEight}{263}
6455+\newcommand{\SecCoocBlockchainLlmTwoZeroOneEightPct}{2.4}
6456+\newcommand{\SecCoocCybersecurityFoundationModelsTwoZeroOneEight}{257}
6457+\newcommand{\SecCoocCybersecurityFoundationModelsTwoZeroOneEightPct}{0.4}
6458+\newcommand{\SecCoocAppliedCryptographyFederatedLearningTwoZeroOneEight}{181}
6459+\newcommand{\SecCoocAppliedCryptographyFederatedLearningTwoZeroOneEightPct}{0.6}
6460+\newcommand{\SecCoocBlockchainAiAgentsTwoZeroOneEight}{155}
6461+\newcommand{\SecCoocBlockchainAiAgentsTwoZeroOneEightPct}{1.4}
6462+\newcommand{\SecCoocAppliedCryptographyAiAgentsTwoZeroOneEight}{123}
6463+\newcommand{\SecCoocAppliedCryptographyAiAgentsTwoZeroOneEightPct}{0.4}
6464+\newcommand{\SecCoocPrivacyPreservingComputationGenerativeAiTwoZeroOneEight}{88}
6465+\newcommand{\SecCoocPrivacyPreservingComputationGenerativeAiTwoZeroOneEightPct}{0.8}
6466+\newcommand{\SecCoocPrivacyPreservingComputationAiAgentsTwoZeroOneEight}{83}
6467+\newcommand{\SecCoocPrivacyPreservingComputationAiAgentsTwoZeroOneEightPct}{0.7}
6468+\newcommand{\SecCoocFederatedLearningAiSafetyTwoZeroOneEight}{81}
6469+\newcommand{\SecCoocFederatedLearningAiSafetyTwoZeroOneEightPct}{7.8}
6470+\newcommand{\SecCoocAppliedCryptographyAiSafetyTwoZeroOneEight}{76}
6471+\newcommand{\SecCoocAppliedCryptographyAiSafetyTwoZeroOneEightPct}{0.3}
6472+\newcommand{\SecCoocAppliedCryptographyLlmTwoZeroOneEight}{66}
6473+\newcommand{\SecCoocAppliedCryptographyLlmTwoZeroOneEightPct}{0.2}
6474+\newcommand{\SecCoocPrivacyPreservingComputationLlmTwoZeroOneEight}{58}
6475+\newcommand{\SecCoocPrivacyPreservingComputationLlmTwoZeroOneEightPct}{0.5}
6476+\newcommand{\SecCoocAppliedCryptographyGenerativeAiTwoZeroOneEight}{47}
6477+\newcommand{\SecCoocAppliedCryptographyGenerativeAiTwoZeroOneEightPct}{0.2}
6478+\newcommand{\SecCoocPrivacyPreservingComputationFoundationModelsTwoZeroOneEight}{42}
6479+\newcommand{\SecCoocPrivacyPreservingComputationFoundationModelsTwoZeroOneEightPct}{0.4}
6480+\newcommand{\SecCoocBlockchainFederatedLearningTwoZeroOneEight}{38}
6481+\newcommand{\SecCoocBlockchainFederatedLearningTwoZeroOneEightPct}{0.3}
6482+\newcommand{\SecCoocAppliedCryptographyFoundationModelsTwoZeroOneEight}{33}
6483+\newcommand{\SecCoocAppliedCryptographyFoundationModelsTwoZeroOneEightPct}{0.1}
6484+\newcommand{\SecCoocBlockchainAiSafetyTwoZeroOneEight}{29}
6485+\newcommand{\SecCoocBlockchainAiSafetyTwoZeroOneEightPct}{0.3}
6486+\newcommand{\SecCoocFederatedLearningFoundationModelsTwoZeroOneEight}{28}
6487+\newcommand{\SecCoocFederatedLearningFoundationModelsTwoZeroOneEightPct}{2.7}
6488+\newcommand{\SecCoocFederatedLearningAiAgentsTwoZeroOneEight}{22}
6489+\newcommand{\SecCoocFederatedLearningAiAgentsTwoZeroOneEightPct}{2.1}
6490+\newcommand{\SecCoocFederatedLearningGenerativeAiTwoZeroOneEight}{21}
6491+\newcommand{\SecCoocFederatedLearningGenerativeAiTwoZeroOneEightPct}{2.0}
6492+\newcommand{\SecCoocFederatedLearningLlmTwoZeroOneEight}{13}
6493+\newcommand{\SecCoocFederatedLearningLlmTwoZeroOneEightPct}{1.2}
6494+\newcommand{\SecCoocBlockchainFoundationModelsTwoZeroOneEight}{10}
6495+\newcommand{\SecCoocBlockchainFoundationModelsTwoZeroOneEightPct}{0.1}
6496+\newcommand{\SecCoocBlockchainGenerativeAiTwoZeroOneEight}{9}
6497+\newcommand{\SecCoocBlockchainGenerativeAiTwoZeroOneEightPct}{0.1}
6498+\newcommand{\SecCoocAppliedCryptographyCybersecurityTwoZeroOneNine}{17{,}684}
6499+\newcommand{\SecCoocAppliedCryptographyCybersecurityTwoZeroOneNinePct}{56.2}
6500+\newcommand{\SecCoocCybersecurityPrivacyPreservingComputationTwoZeroOneNine}{7{,}395}
6501+\newcommand{\SecCoocCybersecurityPrivacyPreservingComputationTwoZeroOneNinePct}{10.1}
6502+\newcommand{\SecCoocBlockchainCybersecurityTwoZeroOneNine}{4{,}916}
6503+\newcommand{\SecCoocBlockchainCybersecurityTwoZeroOneNinePct}{32.1}
6504+\newcommand{\SecCoocAppliedCryptographyPrivacyPreservingComputationTwoZeroOneNine}{4{,}545}
6505+\newcommand{\SecCoocAppliedCryptographyPrivacyPreservingComputationTwoZeroOneNinePct}{14.5}
6506+\newcommand{\SecCoocAppliedCryptographyBlockchainTwoZeroOneNine}{3{,}813}
6507+\newcommand{\SecCoocAppliedCryptographyBlockchainTwoZeroOneNinePct}{12.1}
6508+\newcommand{\SecCoocCybersecurityAiSafetyTwoZeroOneNine}{1{,}726}
6509+\newcommand{\SecCoocCybersecurityAiSafetyTwoZeroOneNinePct}{2.4}
6510+\newcommand{\SecCoocBlockchainPrivacyPreservingComputationTwoZeroOneNine}{1{,}505}
6511+\newcommand{\SecCoocBlockchainPrivacyPreservingComputationTwoZeroOneNinePct}{9.8}
6512+\newcommand{\SecCoocCybersecurityGenerativeAiTwoZeroOneNine}{905}
6513+\newcommand{\SecCoocCybersecurityGenerativeAiTwoZeroOneNinePct}{1.2}
6514+\newcommand{\SecCoocFederatedLearningPrivacyPreservingComputationTwoZeroOneNine}{851}
6515+\newcommand{\SecCoocFederatedLearningPrivacyPreservingComputationTwoZeroOneNinePct}{53.9}
6516+\newcommand{\SecCoocCybersecurityLlmTwoZeroOneNine}{735}
6517+\newcommand{\SecCoocCybersecurityLlmTwoZeroOneNinePct}{1.0}
6518+\newcommand{\SecCoocCybersecurityAiAgentsTwoZeroOneNine}{621}
6519+\newcommand{\SecCoocCybersecurityAiAgentsTwoZeroOneNinePct}{0.8}
6520+\newcommand{\SecCoocPrivacyPreservingComputationAiSafetyTwoZeroOneNine}{614}
6521+\newcommand{\SecCoocPrivacyPreservingComputationAiSafetyTwoZeroOneNinePct}{4.7}
6522+\newcommand{\SecCoocCybersecurityFederatedLearningTwoZeroOneNine}{562}
6523+\newcommand{\SecCoocCybersecurityFederatedLearningTwoZeroOneNinePct}{0.8}
6524+\newcommand{\SecCoocCybersecurityFoundationModelsTwoZeroOneNine}{533}
6525+\newcommand{\SecCoocCybersecurityFoundationModelsTwoZeroOneNinePct}{0.7}
6526+\newcommand{\SecCoocAppliedCryptographyFederatedLearningTwoZeroOneNine}{312}
6527+\newcommand{\SecCoocAppliedCryptographyFederatedLearningTwoZeroOneNinePct}{1.0}
6528+\newcommand{\SecCoocBlockchainLlmTwoZeroOneNine}{272}
6529+\newcommand{\SecCoocBlockchainLlmTwoZeroOneNinePct}{1.8}
6530+\newcommand{\SecCoocBlockchainAiAgentsTwoZeroOneNine}{215}
6531+\newcommand{\SecCoocBlockchainAiAgentsTwoZeroOneNinePct}{1.4}
6532+\newcommand{\SecCoocPrivacyPreservingComputationGenerativeAiTwoZeroOneNine}{191}
6533+\newcommand{\SecCoocPrivacyPreservingComputationGenerativeAiTwoZeroOneNinePct}{1.5}
6534+\newcommand{\SecCoocFederatedLearningAiSafetyTwoZeroOneNine}{156}
6535+\newcommand{\SecCoocFederatedLearningAiSafetyTwoZeroOneNinePct}{9.9}
6536+\newcommand{\SecCoocAppliedCryptographyAiSafetyTwoZeroOneNine}{129}
6537+\newcommand{\SecCoocAppliedCryptographyAiSafetyTwoZeroOneNinePct}{0.4}
6538+\newcommand{\SecCoocAppliedCryptographyAiAgentsTwoZeroOneNine}{123}
6539+\newcommand{\SecCoocAppliedCryptographyAiAgentsTwoZeroOneNinePct}{0.4}
6540+\newcommand{\SecCoocPrivacyPreservingComputationFoundationModelsTwoZeroOneNine}{102}
6541+\newcommand{\SecCoocPrivacyPreservingComputationFoundationModelsTwoZeroOneNinePct}{0.8}
6542+\newcommand{\SecCoocPrivacyPreservingComputationAiAgentsTwoZeroOneNine}{98}
6543+\newcommand{\SecCoocPrivacyPreservingComputationAiAgentsTwoZeroOneNinePct}{0.8}
6544+\newcommand{\SecCoocPrivacyPreservingComputationLlmTwoZeroOneNine}{85}
6545+\newcommand{\SecCoocPrivacyPreservingComputationLlmTwoZeroOneNinePct}{0.7}
6546+\newcommand{\SecCoocFederatedLearningFoundationModelsTwoZeroOneNine}{80}
6547+\newcommand{\SecCoocFederatedLearningFoundationModelsTwoZeroOneNinePct}{5.1}
6548+\newcommand{\SecCoocAppliedCryptographyFoundationModelsTwoZeroOneNine}{74}
6549+\newcommand{\SecCoocAppliedCryptographyFoundationModelsTwoZeroOneNinePct}{0.2}
6550+\newcommand{\SecCoocAppliedCryptographyGenerativeAiTwoZeroOneNine}{73}
6551+\newcommand{\SecCoocAppliedCryptographyGenerativeAiTwoZeroOneNinePct}{0.2}
6552+\newcommand{\SecCoocAppliedCryptographyLlmTwoZeroOneNine}{69}
6553+\newcommand{\SecCoocAppliedCryptographyLlmTwoZeroOneNinePct}{0.2}
6554+\newcommand{\SecCoocBlockchainFederatedLearningTwoZeroOneNine}{66}
6555+\newcommand{\SecCoocBlockchainFederatedLearningTwoZeroOneNinePct}{0.4}
6556+\newcommand{\SecCoocFederatedLearningGenerativeAiTwoZeroOneNine}{54}
6557+\newcommand{\SecCoocFederatedLearningGenerativeAiTwoZeroOneNinePct}{3.4}
6558+\newcommand{\SecCoocBlockchainAiSafetyTwoZeroOneNine}{51}
6559+\newcommand{\SecCoocBlockchainAiSafetyTwoZeroOneNinePct}{0.3}
6560+\newcommand{\SecCoocBlockchainFoundationModelsTwoZeroOneNine}{36}
6561+\newcommand{\SecCoocBlockchainFoundationModelsTwoZeroOneNinePct}{0.2}
6562+\newcommand{\SecCoocFederatedLearningAiAgentsTwoZeroOneNine}{34}
6563+\newcommand{\SecCoocFederatedLearningAiAgentsTwoZeroOneNinePct}{2.2}
6564+\newcommand{\SecCoocBlockchainGenerativeAiTwoZeroOneNine}{16}
6565+\newcommand{\SecCoocBlockchainGenerativeAiTwoZeroOneNinePct}{0.1}
6566+\newcommand{\SecCoocFederatedLearningLlmTwoZeroOneNine}{15}
6567+\newcommand{\SecCoocFederatedLearningLlmTwoZeroOneNinePct}{1.0}
6568+\newcommand{\SecCoocAppliedCryptographyCybersecurityTwoZeroTwoZero}{18{,}011}
6569+\newcommand{\SecCoocAppliedCryptographyCybersecurityTwoZeroTwoZeroPct}{56.7}
6570+\newcommand{\SecCoocCybersecurityPrivacyPreservingComputationTwoZeroTwoZero}{8{,}457}
6571+\newcommand{\SecCoocCybersecurityPrivacyPreservingComputationTwoZeroTwoZeroPct}{11.0}
6572+\newcommand{\SecCoocBlockchainCybersecurityTwoZeroTwoZero}{5{,}753}
6573+\newcommand{\SecCoocBlockchainCybersecurityTwoZeroTwoZeroPct}{34.4}
6574+\newcommand{\SecCoocAppliedCryptographyPrivacyPreservingComputationTwoZeroTwoZero}{5{,}039}
6575+\newcommand{\SecCoocAppliedCryptographyPrivacyPreservingComputationTwoZeroTwoZeroPct}{15.9}
6576+\newcommand{\SecCoocAppliedCryptographyBlockchainTwoZeroTwoZero}{4{,}349}
6577+\newcommand{\SecCoocAppliedCryptographyBlockchainTwoZeroTwoZeroPct}{13.7}
6578+\newcommand{\SecCoocCybersecurityAiSafetyTwoZeroTwoZero}{2{,}648}
6579+\newcommand{\SecCoocCybersecurityAiSafetyTwoZeroTwoZeroPct}{3.5}
6580+\newcommand{\SecCoocBlockchainPrivacyPreservingComputationTwoZeroTwoZero}{1{,}882}
6581+\newcommand{\SecCoocBlockchainPrivacyPreservingComputationTwoZeroTwoZeroPct}{11.3}
6582+\newcommand{\SecCoocFederatedLearningPrivacyPreservingComputationTwoZeroTwoZero}{1{,}858}
6583+\newcommand{\SecCoocFederatedLearningPrivacyPreservingComputationTwoZeroTwoZeroPct}{69.2}
6584+\newcommand{\SecCoocCybersecurityGenerativeAiTwoZeroTwoZero}{1{,}280}
6585+\newcommand{\SecCoocCybersecurityGenerativeAiTwoZeroTwoZeroPct}{1.7}
6586+\newcommand{\SecCoocPrivacyPreservingComputationAiSafetyTwoZeroTwoZero}{1{,}016}
6587+\newcommand{\SecCoocPrivacyPreservingComputationAiSafetyTwoZeroTwoZeroPct}{6.4}
6588+\newcommand{\SecCoocCybersecurityFederatedLearningTwoZeroTwoZero}{995}
6589+\newcommand{\SecCoocCybersecurityFederatedLearningTwoZeroTwoZeroPct}{1.3}
6590+\newcommand{\SecCoocCybersecurityLlmTwoZeroTwoZero}{985}
6591+\newcommand{\SecCoocCybersecurityLlmTwoZeroTwoZeroPct}{1.3}
6592+\newcommand{\SecCoocCybersecurityFoundationModelsTwoZeroTwoZero}{750}
6593+\newcommand{\SecCoocCybersecurityFoundationModelsTwoZeroTwoZeroPct}{1.0}
6594+\newcommand{\SecCoocCybersecurityAiAgentsTwoZeroTwoZero}{639}
6595+\newcommand{\SecCoocCybersecurityAiAgentsTwoZeroTwoZeroPct}{0.8}
6596+\newcommand{\SecCoocAppliedCryptographyFederatedLearningTwoZeroTwoZero}{563}
6597+\newcommand{\SecCoocAppliedCryptographyFederatedLearningTwoZeroTwoZeroPct}{1.8}
6598+\newcommand{\SecCoocPrivacyPreservingComputationGenerativeAiTwoZeroTwoZero}{311}
6599+\newcommand{\SecCoocPrivacyPreservingComputationGenerativeAiTwoZeroTwoZeroPct}{2.0}
6600+\newcommand{\SecCoocFederatedLearningAiSafetyTwoZeroTwoZero}{309}
6601+\newcommand{\SecCoocFederatedLearningAiSafetyTwoZeroTwoZeroPct}{11.5}
6602+\newcommand{\SecCoocBlockchainLlmTwoZeroTwoZero}{281}
6603+\newcommand{\SecCoocBlockchainLlmTwoZeroTwoZeroPct}{1.7}
6604+\newcommand{\SecCoocPrivacyPreservingComputationFoundationModelsTwoZeroTwoZero}{252}
6605+\newcommand{\SecCoocPrivacyPreservingComputationFoundationModelsTwoZeroTwoZeroPct}{1.6}
6606+\newcommand{\SecCoocAppliedCryptographyAiSafetyTwoZeroTwoZero}{218}
6607+\newcommand{\SecCoocAppliedCryptographyAiSafetyTwoZeroTwoZeroPct}{0.7}
6608+\newcommand{\SecCoocFederatedLearningFoundationModelsTwoZeroTwoZero}{199}
6609+\newcommand{\SecCoocFederatedLearningFoundationModelsTwoZeroTwoZeroPct}{7.4}
6610+\newcommand{\SecCoocBlockchainFederatedLearningTwoZeroTwoZero}{188}
6611+\newcommand{\SecCoocBlockchainFederatedLearningTwoZeroTwoZeroPct}{1.1}
6612+\newcommand{\SecCoocBlockchainAiAgentsTwoZeroTwoZero}{170}
6613+\newcommand{\SecCoocBlockchainAiAgentsTwoZeroTwoZeroPct}{1.0}
6614+\newcommand{\SecCoocPrivacyPreservingComputationAiAgentsTwoZeroTwoZero}{145}
6615+\newcommand{\SecCoocPrivacyPreservingComputationAiAgentsTwoZeroTwoZeroPct}{0.9}
6616+\newcommand{\SecCoocAppliedCryptographyAiAgentsTwoZeroTwoZero}{137}
6617+\newcommand{\SecCoocAppliedCryptographyAiAgentsTwoZeroTwoZeroPct}{0.4}
6618+\newcommand{\SecCoocPrivacyPreservingComputationLlmTwoZeroTwoZero}{130}
6619+\newcommand{\SecCoocPrivacyPreservingComputationLlmTwoZeroTwoZeroPct}{0.8}
6620+\newcommand{\SecCoocAppliedCryptographyFoundationModelsTwoZeroTwoZero}{128}
6621+\newcommand{\SecCoocAppliedCryptographyFoundationModelsTwoZeroTwoZeroPct}{0.4}
6622+\newcommand{\SecCoocAppliedCryptographyGenerativeAiTwoZeroTwoZero}{112}
6623+\newcommand{\SecCoocAppliedCryptographyGenerativeAiTwoZeroTwoZeroPct}{0.4}
6624+\newcommand{\SecCoocAppliedCryptographyLlmTwoZeroTwoZero}{108}
6625+\newcommand{\SecCoocAppliedCryptographyLlmTwoZeroTwoZeroPct}{0.3}
6626+\newcommand{\SecCoocFederatedLearningGenerativeAiTwoZeroTwoZero}{89}
6627+\newcommand{\SecCoocFederatedLearningGenerativeAiTwoZeroTwoZeroPct}{3.3}
6628+\newcommand{\SecCoocBlockchainAiSafetyTwoZeroTwoZero}{79}
6629+\newcommand{\SecCoocBlockchainAiSafetyTwoZeroTwoZeroPct}{0.5}
6630+\newcommand{\SecCoocFederatedLearningAiAgentsTwoZeroTwoZero}{51}
6631+\newcommand{\SecCoocFederatedLearningAiAgentsTwoZeroTwoZeroPct}{1.9}
6632+\newcommand{\SecCoocBlockchainFoundationModelsTwoZeroTwoZero}{42}
6633+\newcommand{\SecCoocBlockchainFoundationModelsTwoZeroTwoZeroPct}{0.3}
6634+\newcommand{\SecCoocFederatedLearningLlmTwoZeroTwoZero}{36}
6635+\newcommand{\SecCoocFederatedLearningLlmTwoZeroTwoZeroPct}{1.3}
6636+\newcommand{\SecCoocBlockchainGenerativeAiTwoZeroTwoZero}{25}
6637+\newcommand{\SecCoocBlockchainGenerativeAiTwoZeroTwoZeroPct}{0.1}
6638+\newcommand{\SecCoocAppliedCryptographyCybersecurityTwoZeroTwoOne}{18{,}080}
6639+\newcommand{\SecCoocAppliedCryptographyCybersecurityTwoZeroTwoOnePct}{56.3}
6640+\newcommand{\SecCoocCybersecurityPrivacyPreservingComputationTwoZeroTwoOne}{9{,}270}
6641+\newcommand{\SecCoocCybersecurityPrivacyPreservingComputationTwoZeroTwoOnePct}{12.0}
6642+\newcommand{\SecCoocBlockchainCybersecurityTwoZeroTwoOne}{6{,}334}
6643+\newcommand{\SecCoocBlockchainCybersecurityTwoZeroTwoOnePct}{34.9}
6644+\newcommand{\SecCoocAppliedCryptographyPrivacyPreservingComputationTwoZeroTwoOne}{5{,}505}
6645+\newcommand{\SecCoocAppliedCryptographyPrivacyPreservingComputationTwoZeroTwoOnePct}{17.2}
6646+\newcommand{\SecCoocAppliedCryptographyBlockchainTwoZeroTwoOne}{4{,}598}
6647+\newcommand{\SecCoocAppliedCryptographyBlockchainTwoZeroTwoOnePct}{14.3}
6648+\newcommand{\SecCoocCybersecurityAiSafetyTwoZeroTwoOne}{3{,}438}
6649+\newcommand{\SecCoocCybersecurityAiSafetyTwoZeroTwoOnePct}{4.4}
6650+\newcommand{\SecCoocFederatedLearningPrivacyPreservingComputationTwoZeroTwoOne}{3{,}147}
6651+\newcommand{\SecCoocFederatedLearningPrivacyPreservingComputationTwoZeroTwoOnePct}{74.1}
6652+\newcommand{\SecCoocBlockchainPrivacyPreservingComputationTwoZeroTwoOne}{2{,}408}
6653+\newcommand{\SecCoocBlockchainPrivacyPreservingComputationTwoZeroTwoOnePct}{13.3}
6654+\newcommand{\SecCoocCybersecurityGenerativeAiTwoZeroTwoOne}{1{,}630}
6655+\newcommand{\SecCoocCybersecurityGenerativeAiTwoZeroTwoOnePct}{2.1}
6656+\newcommand{\SecCoocCybersecurityFederatedLearningTwoZeroTwoOne}{1{,}420}
6657+\newcommand{\SecCoocCybersecurityFederatedLearningTwoZeroTwoOnePct}{1.8}
6658+\newcommand{\SecCoocPrivacyPreservingComputationAiSafetyTwoZeroTwoOne}{1{,}315}
6659+\newcommand{\SecCoocPrivacyPreservingComputationAiSafetyTwoZeroTwoOnePct}{7.4}
6660+\newcommand{\SecCoocCybersecurityLlmTwoZeroTwoOne}{1{,}231}
6661+\newcommand{\SecCoocCybersecurityLlmTwoZeroTwoOnePct}{1.6}
6662+\newcommand{\SecCoocCybersecurityFoundationModelsTwoZeroTwoOne}{1{,}097}
6663+\newcommand{\SecCoocCybersecurityFoundationModelsTwoZeroTwoOnePct}{1.4}
6664+\newcommand{\SecCoocAppliedCryptographyFederatedLearningTwoZeroTwoOne}{774}
6665+\newcommand{\SecCoocAppliedCryptographyFederatedLearningTwoZeroTwoOnePct}{2.4}
6666+\newcommand{\SecCoocCybersecurityAiAgentsTwoZeroTwoOne}{645}
6667+\newcommand{\SecCoocCybersecurityAiAgentsTwoZeroTwoOnePct}{0.8}
6668+\newcommand{\SecCoocFederatedLearningAiSafetyTwoZeroTwoOne}{461}
6669+\newcommand{\SecCoocFederatedLearningAiSafetyTwoZeroTwoOnePct}{10.9}
6670+\newcommand{\SecCoocPrivacyPreservingComputationFoundationModelsTwoZeroTwoOne}{451}
6671+\newcommand{\SecCoocPrivacyPreservingComputationFoundationModelsTwoZeroTwoOnePct}{2.5}
6672+\newcommand{\SecCoocPrivacyPreservingComputationGenerativeAiTwoZeroTwoOne}{407}
6673+\newcommand{\SecCoocPrivacyPreservingComputationGenerativeAiTwoZeroTwoOnePct}{2.3}
6674+\newcommand{\SecCoocFederatedLearningFoundationModelsTwoZeroTwoOne}{344}
6675+\newcommand{\SecCoocFederatedLearningFoundationModelsTwoZeroTwoOnePct}{8.1}
6676+\newcommand{\SecCoocBlockchainFederatedLearningTwoZeroTwoOne}{327}
6677+\newcommand{\SecCoocBlockchainFederatedLearningTwoZeroTwoOnePct}{1.8}
6678+\newcommand{\SecCoocAppliedCryptographyAiSafetyTwoZeroTwoOne}{285}
6679+\newcommand{\SecCoocAppliedCryptographyAiSafetyTwoZeroTwoOnePct}{0.9}
6680+\newcommand{\SecCoocBlockchainLlmTwoZeroTwoOne}{283}
6681+\newcommand{\SecCoocBlockchainLlmTwoZeroTwoOnePct}{1.6}
6682+\newcommand{\SecCoocPrivacyPreservingComputationLlmTwoZeroTwoOne}{233}
6683+\newcommand{\SecCoocPrivacyPreservingComputationLlmTwoZeroTwoOnePct}{1.3}
6684+\newcommand{\SecCoocAppliedCryptographyFoundationModelsTwoZeroTwoOne}{215}
6685+\newcommand{\SecCoocAppliedCryptographyFoundationModelsTwoZeroTwoOnePct}{0.7}
6686+\newcommand{\SecCoocPrivacyPreservingComputationAiAgentsTwoZeroTwoOne}{163}
6687+\newcommand{\SecCoocPrivacyPreservingComputationAiAgentsTwoZeroTwoOnePct}{0.9}
6688+\newcommand{\SecCoocBlockchainAiAgentsTwoZeroTwoOne}{159}
6689+\newcommand{\SecCoocBlockchainAiAgentsTwoZeroTwoOnePct}{0.9}
6690+\newcommand{\SecCoocAppliedCryptographyGenerativeAiTwoZeroTwoOne}{145}
6691+\newcommand{\SecCoocAppliedCryptographyGenerativeAiTwoZeroTwoOnePct}{0.5}
6692+\newcommand{\SecCoocAppliedCryptographyLlmTwoZeroTwoOne}{136}
6693+\newcommand{\SecCoocAppliedCryptographyLlmTwoZeroTwoOnePct}{0.4}
6694+\newcommand{\SecCoocAppliedCryptographyAiAgentsTwoZeroTwoOne}{127}
6695+\newcommand{\SecCoocAppliedCryptographyAiAgentsTwoZeroTwoOnePct}{0.4}
6696+\newcommand{\SecCoocFederatedLearningGenerativeAiTwoZeroTwoOne}{120}
6697+\newcommand{\SecCoocFederatedLearningGenerativeAiTwoZeroTwoOnePct}{2.8}
6698+\newcommand{\SecCoocFederatedLearningAiAgentsTwoZeroTwoOne}{78}
6699+\newcommand{\SecCoocFederatedLearningAiAgentsTwoZeroTwoOnePct}{1.8}
6700+\newcommand{\SecCoocBlockchainAiSafetyTwoZeroTwoOne}{75}
6701+\newcommand{\SecCoocBlockchainAiSafetyTwoZeroTwoOnePct}{0.4}
6702+\newcommand{\SecCoocFederatedLearningLlmTwoZeroTwoOne}{66}
6703+\newcommand{\SecCoocFederatedLearningLlmTwoZeroTwoOnePct}{1.6}
6704+\newcommand{\SecCoocBlockchainFoundationModelsTwoZeroTwoOne}{41}
6705+\newcommand{\SecCoocBlockchainFoundationModelsTwoZeroTwoOnePct}{0.2}
6706+\newcommand{\SecCoocBlockchainGenerativeAiTwoZeroTwoOne}{38}
6707+\newcommand{\SecCoocBlockchainGenerativeAiTwoZeroTwoOnePct}{0.2}
6708+\newcommand{\SecCoocAppliedCryptographyCybersecurityTwoZeroTwoTwo}{18{,}127}
6709+\newcommand{\SecCoocAppliedCryptographyCybersecurityTwoZeroTwoTwoPct}{56.5}
6710+\newcommand{\SecCoocCybersecurityPrivacyPreservingComputationTwoZeroTwoTwo}{10{,}184}
6711+\newcommand{\SecCoocCybersecurityPrivacyPreservingComputationTwoZeroTwoTwoPct}{13.0}
6712+\newcommand{\SecCoocBlockchainCybersecurityTwoZeroTwoTwo}{7{,}231}
6713+\newcommand{\SecCoocBlockchainCybersecurityTwoZeroTwoTwoPct}{35.9}
6714+\newcommand{\SecCoocAppliedCryptographyPrivacyPreservingComputationTwoZeroTwoTwo}{5{,}904}
6715+\newcommand{\SecCoocAppliedCryptographyPrivacyPreservingComputationTwoZeroTwoTwoPct}{18.4}
6716+\newcommand{\SecCoocAppliedCryptographyBlockchainTwoZeroTwoTwo}{5{,}171}
6717+\newcommand{\SecCoocAppliedCryptographyBlockchainTwoZeroTwoTwoPct}{16.1}
6718+\newcommand{\SecCoocFederatedLearningPrivacyPreservingComputationTwoZeroTwoTwo}{4{,}648}
6719+\newcommand{\SecCoocFederatedLearningPrivacyPreservingComputationTwoZeroTwoTwoPct}{78.9}
6720+\newcommand{\SecCoocCybersecurityAiSafetyTwoZeroTwoTwo}{4{,}056}
6721+\newcommand{\SecCoocCybersecurityAiSafetyTwoZeroTwoTwoPct}{5.2}
6722+\newcommand{\SecCoocBlockchainPrivacyPreservingComputationTwoZeroTwoTwo}{2{,}807}
6723+\newcommand{\SecCoocBlockchainPrivacyPreservingComputationTwoZeroTwoTwoPct}{13.9}
6724+\newcommand{\SecCoocCybersecurityFederatedLearningTwoZeroTwoTwo}{2{,}001}
6725+\newcommand{\SecCoocCybersecurityFederatedLearningTwoZeroTwoTwoPct}{2.6}
6726+\newcommand{\SecCoocCybersecurityGenerativeAiTwoZeroTwoTwo}{1{,}920}
6727+\newcommand{\SecCoocCybersecurityGenerativeAiTwoZeroTwoTwoPct}{2.5}
6728+\newcommand{\SecCoocPrivacyPreservingComputationAiSafetyTwoZeroTwoTwo}{1{,}735}
6729+\newcommand{\SecCoocPrivacyPreservingComputationAiSafetyTwoZeroTwoTwoPct}{8.7}
6730+\newcommand{\SecCoocCybersecurityLlmTwoZeroTwoTwo}{1{,}337}
6731+\newcommand{\SecCoocCybersecurityLlmTwoZeroTwoTwoPct}{1.7}
6732+\newcommand{\SecCoocCybersecurityFoundationModelsTwoZeroTwoTwo}{1{,}281}
6733+\newcommand{\SecCoocCybersecurityFoundationModelsTwoZeroTwoTwoPct}{1.6}
6734+\newcommand{\SecCoocAppliedCryptographyFederatedLearningTwoZeroTwoTwo}{1{,}008}
6735+\newcommand{\SecCoocAppliedCryptographyFederatedLearningTwoZeroTwoTwoPct}{3.1}
6736+\newcommand{\SecCoocPrivacyPreservingComputationFoundationModelsTwoZeroTwoTwo}{681}
6737+\newcommand{\SecCoocPrivacyPreservingComputationFoundationModelsTwoZeroTwoTwoPct}{3.4}
6738+\newcommand{\SecCoocFederatedLearningAiSafetyTwoZeroTwoTwo}{626}
6739+\newcommand{\SecCoocFederatedLearningAiSafetyTwoZeroTwoTwoPct}{10.6}
6740+\newcommand{\SecCoocCybersecurityAiAgentsTwoZeroTwoTwo}{596}
6741+\newcommand{\SecCoocCybersecurityAiAgentsTwoZeroTwoTwoPct}{0.8}
6742+\newcommand{\SecCoocBlockchainFederatedLearningTwoZeroTwoTwo}{542}
6743+\newcommand{\SecCoocBlockchainFederatedLearningTwoZeroTwoTwoPct}{2.7}
6744+\newcommand{\SecCoocPrivacyPreservingComputationGenerativeAiTwoZeroTwoTwo}{526}
6745+\newcommand{\SecCoocPrivacyPreservingComputationGenerativeAiTwoZeroTwoTwoPct}{2.6}
6746+\newcommand{\SecCoocFederatedLearningFoundationModelsTwoZeroTwoTwo}{521}
6747+\newcommand{\SecCoocFederatedLearningFoundationModelsTwoZeroTwoTwoPct}{8.8}
6748+\newcommand{\SecCoocAppliedCryptographyFoundationModelsTwoZeroTwoTwo}{377}
6749+\newcommand{\SecCoocAppliedCryptographyFoundationModelsTwoZeroTwoTwoPct}{1.2}
6750+\newcommand{\SecCoocAppliedCryptographyAiSafetyTwoZeroTwoTwo}{353}
6751+\newcommand{\SecCoocAppliedCryptographyAiSafetyTwoZeroTwoTwoPct}{1.1}
6752+\newcommand{\SecCoocPrivacyPreservingComputationLlmTwoZeroTwoTwo}{278}
6753+\newcommand{\SecCoocPrivacyPreservingComputationLlmTwoZeroTwoTwoPct}{1.4}
6754+\newcommand{\SecCoocBlockchainLlmTwoZeroTwoTwo}{275}
6755+\newcommand{\SecCoocBlockchainLlmTwoZeroTwoTwoPct}{1.4}
6756+\newcommand{\SecCoocAppliedCryptographyGenerativeAiTwoZeroTwoTwo}{205}
6757+\newcommand{\SecCoocAppliedCryptographyGenerativeAiTwoZeroTwoTwoPct}{0.6}
6758+\newcommand{\SecCoocAppliedCryptographyLlmTwoZeroTwoTwo}{198}
6759+\newcommand{\SecCoocAppliedCryptographyLlmTwoZeroTwoTwoPct}{0.6}
6760+\newcommand{\SecCoocPrivacyPreservingComputationAiAgentsTwoZeroTwoTwo}{171}
6761+\newcommand{\SecCoocPrivacyPreservingComputationAiAgentsTwoZeroTwoTwoPct}{0.9}
6762+\newcommand{\SecCoocFederatedLearningGenerativeAiTwoZeroTwoTwo}{165}
6763+\newcommand{\SecCoocFederatedLearningGenerativeAiTwoZeroTwoTwoPct}{2.8}
6764+\newcommand{\SecCoocBlockchainAiAgentsTwoZeroTwoTwo}{145}
6765+\newcommand{\SecCoocBlockchainAiAgentsTwoZeroTwoTwoPct}{0.7}
6766+\newcommand{\SecCoocAppliedCryptographyAiAgentsTwoZeroTwoTwo}{135}
6767+\newcommand{\SecCoocAppliedCryptographyAiAgentsTwoZeroTwoTwoPct}{0.4}
6768+\newcommand{\SecCoocBlockchainAiSafetyTwoZeroTwoTwo}{113}
6769+\newcommand{\SecCoocBlockchainAiSafetyTwoZeroTwoTwoPct}{0.6}
6770+\newcommand{\SecCoocFederatedLearningLlmTwoZeroTwoTwo}{91}
6771+\newcommand{\SecCoocFederatedLearningLlmTwoZeroTwoTwoPct}{1.5}
6772+\newcommand{\SecCoocFederatedLearningAiAgentsTwoZeroTwoTwo}{86}
6773+\newcommand{\SecCoocFederatedLearningAiAgentsTwoZeroTwoTwoPct}{1.5}
6774+\newcommand{\SecCoocBlockchainGenerativeAiTwoZeroTwoTwo}{80}
6775+\newcommand{\SecCoocBlockchainGenerativeAiTwoZeroTwoTwoPct}{0.4}
6776+\newcommand{\SecCoocBlockchainFoundationModelsTwoZeroTwoTwo}{53}
6777+\newcommand{\SecCoocBlockchainFoundationModelsTwoZeroTwoTwoPct}{0.3}
6778+\newcommand{\SecCoocAppliedCryptographyCybersecurityTwoZeroTwoThree}{20{,}517}
6779+\newcommand{\SecCoocAppliedCryptographyCybersecurityTwoZeroTwoThreePct}{57.4}
6780+\newcommand{\SecCoocCybersecurityPrivacyPreservingComputationTwoZeroTwoThree}{12{,}213}
6781+\newcommand{\SecCoocCybersecurityPrivacyPreservingComputationTwoZeroTwoThreePct}{13.6}
6782+\newcommand{\SecCoocBlockchainCybersecurityTwoZeroTwoThree}{9{,}090}
6783+\newcommand{\SecCoocBlockchainCybersecurityTwoZeroTwoThreePct}{37.4}
6784+\newcommand{\SecCoocAppliedCryptographyPrivacyPreservingComputationTwoZeroTwoThree}{7{,}163}
6785+\newcommand{\SecCoocAppliedCryptographyPrivacyPreservingComputationTwoZeroTwoThreePct}{20.1}
6786+\newcommand{\SecCoocFederatedLearningPrivacyPreservingComputationTwoZeroTwoThree}{7{,}023}
6787+\newcommand{\SecCoocFederatedLearningPrivacyPreservingComputationTwoZeroTwoThreePct}{82.5}
6788+\newcommand{\SecCoocAppliedCryptographyBlockchainTwoZeroTwoThree}{6{,}123}
6789+\newcommand{\SecCoocAppliedCryptographyBlockchainTwoZeroTwoThreePct}{17.1}
6790+\newcommand{\SecCoocCybersecurityAiSafetyTwoZeroTwoThree}{5{,}288}
6791+\newcommand{\SecCoocCybersecurityAiSafetyTwoZeroTwoThreePct}{5.9}
6792+\newcommand{\SecCoocBlockchainPrivacyPreservingComputationTwoZeroTwoThree}{3{,}585}
6793+\newcommand{\SecCoocBlockchainPrivacyPreservingComputationTwoZeroTwoThreePct}{14.7}
6794+\newcommand{\SecCoocCybersecurityFederatedLearningTwoZeroTwoThree}{2{,}852}
6795+\newcommand{\SecCoocCybersecurityFederatedLearningTwoZeroTwoThreePct}{3.2}
6796+\newcommand{\SecCoocCybersecurityGenerativeAiTwoZeroTwoThree}{2{,}765}
6797+\newcommand{\SecCoocCybersecurityGenerativeAiTwoZeroTwoThreePct}{3.1}
6798+\newcommand{\SecCoocPrivacyPreservingComputationAiSafetyTwoZeroTwoThree}{2{,}410}
6799+\newcommand{\SecCoocPrivacyPreservingComputationAiSafetyTwoZeroTwoThreePct}{9.6}
6800+\newcommand{\SecCoocCybersecurityLlmTwoZeroTwoThree}{2{,}304}
6801+\newcommand{\SecCoocCybersecurityLlmTwoZeroTwoThreePct}{2.6}
6802+\newcommand{\SecCoocCybersecurityFoundationModelsTwoZeroTwoThree}{1{,}764}
6803+\newcommand{\SecCoocCybersecurityFoundationModelsTwoZeroTwoThreePct}{2.0}
6804+\newcommand{\SecCoocAppliedCryptographyFederatedLearningTwoZeroTwoThree}{1{,}473}
6805+\newcommand{\SecCoocAppliedCryptographyFederatedLearningTwoZeroTwoThreePct}{4.1}
6806+\newcommand{\SecCoocPrivacyPreservingComputationFoundationModelsTwoZeroTwoThree}{1{,}083}
6807+\newcommand{\SecCoocPrivacyPreservingComputationFoundationModelsTwoZeroTwoThreePct}{4.3}
6808+\newcommand{\SecCoocPrivacyPreservingComputationGenerativeAiTwoZeroTwoThree}{897}
6809+\newcommand{\SecCoocPrivacyPreservingComputationGenerativeAiTwoZeroTwoThreePct}{3.6}
6810+\newcommand{\SecCoocBlockchainFederatedLearningTwoZeroTwoThree}{872}
6811+\newcommand{\SecCoocBlockchainFederatedLearningTwoZeroTwoThreePct}{3.6}
6812+\newcommand{\SecCoocFederatedLearningAiSafetyTwoZeroTwoThree}{857}
6813+\newcommand{\SecCoocFederatedLearningAiSafetyTwoZeroTwoThreePct}{10.1}
6814+\newcommand{\SecCoocCybersecurityAiAgentsTwoZeroTwoThree}{807}
6815+\newcommand{\SecCoocCybersecurityAiAgentsTwoZeroTwoThreePct}{0.9}
6816+\newcommand{\SecCoocFederatedLearningFoundationModelsTwoZeroTwoThree}{801}
6817+\newcommand{\SecCoocFederatedLearningFoundationModelsTwoZeroTwoThreePct}{9.4}
6818+\newcommand{\SecCoocPrivacyPreservingComputationLlmTwoZeroTwoThree}{717}
6819+\newcommand{\SecCoocPrivacyPreservingComputationLlmTwoZeroTwoThreePct}{2.8}
6820+\newcommand{\SecCoocAppliedCryptographyFoundationModelsTwoZeroTwoThree}{571}
6821+\newcommand{\SecCoocAppliedCryptographyFoundationModelsTwoZeroTwoThreePct}{1.6}
6822+\newcommand{\SecCoocAppliedCryptographyAiSafetyTwoZeroTwoThree}{530}
6823+\newcommand{\SecCoocAppliedCryptographyAiSafetyTwoZeroTwoThreePct}{1.5}
6824+\newcommand{\SecCoocAppliedCryptographyLlmTwoZeroTwoThree}{403}
6825+\newcommand{\SecCoocAppliedCryptographyLlmTwoZeroTwoThreePct}{1.1}
6826+\newcommand{\SecCoocBlockchainLlmTwoZeroTwoThree}{357}
6827+\newcommand{\SecCoocBlockchainLlmTwoZeroTwoThreePct}{1.5}
6828+\newcommand{\SecCoocAppliedCryptographyGenerativeAiTwoZeroTwoThree}{286}
6829+\newcommand{\SecCoocAppliedCryptographyGenerativeAiTwoZeroTwoThreePct}{0.8}
6830+\newcommand{\SecCoocPrivacyPreservingComputationAiAgentsTwoZeroTwoThree}{267}
6831+\newcommand{\SecCoocPrivacyPreservingComputationAiAgentsTwoZeroTwoThreePct}{1.1}
6832+\newcommand{\SecCoocFederatedLearningGenerativeAiTwoZeroTwoThree}{258}
6833+\newcommand{\SecCoocFederatedLearningGenerativeAiTwoZeroTwoThreePct}{3.0}
6834+\newcommand{\SecCoocFederatedLearningLlmTwoZeroTwoThree}{226}
6835+\newcommand{\SecCoocFederatedLearningLlmTwoZeroTwoThreePct}{2.7}
6836+\newcommand{\SecCoocBlockchainAiAgentsTwoZeroTwoThree}{191}
6837+\newcommand{\SecCoocBlockchainAiAgentsTwoZeroTwoThreePct}{0.8}
6838+\newcommand{\SecCoocBlockchainAiSafetyTwoZeroTwoThree}{184}
6839+\newcommand{\SecCoocBlockchainAiSafetyTwoZeroTwoThreePct}{0.8}
6840+\newcommand{\SecCoocAppliedCryptographyAiAgentsTwoZeroTwoThree}{175}
6841+\newcommand{\SecCoocAppliedCryptographyAiAgentsTwoZeroTwoThreePct}{0.5}
6842+\newcommand{\SecCoocFederatedLearningAiAgentsTwoZeroTwoThree}{144}
6843+\newcommand{\SecCoocFederatedLearningAiAgentsTwoZeroTwoThreePct}{1.7}
6844+\newcommand{\SecCoocBlockchainGenerativeAiTwoZeroTwoThree}{116}
6845+\newcommand{\SecCoocBlockchainGenerativeAiTwoZeroTwoThreePct}{0.5}
6846+\newcommand{\SecCoocBlockchainFoundationModelsTwoZeroTwoThree}{63}
6847+\newcommand{\SecCoocBlockchainFoundationModelsTwoZeroTwoThreePct}{0.3}
6848+\newcommand{\SecCoocAppliedCryptographyCybersecurityTwoZeroTwoFour}{24{,}606}
6849+\newcommand{\SecCoocAppliedCryptographyCybersecurityTwoZeroTwoFourPct}{59.8}
6850+\newcommand{\SecCoocCybersecurityPrivacyPreservingComputationTwoZeroTwoFour}{16{,}115}
6851+\newcommand{\SecCoocCybersecurityPrivacyPreservingComputationTwoZeroTwoFourPct}{15.0}
6852+\newcommand{\SecCoocBlockchainCybersecurityTwoZeroTwoFour}{10{,}704}
6853+\newcommand{\SecCoocBlockchainCybersecurityTwoZeroTwoFourPct}{39.8}
6854+\newcommand{\SecCoocFederatedLearningPrivacyPreservingComputationTwoZeroTwoFour}{9{,}966}
6855+\newcommand{\SecCoocFederatedLearningPrivacyPreservingComputationTwoZeroTwoFourPct}{83.6}
6856+\newcommand{\SecCoocAppliedCryptographyPrivacyPreservingComputationTwoZeroTwoFour}{9{,}178}
6857+\newcommand{\SecCoocAppliedCryptographyPrivacyPreservingComputationTwoZeroTwoFourPct}{22.3}
6858+\newcommand{\SecCoocCybersecurityAiSafetyTwoZeroTwoFour}{7{,}850}
6859+\newcommand{\SecCoocCybersecurityAiSafetyTwoZeroTwoFourPct}{7.3}
6860+\newcommand{\SecCoocAppliedCryptographyBlockchainTwoZeroTwoFour}{7{,}116}
6861+\newcommand{\SecCoocAppliedCryptographyBlockchainTwoZeroTwoFourPct}{17.3}
6862+\newcommand{\SecCoocCybersecurityLlmTwoZeroTwoFour}{4{,}977}
6863+\newcommand{\SecCoocCybersecurityLlmTwoZeroTwoFourPct}{4.6}
6864+\newcommand{\SecCoocCybersecurityGenerativeAiTwoZeroTwoFour}{4{,}706}
6865+\newcommand{\SecCoocCybersecurityGenerativeAiTwoZeroTwoFourPct}{4.4}
6866+\newcommand{\SecCoocBlockchainPrivacyPreservingComputationTwoZeroTwoFour}{4{,}636}
6867+\newcommand{\SecCoocBlockchainPrivacyPreservingComputationTwoZeroTwoFourPct}{17.2}
6868+\newcommand{\SecCoocCybersecurityFederatedLearningTwoZeroTwoFour}{4{,}126}
6869+\newcommand{\SecCoocCybersecurityFederatedLearningTwoZeroTwoFourPct}{3.8}
6870+\newcommand{\SecCoocPrivacyPreservingComputationAiSafetyTwoZeroTwoFour}{4{,}069}
6871+\newcommand{\SecCoocPrivacyPreservingComputationAiSafetyTwoZeroTwoFourPct}{11.8}
6872+\newcommand{\SecCoocCybersecurityFoundationModelsTwoZeroTwoFour}{2{,}457}
6873+\newcommand{\SecCoocCybersecurityFoundationModelsTwoZeroTwoFourPct}{2.3}
6874+\newcommand{\SecCoocAppliedCryptographyFederatedLearningTwoZeroTwoFour}{2{,}169}
6875+\newcommand{\SecCoocAppliedCryptographyFederatedLearningTwoZeroTwoFourPct}{5.3}
6876+\newcommand{\SecCoocPrivacyPreservingComputationLlmTwoZeroTwoFour}{2{,}022}
6877+\newcommand{\SecCoocPrivacyPreservingComputationLlmTwoZeroTwoFourPct}{5.8}
6878+\newcommand{\SecCoocPrivacyPreservingComputationFoundationModelsTwoZeroTwoFour}{1{,}778}
6879+\newcommand{\SecCoocPrivacyPreservingComputationFoundationModelsTwoZeroTwoFourPct}{5.1}
6880+\newcommand{\SecCoocPrivacyPreservingComputationGenerativeAiTwoZeroTwoFour}{1{,}720}
6881+\newcommand{\SecCoocPrivacyPreservingComputationGenerativeAiTwoZeroTwoFourPct}{5.0}
6882+\newcommand{\SecCoocFederatedLearningAiSafetyTwoZeroTwoFour}{1{,}314}
6883+\newcommand{\SecCoocFederatedLearningAiSafetyTwoZeroTwoFourPct}{11.0}
6884+\newcommand{\SecCoocBlockchainFederatedLearningTwoZeroTwoFour}{1{,}260}
6885+\newcommand{\SecCoocBlockchainFederatedLearningTwoZeroTwoFourPct}{4.7}
6886+\newcommand{\SecCoocFederatedLearningFoundationModelsTwoZeroTwoFour}{1{,}246}
6887+\newcommand{\SecCoocFederatedLearningFoundationModelsTwoZeroTwoFourPct}{10.5}
6888+\newcommand{\SecCoocCybersecurityAiAgentsTwoZeroTwoFour}{1{,}233}
6889+\newcommand{\SecCoocCybersecurityAiAgentsTwoZeroTwoFourPct}{1.1}
6890+\newcommand{\SecCoocAppliedCryptographyAiSafetyTwoZeroTwoFour}{874}
6891+\newcommand{\SecCoocAppliedCryptographyAiSafetyTwoZeroTwoFourPct}{2.1}
6892+\newcommand{\SecCoocAppliedCryptographyLlmTwoZeroTwoFour}{839}
6893+\newcommand{\SecCoocAppliedCryptographyLlmTwoZeroTwoFourPct}{2.0}
6894+\newcommand{\SecCoocAppliedCryptographyFoundationModelsTwoZeroTwoFour}{755}
6895+\newcommand{\SecCoocAppliedCryptographyFoundationModelsTwoZeroTwoFourPct}{1.8}
6896+\newcommand{\SecCoocFederatedLearningLlmTwoZeroTwoFour}{601}
6897+\newcommand{\SecCoocFederatedLearningLlmTwoZeroTwoFourPct}{5.0}
6898+\newcommand{\SecCoocBlockchainLlmTwoZeroTwoFour}{508}
6899+\newcommand{\SecCoocBlockchainLlmTwoZeroTwoFourPct}{1.9}
6900+\newcommand{\SecCoocAppliedCryptographyGenerativeAiTwoZeroTwoFour}{497}
6901+\newcommand{\SecCoocAppliedCryptographyGenerativeAiTwoZeroTwoFourPct}{1.2}
6902+\newcommand{\SecCoocFederatedLearningGenerativeAiTwoZeroTwoFour}{480}
6903+\newcommand{\SecCoocFederatedLearningGenerativeAiTwoZeroTwoFourPct}{4.0}
6904+\newcommand{\SecCoocPrivacyPreservingComputationAiAgentsTwoZeroTwoFour}{420}
6905+\newcommand{\SecCoocPrivacyPreservingComputationAiAgentsTwoZeroTwoFourPct}{1.2}
6906+\newcommand{\SecCoocBlockchainAiSafetyTwoZeroTwoFour}{331}
6907+\newcommand{\SecCoocBlockchainAiSafetyTwoZeroTwoFourPct}{1.2}
6908+\newcommand{\SecCoocAppliedCryptographyAiAgentsTwoZeroTwoFour}{299}
6909+\newcommand{\SecCoocAppliedCryptographyAiAgentsTwoZeroTwoFourPct}{0.7}
6910+\newcommand{\SecCoocBlockchainAiAgentsTwoZeroTwoFour}{278}
6911+\newcommand{\SecCoocBlockchainAiAgentsTwoZeroTwoFourPct}{1.0}
6912+\newcommand{\SecCoocBlockchainGenerativeAiTwoZeroTwoFour}{244}
6913+\newcommand{\SecCoocBlockchainGenerativeAiTwoZeroTwoFourPct}{0.9}
6914+\newcommand{\SecCoocFederatedLearningAiAgentsTwoZeroTwoFour}{158}
6915+\newcommand{\SecCoocFederatedLearningAiAgentsTwoZeroTwoFourPct}{1.3}
6916+\newcommand{\SecCoocBlockchainFoundationModelsTwoZeroTwoFour}{92}
6917+\newcommand{\SecCoocBlockchainFoundationModelsTwoZeroTwoFourPct}{0.3}
6918+\newcommand{\SecCoocAppliedCryptographyCybersecurityTwoZeroTwoFive}{31{,}593}
6919+\newcommand{\SecCoocAppliedCryptographyCybersecurityTwoZeroTwoFivePct}{61.0}
6920+\newcommand{\SecCoocCybersecurityPrivacyPreservingComputationTwoZeroTwoFive}{22{,}170}
6921+\newcommand{\SecCoocCybersecurityPrivacyPreservingComputationTwoZeroTwoFivePct}{15.9}
6922+\newcommand{\SecCoocBlockchainCybersecurityTwoZeroTwoFive}{14{,}843}
6923+\newcommand{\SecCoocBlockchainCybersecurityTwoZeroTwoFivePct}{44.4}
6924+\newcommand{\SecCoocFederatedLearningPrivacyPreservingComputationTwoZeroTwoFive}{14{,}526}
6925+\newcommand{\SecCoocFederatedLearningPrivacyPreservingComputationTwoZeroTwoFivePct}{80.3}
6926+\newcommand{\SecCoocCybersecurityAiSafetyTwoZeroTwoFive}{12{,}933}
6927+\newcommand{\SecCoocCybersecurityAiSafetyTwoZeroTwoFivePct}{9.3}
6928+\newcommand{\SecCoocAppliedCryptographyPrivacyPreservingComputationTwoZeroTwoFive}{12{,}475}
6929+\newcommand{\SecCoocAppliedCryptographyPrivacyPreservingComputationTwoZeroTwoFivePct}{24.1}
6930+\newcommand{\SecCoocCybersecurityLlmTwoZeroTwoFive}{10{,}549}
6931+\newcommand{\SecCoocCybersecurityLlmTwoZeroTwoFivePct}{7.6}
6932+\newcommand{\SecCoocAppliedCryptographyBlockchainTwoZeroTwoFive}{9{,}947}
6933+\newcommand{\SecCoocAppliedCryptographyBlockchainTwoZeroTwoFivePct}{19.2}
6934+\newcommand{\SecCoocCybersecurityGenerativeAiTwoZeroTwoFive}{7{,}969}
6935+\newcommand{\SecCoocCybersecurityGenerativeAiTwoZeroTwoFivePct}{5.7}
6936+\newcommand{\SecCoocPrivacyPreservingComputationAiSafetyTwoZeroTwoFive}{7{,}289}
6937+\newcommand{\SecCoocPrivacyPreservingComputationAiSafetyTwoZeroTwoFivePct}{14.8}
6938+\newcommand{\SecCoocBlockchainPrivacyPreservingComputationTwoZeroTwoFive}{6{,}742}
6939+\newcommand{\SecCoocBlockchainPrivacyPreservingComputationTwoZeroTwoFivePct}{20.2}
6940+\newcommand{\SecCoocCybersecurityFederatedLearningTwoZeroTwoFive}{6{,}679}
6941+\newcommand{\SecCoocCybersecurityFederatedLearningTwoZeroTwoFivePct}{4.8}
6942+\newcommand{\SecCoocPrivacyPreservingComputationLlmTwoZeroTwoFive}{4{,}584}
6943+\newcommand{\SecCoocPrivacyPreservingComputationLlmTwoZeroTwoFivePct}{9.3}
6944+\newcommand{\SecCoocCybersecurityFoundationModelsTwoZeroTwoFive}{3{,}561}
6945+\newcommand{\SecCoocCybersecurityFoundationModelsTwoZeroTwoFivePct}{2.6}
6946+\newcommand{\SecCoocAppliedCryptographyFederatedLearningTwoZeroTwoFive}{3{,}365}
6947+\newcommand{\SecCoocAppliedCryptographyFederatedLearningTwoZeroTwoFivePct}{6.5}
6948+\newcommand{\SecCoocCybersecurityAiAgentsTwoZeroTwoFive}{3{,}207}
6949+\newcommand{\SecCoocCybersecurityAiAgentsTwoZeroTwoFivePct}{2.3}
6950+\newcommand{\SecCoocPrivacyPreservingComputationGenerativeAiTwoZeroTwoFive}{3{,}010}
6951+\newcommand{\SecCoocPrivacyPreservingComputationGenerativeAiTwoZeroTwoFivePct}{6.1}
6952+\newcommand{\SecCoocPrivacyPreservingComputationFoundationModelsTwoZeroTwoFive}{2{,}663}
6953+\newcommand{\SecCoocPrivacyPreservingComputationFoundationModelsTwoZeroTwoFivePct}{5.4}
6954+\newcommand{\SecCoocFederatedLearningAiSafetyTwoZeroTwoFive}{2{,}295}
6955+\newcommand{\SecCoocFederatedLearningAiSafetyTwoZeroTwoFivePct}{12.7}
6956+\newcommand{\SecCoocBlockchainFederatedLearningTwoZeroTwoFive}{2{,}200}
6957+\newcommand{\SecCoocBlockchainFederatedLearningTwoZeroTwoFivePct}{6.6}
6958+\newcommand{\SecCoocAppliedCryptographyLlmTwoZeroTwoFive}{1{,}745}
6959+\newcommand{\SecCoocAppliedCryptographyLlmTwoZeroTwoFivePct}{3.4}
6960+\newcommand{\SecCoocFederatedLearningFoundationModelsTwoZeroTwoFive}{1{,}728}
6961+\newcommand{\SecCoocFederatedLearningFoundationModelsTwoZeroTwoFivePct}{9.6}
6962+\newcommand{\SecCoocAppliedCryptographyAiSafetyTwoZeroTwoFive}{1{,}539}
6963+\newcommand{\SecCoocAppliedCryptographyAiSafetyTwoZeroTwoFivePct}{3.0}
6964+\newcommand{\SecCoocFederatedLearningLlmTwoZeroTwoFive}{1{,}374}
6965+\newcommand{\SecCoocFederatedLearningLlmTwoZeroTwoFivePct}{7.6}
6966+\newcommand{\SecCoocPrivacyPreservingComputationAiAgentsTwoZeroTwoFive}{1{,}112}
6967+\newcommand{\SecCoocPrivacyPreservingComputationAiAgentsTwoZeroTwoFivePct}{2.3}
6968+\newcommand{\SecCoocBlockchainLlmTwoZeroTwoFive}{1{,}033}
6969+\newcommand{\SecCoocBlockchainLlmTwoZeroTwoFivePct}{3.1}
6970+\newcommand{\SecCoocAppliedCryptographyFoundationModelsTwoZeroTwoFive}{993}
6971+\newcommand{\SecCoocAppliedCryptographyFoundationModelsTwoZeroTwoFivePct}{1.9}
6972+\newcommand{\SecCoocAppliedCryptographyGenerativeAiTwoZeroTwoFive}{895}
6973+\newcommand{\SecCoocAppliedCryptographyGenerativeAiTwoZeroTwoFivePct}{1.7}
6974+\newcommand{\SecCoocFederatedLearningGenerativeAiTwoZeroTwoFive}{883}
6975+\newcommand{\SecCoocFederatedLearningGenerativeAiTwoZeroTwoFivePct}{4.9}
6976+\newcommand{\SecCoocBlockchainAiSafetyTwoZeroTwoFive}{770}
6977+\newcommand{\SecCoocBlockchainAiSafetyTwoZeroTwoFivePct}{2.3}
6978+\newcommand{\SecCoocAppliedCryptographyAiAgentsTwoZeroTwoFive}{700}
6979+\newcommand{\SecCoocAppliedCryptographyAiAgentsTwoZeroTwoFivePct}{1.4}
6980+\newcommand{\SecCoocBlockchainAiAgentsTwoZeroTwoFive}{664}
6981+\newcommand{\SecCoocBlockchainAiAgentsTwoZeroTwoFivePct}{2.0}
6982+\newcommand{\SecCoocBlockchainGenerativeAiTwoZeroTwoFive}{466}
6983+\newcommand{\SecCoocBlockchainGenerativeAiTwoZeroTwoFivePct}{1.4}
6984+\newcommand{\SecCoocFederatedLearningAiAgentsTwoZeroTwoFive}{462}
6985+\newcommand{\SecCoocFederatedLearningAiAgentsTwoZeroTwoFivePct}{2.6}
6986+\newcommand{\SecCoocBlockchainFoundationModelsTwoZeroTwoFive}{152}
6987+\newcommand{\SecCoocBlockchainFoundationModelsTwoZeroTwoFivePct}{0.5}
6988+\newcommand{\SecCoocAppliedCryptographyCybersecurityTwoZeroTwoSix}{25{,}111}
6989+\newcommand{\SecCoocAppliedCryptographyCybersecurityTwoZeroTwoSixPct}{57.1}
6990+\newcommand{\SecCoocCybersecurityPrivacyPreservingComputationTwoZeroTwoSix}{17{,}928}
6991+\newcommand{\SecCoocCybersecurityPrivacyPreservingComputationTwoZeroTwoSixPct}{14.9}
6992+\newcommand{\SecCoocCybersecurityLlmTwoZeroTwoSix}{14{,}410}
6993+\newcommand{\SecCoocCybersecurityLlmTwoZeroTwoSixPct}{12.0}
6994+\newcommand{\SecCoocCybersecurityAiSafetyTwoZeroTwoSix}{14{,}029}
6995+\newcommand{\SecCoocCybersecurityAiSafetyTwoZeroTwoSixPct}{11.6}
6996+\newcommand{\SecCoocFederatedLearningPrivacyPreservingComputationTwoZeroTwoSix}{12{,}665}
6997+\newcommand{\SecCoocFederatedLearningPrivacyPreservingComputationTwoZeroTwoSixPct}{74.2}
6998+\newcommand{\SecCoocBlockchainCybersecurityTwoZeroTwoSix}{10{,}861}
6999+\newcommand{\SecCoocBlockchainCybersecurityTwoZeroTwoSixPct}{43.8}
7000+\newcommand{\SecCoocAppliedCryptographyPrivacyPreservingComputationTwoZeroTwoSix}{9{,}424}
7001+\newcommand{\SecCoocAppliedCryptographyPrivacyPreservingComputationTwoZeroTwoSixPct}{21.4}
7002+\newcommand{\SecCoocCybersecurityGenerativeAiTwoZeroTwoSix}{7{,}571}
7003+\newcommand{\SecCoocCybersecurityGenerativeAiTwoZeroTwoSixPct}{6.3}
7004+\newcommand{\SecCoocAppliedCryptographyBlockchainTwoZeroTwoSix}{7{,}508}
7005+\newcommand{\SecCoocAppliedCryptographyBlockchainTwoZeroTwoSixPct}{17.1}
7006+\newcommand{\SecCoocPrivacyPreservingComputationAiSafetyTwoZeroTwoSix}{7{,}075}
7007+\newcommand{\SecCoocPrivacyPreservingComputationAiSafetyTwoZeroTwoSixPct}{17.0}
7008+\newcommand{\SecCoocCybersecurityAiAgentsTwoZeroTwoSix}{6{,}329}
7009+\newcommand{\SecCoocCybersecurityAiAgentsTwoZeroTwoSixPct}{5.3}
7010+\newcommand{\SecCoocCybersecurityFederatedLearningTwoZeroTwoSix}{5{,}838}
7011+\newcommand{\SecCoocCybersecurityFederatedLearningTwoZeroTwoSixPct}{4.8}
7012+\newcommand{\SecCoocPrivacyPreservingComputationLlmTwoZeroTwoSix}{5{,}665}
7013+\newcommand{\SecCoocPrivacyPreservingComputationLlmTwoZeroTwoSixPct}{13.6}
7014+\newcommand{\SecCoocBlockchainPrivacyPreservingComputationTwoZeroTwoSix}{4{,}759}
7015+\newcommand{\SecCoocBlockchainPrivacyPreservingComputationTwoZeroTwoSixPct}{19.2}
7016+\newcommand{\SecCoocCybersecurityFoundationModelsTwoZeroTwoSix}{4{,}094}
7017+\newcommand{\SecCoocCybersecurityFoundationModelsTwoZeroTwoSixPct}{3.4}
7018+\newcommand{\SecCoocAppliedCryptographyFederatedLearningTwoZeroTwoSix}{2{,}760}
7019+\newcommand{\SecCoocAppliedCryptographyFederatedLearningTwoZeroTwoSixPct}{6.3}
7020+\newcommand{\SecCoocPrivacyPreservingComputationGenerativeAiTwoZeroTwoSix}{2{,}673}
7021+\newcommand{\SecCoocPrivacyPreservingComputationGenerativeAiTwoZeroTwoSixPct}{6.4}
7022+\newcommand{\SecCoocPrivacyPreservingComputationFoundationModelsTwoZeroTwoSix}{2{,}650}
7023+\newcommand{\SecCoocPrivacyPreservingComputationFoundationModelsTwoZeroTwoSixPct}{6.4}
7024+\newcommand{\SecCoocAppliedCryptographyLlmTwoZeroTwoSix}{2{,}423}
7025+\newcommand{\SecCoocAppliedCryptographyLlmTwoZeroTwoSixPct}{5.5}
7026+\newcommand{\SecCoocFederatedLearningAiSafetyTwoZeroTwoSix}{2{,}329}
7027+\newcommand{\SecCoocFederatedLearningAiSafetyTwoZeroTwoSixPct}{13.6}
7028+\newcommand{\SecCoocFederatedLearningLlmTwoZeroTwoSix}{1{,}990}
7029+\newcommand{\SecCoocFederatedLearningLlmTwoZeroTwoSixPct}{11.7}
7030+\newcommand{\SecCoocPrivacyPreservingComputationAiAgentsTwoZeroTwoSix}{1{,}926}
7031+\newcommand{\SecCoocPrivacyPreservingComputationAiAgentsTwoZeroTwoSixPct}{4.6}
7032+\newcommand{\SecCoocBlockchainFederatedLearningTwoZeroTwoSix}{1{,}848}
7033+\newcommand{\SecCoocBlockchainFederatedLearningTwoZeroTwoSixPct}{7.4}
7034+\newcommand{\SecCoocFederatedLearningFoundationModelsTwoZeroTwoSix}{1{,}824}
7035+\newcommand{\SecCoocFederatedLearningFoundationModelsTwoZeroTwoSixPct}{10.7}
7036+\newcommand{\SecCoocAppliedCryptographyAiSafetyTwoZeroTwoSix}{1{,}792}
7037+\newcommand{\SecCoocAppliedCryptographyAiSafetyTwoZeroTwoSixPct}{4.1}
7038+\newcommand{\SecCoocAppliedCryptographyAiAgentsTwoZeroTwoSix}{1{,}341}
7039+\newcommand{\SecCoocAppliedCryptographyAiAgentsTwoZeroTwoSixPct}{3.1}
7040+\newcommand{\SecCoocAppliedCryptographyFoundationModelsTwoZeroTwoSix}{1{,}104}
7041+\newcommand{\SecCoocAppliedCryptographyFoundationModelsTwoZeroTwoSixPct}{2.5}
7042+\newcommand{\SecCoocBlockchainLlmTwoZeroTwoSix}{1{,}067}
7043+\newcommand{\SecCoocBlockchainLlmTwoZeroTwoSixPct}{4.3}
7044+\newcommand{\SecCoocFederatedLearningGenerativeAiTwoZeroTwoSix}{1{,}004}
7045+\newcommand{\SecCoocFederatedLearningGenerativeAiTwoZeroTwoSixPct}{5.9}
7046+\newcommand{\SecCoocBlockchainAiAgentsTwoZeroTwoSix}{920}
7047+\newcommand{\SecCoocBlockchainAiAgentsTwoZeroTwoSixPct}{3.7}
7048+\newcommand{\SecCoocAppliedCryptographyGenerativeAiTwoZeroTwoSix}{814}
7049+\newcommand{\SecCoocAppliedCryptographyGenerativeAiTwoZeroTwoSixPct}{1.9}
7050+\newcommand{\SecCoocBlockchainAiSafetyTwoZeroTwoSix}{749}
7051+\newcommand{\SecCoocBlockchainAiSafetyTwoZeroTwoSixPct}{3.0}
7052+\newcommand{\SecCoocFederatedLearningAiAgentsTwoZeroTwoSix}{701}
7053+\newcommand{\SecCoocFederatedLearningAiAgentsTwoZeroTwoSixPct}{4.1}
7054+\newcommand{\SecCoocBlockchainGenerativeAiTwoZeroTwoSix}{337}
7055+\newcommand{\SecCoocBlockchainGenerativeAiTwoZeroTwoSixPct}{1.4}
7056+\newcommand{\SecCoocBlockchainFoundationModelsTwoZeroTwoSix}{163}
7057+\newcommand{\SecCoocBlockchainFoundationModelsTwoZeroTwoSixPct}{0.7}
7058+\newcommand{\SecLandN}{125}
7059+\newcommand{\SecLandWithAbstract}{92}
7060+\newcommand{\SecLandWithAbstractPct}{74}
7061+\newcommand{\SecLandOaPct}{62}
7062+\newcommand{\SecLandPost}{66}
7063+\newcommand{\SecLandBaseline}{58}
7064+\newcommand{\SecLandYearTwoZeroTwoSix}{33}
7065+\newcommand{\SecLandNoDoi}{9}
7066+\newcommand{\SecLandTypePreprint}{7}
7067+\newcommand{\SecLandTypeDissertation}{13}
7068+\newcommand{\SecLandTypeBookChapter}{12}
7069+\newcommand{\SecLandTypeArticle}{69}
7070+\newcommand{\SecLandTypeConferencePaper}{11}
7071+\newcommand{\SecLandTypeReport}{9}
7072+\newcommand{\SecLandTypeReview}{3}
7073+\newcommand{\SecLandTypeBook}{1}
7074+\newcommand{\SecLandCybersecurityN}{25}
7075+\newcommand{\SecLandCybersecurityPost}{16}
7076+\newcommand{\SecLandCybersecurityWithAbstract}{20}
7077+\newcommand{\SecLandCybersecurityYearTwoZeroTwoSix}{9}
7078+\newcommand{\SecLandCybersecurityPreprints}{1}
7079+\newcommand{\SecLandCybersecurityArticles}{13}
7080+\newcommand{\SecLandCybersecurityConference}{1}
7081+\newcommand{\SecLandCybersecurityGrey}{10}
7082+\newcommand{\SecLandAppliedCryptographyN}{25}
7083+\newcommand{\SecLandAppliedCryptographyPost}{15}
7084+\newcommand{\SecLandAppliedCryptographyWithAbstract}{18}
7085+\newcommand{\SecLandAppliedCryptographyYearTwoZeroTwoSix}{8}
7086+\newcommand{\SecLandAppliedCryptographyPreprints}{0}
7087+\newcommand{\SecLandAppliedCryptographyArticles}{11}
7088+\newcommand{\SecLandAppliedCryptographyConference}{2}
7089+\newcommand{\SecLandAppliedCryptographyGrey}{12}
7090+\newcommand{\SecLandBlockchainN}{25}
7091+\newcommand{\SecLandBlockchainPost}{6}
7092+\newcommand{\SecLandBlockchainWithAbstract}{19}
7093+\newcommand{\SecLandBlockchainYearTwoZeroTwoSix}{2}
7094+\newcommand{\SecLandBlockchainPreprints}{5}
7095+\newcommand{\SecLandBlockchainArticles}{14}
7096+\newcommand{\SecLandBlockchainConference}{1}
7097+\newcommand{\SecLandBlockchainGrey}{5}
7098+\newcommand{\SecLandPrivacyPreservingComputationN}{25}
7099+\newcommand{\SecLandPrivacyPreservingComputationPost}{14}
7100+\newcommand{\SecLandPrivacyPreservingComputationWithAbstract}{21}
7101+\newcommand{\SecLandPrivacyPreservingComputationYearTwoZeroTwoSix}{8}
7102+\newcommand{\SecLandPrivacyPreservingComputationPreprints}{1}
7103+\newcommand{\SecLandPrivacyPreservingComputationArticles}{16}
7104+\newcommand{\SecLandPrivacyPreservingComputationConference}{3}
7105+\newcommand{\SecLandPrivacyPreservingComputationGrey}{5}
7106+\newcommand{\SecLandFederatedLearningN}{25}
7107+\newcommand{\SecLandFederatedLearningPost}{15}
7108+\newcommand{\SecLandFederatedLearningWithAbstract}{14}
7109+\newcommand{\SecLandFederatedLearningYearTwoZeroTwoSix}{6}
7110+\newcommand{\SecLandFederatedLearningPreprints}{0}
7111+\newcommand{\SecLandFederatedLearningArticles}{18}
7112+\newcommand{\SecLandFederatedLearningConference}{4}
7113+\newcommand{\SecLandFederatedLearningGrey}{3}
modified manuscript/tables/tab_cagr.tex +40 −40
@@ -9,45 +9,45 @@
9 \textbf{category} & \textbf{n\_2018} & \textbf{n\_2022} & \textbf{n\_2025} & \textbf{cagr\_baseline} & \textbf{cagr\_post} & \textbf{share\_2018} & \textbf{share\_2025} \\ 9 \textbf{category} & \textbf{n\_2018} & \textbf{n\_2022} & \textbf{n\_2025} & \textbf{cagr\_baseline} & \textbf{cagr\_post} & \textbf{share\_2018} & \textbf{share\_2025} \\
10 \midrule 10 \midrule
11 \endhead 11 \endhead
12−ai\_accelerators & 409 & 710 & 953 & 0.148 & 0.092 & 0.002 & 0.002 \\ 12+ai\_accelerators & 818 & 1{,}421 & 1{,}906 & 0.148 & 0.092 & 0.002 & 0.002 \\
13−ai\_agents & 2{,}718 & 2{,}260 & 4{,}409 & -0.045 & 0.329 & 0.016 & 0.010 \\ 13+ai\_agents & 5{,}436 & 4{,}521 & 8{,}819 & -0.045 & 0.329 & 0.011 & 0.010 \\
14−ai\_safety & 526 & 1{,}477 & 5{,}442 & 0.294 & 0.647 & 0.003 & 0.013 \\ 14+ai\_safety & 1{,}052 & 2{,}954 & 10{,}885 & 0.294 & 0.647 & 0.002 & 0.012 \\
15−algorithms & 9{,}716 & 9{,}400 & 11{,}265 & -0.008 & 0.088 & 0.057 & 0.026 \\ 15+algorithms & 29{,}148 & 28{,}202 & 33{,}795 & -0.008 & 0.088 & 0.058 & 0.038 \\
16−applied\_cryptography & 3{,}594 & 5{,}149 & 7{,}306 & 0.094 & 0.398 & 0.021 & 0.017 \\ 16+applied\_cryptography & 10{,}782 & 10{,}299 & 14{,}612 & -0.011 & 0.141 & 0.022 & 0.016 \\
17−autonomous\_scientific\_discovery & 43 & 118 & 382 & 0.287 & 0.382 & 0.000 & 0.001 \\ 17+autonomous\_scientific\_discovery & 86 & 236 & 764 & 0.287 & 0.382 & 0.000 & 0.001 \\
18−blockchain & 2{,}147 & 5{,}885 & 9{,}006 & 0.287 & 0.120 & 0.013 & 0.021 \\ 18+blockchain & 6{,}441 & 11{,}770 & 18{,}012 & 0.163 & 0.120 & 0.013 & 0.020 \\
19−cloud\_computing & 3{,}132 & 3{,}551 & 3{,}658 & 0.032 & -0.022 & 0.018 & 0.009 \\ 19+cloud\_computing & 9{,}396 & 7{,}103 & 10{,}974 & -0.068 & 0.198 & 0.019 & 0.012 \\
20−compilers & 241 & 321 & 383 & 0.074 & 0.053 & 0.001 & 0.001 \\ 20+compilers & 725 & 642 & 767 & -0.030 & 0.053 & 0.001 & 0.001 \\
21−computational\_biology & 4{,}322 & 9{,}768 & 14{,}077 & 0.226 & 0.170 & 0.025 & 0.033 \\ 21+computational\_biology & 12{,}968 & 19{,}536 & 28{,}154 & 0.108 & 0.170 & 0.026 & 0.032 \\
22−computational\_complexity & 1{,}279 & 1{,}770 & 2{,}104 & 0.085 & 0.343 & 0.008 & 0.005 \\ 22+computational\_complexity & 3{,}837 & 3{,}541 & 4{,}209 & -0.020 & 0.097 & 0.008 & 0.005 \\
23−computer\_architecture & 6{,}518 & 7{,}587 & 6{,}736 & 0.039 & -0.093 & 0.038 & 0.016 \\ 23+computer\_architecture & 19{,}554 & 15{,}174 & 20{,}209 & -0.061 & 0.111 & 0.039 & 0.023 \\
24−computer\_graphics & 3{,}399 & 4{,}143 & 6{,}220 & 0.051 & 0.417 & 0.020 & 0.014 \\ 24+computer\_graphics & 10{,}198 & 8{,}287 & 12{,}440 & -0.051 & 0.157 & 0.020 & 0.014 \\
25−computer\_networks & 19{,}581 & 16{,}380 & 29{,}777 & -0.044 & 0.312 & 0.115 & 0.069 \\ 25+computer\_networks & 58{,}744 & 49{,}141 & 59{,}555 & -0.044 & 0.071 & 0.118 & 0.067 \\
26−computer\_vision & 23{,}246 & 49{,}352 & 75{,}378 & 0.207 & 0.149 & 0.137 & 0.175 \\ 26+computer\_vision & 69{,}738 & 98{,}704 & 150{,}757 & 0.091 & 0.149 & 0.140 & 0.169 \\
27−cybersecurity & 9{,}813 & 15{,}777 & 29{,}522 & 0.126 & 0.558 & 0.058 & 0.069 \\ 27+cybersecurity & 29{,}440 & 31{,}554 & 59{,}044 & 0.017 & 0.272 & 0.059 & 0.066 \\
28−database\_systems & 5{,}421 & 6{,}812 & 8{,}340 & 0.059 & 0.073 & 0.032 & 0.019 \\ 28+database\_systems & 16{,}265 & 13{,}625 & 16{,}681 & -0.043 & 0.073 & 0.033 & 0.019 \\
29−deep\_learning\_architectures & 3{,}997 & 7{,}423 & 17{,}070 & 0.167 & 0.155 & 0.024 & 0.040 \\ 29+deep\_learning\_architectures & 11{,}992 & 22{,}270 & 34{,}141 & 0.167 & 0.155 & 0.024 & 0.038 \\
30−distributed\_computing & 3{,}839 & 4{,}689 & 4{,}183 & 0.051 & 0.132 & 0.023 & 0.010 \\ 30+distributed\_computing & 11{,}519 & 9{,}379 & 12{,}550 & -0.050 & 0.132 & 0.023 & 0.014 \\
31−edge\_computing & 1{,}728 & 4{,}277 & 5{,}777 & 0.254 & 0.379 & 0.010 & 0.013 \\ 31+edge\_computing & 5{,}185 & 8{,}554 & 11{,}554 & 0.133 & 0.126 & 0.010 & 0.013 \\
32−federated\_learning & 71 & 487 & 1{,}272 & 0.615 & 0.304 & 0.000 & 0.003 \\ 32+federated\_learning & 143 & 974 & 2{,}544 & 0.615 & 0.304 & 0.000 & 0.003 \\
33−formal\_verification & 1{,}769 & 1{,}997 & 2{,}311 & 0.031 & 0.258 & 0.010 & 0.005 \\ 33+formal\_verification & 5{,}307 & 3{,}995 & 4{,}623 & -0.069 & 0.027 & 0.011 & 0.005 \\
34−foundation\_models & 23 & 175 & 927 & 0.661 & 0.802 & 0.000 & 0.002 \\ 34+foundation\_models & 46 & 350 & 1{,}854 & 0.661 & 0.802 & 0.000 & 0.002 \\
35−generative\_ai & 255 & 1{,}080 & 6{,}371 & 0.434 & 0.870 & 0.002 & 0.015 \\ 35+generative\_ai & 766 & 2{,}160 & 12{,}743 & 0.296 & 0.870 & 0.002 & 0.014 \\
36−hci & 11{,}408 & 17{,}400 & 29{,}074 & 0.111 & 0.197 & 0.067 & 0.068 \\ 36+hci & 34{,}225 & 34{,}801 & 58{,}148 & 0.004 & 0.197 & 0.069 & 0.065 \\
37−information\_retrieval & 4{,}031 & 5{,}621 & 8{,}689 & 0.087 & 0.176 & 0.024 & 0.020 \\ 37+information\_retrieval & 12{,}095 & 11{,}243 & 17{,}379 & -0.018 & 0.176 & 0.024 & 0.019 \\
38−llm & 545 & 1{,}088 & 22{,}579 & 0.189 & 1.085 & 0.003 & 0.053 \\ 38+llm & 1{,}635 & 3{,}266 & 45{,}158 & 0.189 & 1.085 & 0.003 & 0.051 \\
39−ml\_theory & 4{,}511 & 10{,}065 & 16{,}399 & 0.222 & 0.197 & 0.027 & 0.038 \\ 39+ml\_theory & 13{,}534 & 20{,}131 & 32{,}798 & 0.104 & 0.197 & 0.027 & 0.037 \\
40−multimodal\_ai & 156 & 639 & 3{,}374 & 0.421 & 0.869 & 0.001 & 0.008 \\ 40+multimodal\_ai & 470 & 1{,}278 & 6{,}749 & 0.284 & 0.869 & 0.001 & 0.008 \\
41−nlp & 7{,}790 & 9{,}892 & 21{,}457 & 0.062 & 0.124 & 0.046 & 0.050 \\ 41+nlp & 23{,}372 & 29{,}678 & 42{,}914 & 0.062 & 0.124 & 0.047 & 0.048 \\
42−operating\_systems & 778 & 541 & 681 & -0.087 & 0.078 & 0.005 & 0.002 \\ 42+operating\_systems & 1{,}556 & 1{,}083 & 1{,}363 & -0.087 & 0.078 & 0.003 & 0.002 \\
43−privacy\_preserving\_computation & 1{,}035 & 2{,}188 & 4{,}408 & 0.206 & 0.327 & 0.006 & 0.010 \\ 43+privacy\_preserving\_computation & 3{,}105 & 4{,}376 & 8{,}817 & 0.090 & 0.327 & 0.006 & 0.010 \\
44−programming\_languages & 1{,}463 & 1{,}832 & 2{,}096 & 0.058 & 0.040 & 0.009 & 0.005 \\ 44+programming\_languages & 4{,}390 & 3{,}664 & 4{,}193 & -0.044 & 0.040 & 0.009 & 0.005 \\
45−quantum\_computing & 3{,}112 & 4{,}608 & 8{,}735 & 0.103 & 0.287 & 0.018 & 0.020 \\ 45+quantum\_computing & 6{,}224 & 9{,}216 & 17{,}471 & 0.103 & 0.287 & 0.012 & 0.020 \\
46−reinforcement\_learning & 1{,}648 & 3{,}508 & 6{,}605 & 0.208 & 0.252 & 0.010 & 0.015 \\ 46+reinforcement\_learning & 3{,}296 & 7{,}017 & 13{,}210 & 0.208 & 0.252 & 0.007 & 0.015 \\
47−robotics & 5{,}832 & 11{,}551 & 16{,}732 & 0.186 & 0.140 & 0.034 & 0.039 \\ 47+robotics & 17{,}496 & 23{,}103 & 33{,}464 & 0.072 & 0.140 & 0.035 & 0.037 \\
48−scientific\_computing & 5{,}627 & 6{,}644 & 7{,}730 & 0.042 & 0.066 & 0.033 & 0.018 \\ 48+scientific\_computing & 16{,}882 & 19{,}932 & 23{,}191 & 0.042 & 0.066 & 0.034 & 0.026 \\
49−software\_engineering & 9{,}672 & 11{,}862 & 18{,}190 & 0.052 & 0.177 & 0.057 & 0.042 \\ 49+software\_engineering & 29{,}016 & 23{,}725 & 36{,}381 & -0.049 & 0.177 & 0.058 & 0.041 \\
50−speech\_audio & 4{,}032 & 6{,}349 & 9{,}186 & 0.120 & 0.118 & 0.024 & 0.021 \\ 50+speech\_audio & 12{,}096 & 12{,}698 & 18{,}373 & 0.012 & 0.118 & 0.024 & 0.021 \\
51−sustainable\_computing & 291 & 276 & 819 & -0.013 & 0.233 & 0.002 & 0.002 \\ 51+sustainable\_computing & 583 & 830 & 1{,}639 & 0.092 & 0.233 & 0.001 & 0.002 \\
52 \bottomrule 52 \bottomrule
53 \end{longtable} 53 \end{longtable}
modified manuscript/tables/tab_diversity.tex +8 −8
@@ -5,14 +5,14 @@
5 \toprule 5 \toprule
6 \textbf{year} & \textbf{n} & \textbf{rao\_stirling} & \textbf{rs\_ci\_low} & \textbf{rs\_ci\_high} & \textbf{leinster\_cobbold\_q2} \\ 6 \textbf{year} & \textbf{n} & \textbf{rao\_stirling} & \textbf{rs\_ci\_low} & \textbf{rs\_ci\_high} & \textbf{leinster\_cobbold\_q2} \\
7 \midrule 7 \midrule
8−2{,}018 & 169{,}727 & 0.899 & 0.899 & 0.900 & 9.929 \\ 8+2{,}018 & 499{,}561 & 0.897 & 0.896 & 0.897 & 9.680 \\
9−2{,}019 & 180{,}248 & 0.896 & 0.896 & 0.897 & 9.625 \\ 9+2{,}019 & 539{,}741 & 0.896 & 0.896 & 0.896 & 9.594 \\
10−2{,}020 & 189{,}370 & 0.901 & 0.900 & 0.901 & 10.073 \\ 10+2{,}020 & 549{,}101 & 0.896 & 0.896 & 0.897 & 9.654 \\
11−2{,}021 & 226{,}728 & 0.902 & 0.902 & 0.903 & 10.227 \\ 11+2{,}021 & 558{,}970 & 0.894 & 0.894 & 0.894 & 9.430 \\
12−2{,}022 & 254{,}663 & 0.890 & 0.890 & 0.891 & 9.117 \\ 12+2{,}022 & 560{,}433 & 0.891 & 0.891 & 0.892 & 9.214 \\
13−2{,}023 & 286{,}460 & 0.886 & 0.885 & 0.886 & 8.734 \\ 13+2{,}023 & 634{,}466 & 0.891 & 0.890 & 0.891 & 9.140 \\
14−2{,}024 & 363{,}470 & 0.888 & 0.888 & 0.889 & 8.959 \\ 14+2{,}024 & 739{,}619 & 0.890 & 0.890 & 0.890 & 9.095 \\
15−2{,}025 & 429{,}633 & 0.889 & 0.889 & 0.890 & 9.046 \\ 15+2{,}025 & 892{,}840 & 0.893 & 0.893 & 0.893 & 9.356 \\
16 \bottomrule 16 \bottomrule
17 \end{tabular} 17 \end{tabular}
18 \end{threeparttable}\end{table} 18 \end{threeparttable}\end{table}
added manuscript/tables/tab_p9_ai_shares.tex +21 −0
@@ -0,0 +1,21 @@
1+% generated by tables/make_tables.py from ./results/bibliometrics/category_share_ci_yearly.csv
2+\begin{table}[htbp]\centering\caption{Share of primary-category assignments held by the AI and machine-learning categories in 2018, 2022 and 2025 (per cent, bootstrap 95\% intervals).}\label{tab:p9-ai-shares}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrrr}
5+\toprule
6+\textbf{Category} & \textbf{2018} & \textbf{2022} & \textbf{2025} \\
7+\midrule
8+Large Language Models & 0.33 [0.31, 0.34] & 0.58 [0.56, 0.60] & 5.06 [5.01, 5.10] \\
9+Foundation Models & 0.01 [0.01, 0.01] & 0.06 [0.06, 0.07] & 0.21 [0.20, 0.22] \\
10+Multimodal AI & 0.09 [0.09, 0.10] & 0.23 [0.22, 0.24] & 0.76 [0.74, 0.77] \\
11+Generative AI & 0.15 [0.14, 0.16] & 0.39 [0.37, 0.40] & 1.43 [1.40, 1.45] \\
12+AI Agents and Autonomous Systems & 1.09 [1.06, 1.11] & 0.81 [0.78, 0.83] & 0.99 [0.97, 1.01] \\
13+Reinforcement Learning & 0.66 [0.64, 0.68] & 1.25 [1.22, 1.28] & 1.48 [1.46, 1.50] \\
14+Machine Learning Theory & 2.71 [2.67, 2.76] & 3.59 [3.54, 3.64] & 3.67 [3.63, 3.71] \\
15+Deep Learning Architectures & 2.40 [2.36, 2.44] & 3.97 [3.93, 4.02] & 3.82 [3.79, 3.86] \\
16+AI Safety and Alignment & 0.21 [0.20, 0.22] & 0.53 [0.51, 0.55] & 1.22 [1.19, 1.24] \\
17+AI \\& ML area (sum) & 7.65 & 11.41 & 18.63 \\
18+\bottomrule
19+\end{tabular}
20+\begin{tablenotes}\footnotesize\item Shares are of the year's works with a primary category; intervals from 500 multinomial resamples of the works (seed 44). The area row is the sum of the category shares (no interval). Categories 1--8 and 38 of the mandate.\end{tablenotes}
21+\end{threeparttable}\end{table}
added manuscript/tables/tab_p9_altdates.tex +54 −0
@@ -0,0 +1,54 @@
1+% generated by tables/make_tables.py from ./results/econometrics/its_results.json
2+\begin{longtable}{lrrrrr}
3+\caption{ITS level change (HAC $p$) by category at the five pre-specified event dates.}\label{tab:p9-altdates}\\
4+\toprule
5+\textbf{Category} & \textbf{GPT-3} & \textbf{DALL$\cdot$E 2} & \textbf{ChatGPT} & \textbf{GPT-4} & \textbf{o1} \\
6+\midrule
7+\endfirsthead
8+\toprule
9+\textbf{Category} & \textbf{GPT-3} & \textbf{DALL$\cdot$E 2} & \textbf{ChatGPT} & \textbf{GPT-4} & \textbf{o1} \\
10+\midrule
11+\endhead
12+AI Agents and Autonomous Systems & -0.359 (0.029) & -0.320 (0.003) & -0.210 (0.028) & -0.151 (0.090) & +0.226 (0.003) \\
13+AI Safety and Alignment & -0.226 (0.034) & -0.311 (0.000) & -0.206 (0.002) & -0.080 (0.244) & +0.150 (0.010) \\
14+Accelerators and AI Hardware & -0.099 (0.273) & -0.137 (0.027) & -0.113 (0.027) & -0.043 (0.455) & +0.093 (0.016) \\
15+Algorithms and Data Structures & -0.060 (0.209) & -0.076 (0.010) & -0.045 (0.039) & -0.020 (0.410) & +0.092 (0.002) \\
16+Applied Cryptography & -0.101 (0.132) & -0.052 (0.052) & +0.002 (0.922) & +0.035 (0.082) & +0.142 (0.002) \\
17+Autonomous Scientific Discovery & -0.109 (0.170) & -0.294 (0.000) & -0.155 (0.038) & -0.087 (0.239) & +0.011 (0.793) \\
18+Blockchain Systems & -0.065 (0.362) & -0.023 (0.654) & -0.008 (0.858) & -0.024 (0.571) & +0.007 (0.842) \\
19+Cloud Computing & -0.114 (0.155) & -0.063 (0.155) & +0.003 (0.936) & +0.010 (0.785) & +0.192 (0.002) \\
20+Compilers & -0.098 (0.158) & -0.098 (0.015) & -0.028 (0.471) & +0.003 (0.937) & +0.090 (0.086) \\
21+Computational Biology & -0.012 (0.750) & -0.141 (0.000) & -0.116 (0.000) & -0.084 (0.012) & +0.014 (0.487) \\
22+Computational Complexity & -0.133 (0.043) & -0.116 (0.000) & -0.068 (0.000) & -0.016 (0.438) & +0.173 (0.000) \\
23+Computer Architecture & -0.084 (0.294) & -0.057 (0.151) & -0.009 (0.771) & +0.027 (0.385) & +0.141 (0.003) \\
24+Computer Graphics & -0.119 (0.056) & -0.046 (0.026) & +0.043 (0.090) & +0.099 (0.000) & +0.129 (0.024) \\
25+Computer Networks & -0.045 (0.393) & +0.004 (0.835) & +0.043 (0.021) & +0.037 (0.241) & +0.125 (0.001) \\
26+Computer Vision & -0.043 (0.442) & -0.038 (0.104) & +0.024 (0.305) & +0.057 (0.019) & +0.092 (0.013) \\
27+Cybersecurity & -0.113 (0.121) & -0.083 (0.015) & -0.021 (0.408) & +0.010 (0.713) & +0.149 (0.001) \\
28+Database Systems & -0.107 (0.090) & -0.041 (0.275) & -0.013 (0.615) & +0.019 (0.434) & +0.129 (0.005) \\
29+Deep Learning Architectures & -0.004 (0.937) & -0.114 (0.008) & -0.071 (0.126) & -0.051 (0.260) & +0.008 (0.792) \\
30+Distributed Computing & -0.153 (0.052) & -0.108 (0.008) & -0.034 (0.282) & +0.001 (0.976) & +0.191 (0.000) \\
31+Edge Computing & -0.011 (0.861) & -0.069 (0.037) & -0.044 (0.139) & -0.044 (0.154) & +0.083 (0.027) \\
32+Federated Learning & +0.166 (0.002) & -0.101 (0.067) & -0.126 (0.007) & -0.092 (0.075) & -0.115 (0.010) \\
33+Formal Verification & -0.128 (0.138) & -0.174 (0.003) & -0.109 (0.029) & -0.056 (0.303) & +0.063 (0.246) \\
34+Foundation Models & -0.164 (0.049) & -0.221 (0.002) & -0.075 (0.340) & +0.023 (0.756) & +0.084 (0.173) \\
35+Generative AI & -0.274 (0.028) & -0.173 (0.007) & +0.036 (0.556) & +0.150 (0.017) & +0.324 (0.000) \\
36+Human-Computer Interaction & -0.156 (0.040) & -0.041 (0.043) & +0.049 (0.027) & +0.101 (0.000) & +0.191 (0.004) \\
37+Information Retrieval & -0.177 (0.080) & -0.167 (0.001) & -0.094 (0.019) & -0.034 (0.378) & +0.229 (0.000) \\
38+Large Language Models & -0.380 (0.007) & -0.182 (0.002) & +0.106 (0.075) & +0.236 (0.002) & +0.457 (0.001) \\
39+Machine Learning Theory & -0.179 (0.020) & -0.162 (0.001) & -0.080 (0.090) & -0.047 (0.296) & +0.081 (0.026) \\
40+Multimodal AI & -0.258 (0.016) & -0.129 (0.004) & +0.025 (0.600) & +0.150 (0.002) & +0.330 (0.001) \\
41+Natural Language Processing & -0.204 (0.015) & -0.127 (0.002) & +0.016 (0.694) & +0.080 (0.105) & +0.228 (0.003) \\
42+Operating Systems & -0.139 (0.047) & -0.062 (0.208) & -0.033 (0.417) & -0.036 (0.350) & +0.101 (0.019) \\
43+Privacy-Preserving Computation & -0.044 (0.469) & -0.073 (0.037) & -0.001 (0.965) & +0.042 (0.200) & +0.198 (0.001) \\
44+Programming Languages & -0.073 (0.281) & -0.083 (0.061) & -0.040 (0.381) & -0.030 (0.568) & +0.000 (0.996) \\
45+Quantum Computing & -0.170 (0.024) & -0.134 (0.021) & -0.087 (0.039) & -0.025 (0.582) & +0.150 (0.000) \\
46+Reinforcement Learning & -0.148 (0.069) & -0.216 (0.002) & -0.164 (0.013) & -0.158 (0.006) & +0.054 (0.071) \\
47+Robotics & -0.062 (0.237) & -0.051 (0.064) & +0.012 (0.660) & +0.034 (0.256) & +0.037 (0.326) \\
48+Scientific Computing & -0.027 (0.548) & -0.086 (0.001) & -0.047 (0.054) & -0.036 (0.109) & +0.034 (0.094) \\
49+Software Engineering & -0.097 (0.308) & -0.034 (0.279) & +0.048 (0.124) & +0.102 (0.003) & +0.157 (0.017) \\
50+Speech and Audio Computing & -0.119 (0.038) & -0.074 (0.039) & +0.036 (0.354) & +0.069 (0.151) & +0.049 (0.413) \\
51+Sustainable Computing & -0.089 (0.256) & -0.020 (0.455) & +0.049 (0.048) & +0.056 (0.106) & +0.222 (0.000) \\
52+\bottomrule
53+\multicolumn{6}{p{0.95\linewidth}}{\footnotesize Each column re-estimates equation~\eqref{eq:its} with the event at the stated month; entries are the log level change with the HAC $p$-value in parentheses. The five estimates of a row are not independent: they are the same series with the break moved.}\\
54+\end{longtable}
added manuscript/tables/tab_p9_area_flow_change.tex +19 −0
@@ -0,0 +1,19 @@
1+% generated by tables/make_tables.py from ./results/graph/area_flows_yearly.csv
2+\begin{table}[htbp]\centering\caption{Change in the area citation-flow matrix between 2018--2022 and 2023--2026: post-period row share minus baseline row share (percentage points).}\label{tab:p9-area-flow-change}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrrrrrrrr}
5+\toprule
6+\textbf{Citing area} & \textbf{AI \\& ML} & \textbf{Perception \\& interaction} & \textbf{Software \\& PL} & \textbf{Theory} & \textbf{Data \\& distributed} & \textbf{Security \\& privacy} & \textbf{Hardware \\& emerging} & \textbf{Computational science} \\
7+\midrule
8+AI \\& ML & -3.1 & +3.1 & +0.2 & -0.9 & +0.1 & +0.5 & -0.2 & +0.2 \\
9+Perception \\& interaction & +0.4 & +0.2 & -0.1 & -0.0 & -0.4 & +0.0 & -0.0 & +0.0 \\
10+Software \\& PL & +2.6 & +1.2 & -2.8 & -0.6 & -1.0 & +0.7 & -0.0 & -0.2 \\
11+Theory & +1.1 & +1.1 & -0.5 & -1.0 & -0.6 & -0.0 & +0.2 & -0.2 \\
12+Data \\& distributed & +1.5 & +0.7 & -0.0 & -0.1 & -3.0 & +1.0 & +0.1 & -0.1 \\
13+Security \\& privacy & +1.8 & +0.9 & +0.0 & -0.1 & +0.1 & -2.6 & -0.1 & +0.0 \\
14+Hardware \\& emerging & +0.4 & -0.2 & -0.2 & +0.0 & +0.1 & +0.1 & -0.2 & -0.0 \\
15+Computational science & +2.5 & +0.9 & -0.1 & -0.7 & -0.7 & +0.0 & -0.1 & -1.8 \\
16+\bottomrule
17+\end{tabular}
18+\begin{tablenotes}\footnotesize\item Rows: citing area; columns: cited area; row shares of internal references with both ends classified, split by citing calendar year (2022 entirely in the first span). Figure~\ref{fig:p9-area-flow-matrix} shows the two underlying matrices.\end{tablenotes}
19+\end{threeparttable}\end{table}
added manuscript/tables/tab_p9_area_shares.tex +22 −0
@@ -0,0 +1,22 @@
1+% generated by tables/make_tables.py from ./results/graph/category_counts_yearly.csv
2+\begin{table}[htbp]\centering\caption{Share of primary-category assignments by top-level area and year (per cent), with the two protocol periods.}\label{tab:p9-area-shares}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrrrrrrrr}
5+\toprule
6+\textbf{Year / period} & \textbf{AI \\& ML} & \textbf{Perception \\& interaction} & \textbf{Software \\& PL} & \textbf{Theory} & \textbf{Data \\& distributed} & \textbf{Security \\& privacy} & \textbf{Hardware \\& emerging} & \textbf{Computational science} \\
7+\midrule
8+2018 & 7.65 & 33.45 & 7.89 & 6.60 & 22.66 & 9.99 & 5.75 & 5.99 \\
9+2019 & 8.89 & 34.69 & 7.21 & 6.15 & 21.14 & 10.27 & 5.48 & 6.17 \\
10+2020 & 10.02 & 34.85 & 6.77 & 6.05 & 19.56 & 10.59 & 5.24 & 6.92 \\
11+2021 & 10.94 & 35.95 & 6.06 & 5.87 & 18.47 & 10.42 & 5.18 & 7.11 \\
12+2022 & 11.41 & 36.98 & 5.71 & 5.66 & 17.67 & 10.52 & 4.95 & 7.08 \\
13+2023 & 13.00 & 37.60 & 5.55 & 5.05 & 16.63 & 10.81 & 4.85 & 6.53 \\
14+2024 & 15.44 & 37.48 & 5.32 & 4.57 & 15.27 & 11.10 & 4.61 & 6.20 \\
15+2025 & 18.63 & 35.40 & 5.15 & 4.26 & 14.41 & 11.54 & 4.77 & 5.84 \\
16+2026 & 22.09 & 32.86 & 5.63 & 4.26 & 13.08 & 11.35 & 5.09 & 5.63 \\
17+Baseline (2018-01 to 2022-11) & 9.84 & 35.23 & 6.71 & 6.10 & 19.80 & 10.33 & 5.32 & 6.66 \\
18+Post (2022-12 to 2026-11) & 17.46 & 35.72 & 5.41 & 4.47 & 14.84 & 11.26 & 4.82 & 6.02 \\
19+\bottomrule
20+\end{tabular}
21+\begin{tablenotes}\footnotesize\item Shares of works with a primary category (Table~\ref{tab:p9-classification}); rows sum to 100. The period rows split at the reference date; the year rows split at the calendar year. 2026 is incomplete.\end{tablenotes}
22+\end{threeparttable}\end{table}
added manuscript/tables/tab_p9_category_periods.tex +54 −0
@@ -0,0 +1,54 @@
1+% generated by tables/make_tables.py from ./results/graph/category_counts_yearly.csv, ./results/bibliometrics/cagr_by_category.csv
2+\begin{longtable}{lrrrrrrrr}
3+\caption{The forty categories by protocol period: primary-category output, composition shares, multi-label growth and compound annual growth rates.}\label{tab:p9-category-periods}\\
4+\toprule
5+\textbf{Category} & \textbf{Area} & \textbf{Primary, base} & \textbf{Primary, post} & \textbf{Share base (\%)} & \textbf{Share post (\%)} & \textbf{Multi-label post/base} & \textbf{CAGR base} & \textbf{CAGR post} \\
6+\midrule
7+\endfirsthead
8+\toprule
9+\textbf{Category} & \textbf{Area} & \textbf{Primary, base} & \textbf{Primary, post} & \textbf{Share base (\%)} & \textbf{Share post (\%)} & \textbf{Multi-label post/base} & \textbf{CAGR base} & \textbf{CAGR post} \\
10+\midrule
11+\endhead
12+Computer Vision & Perception \\& interaction & 421{,}385 & 521{,}935 & 15.79 & 16.97 & 1.30 & 0.091 & 0.149 \\
13+Computer Networks & Data \\& distributed & 267{,}896 & 212{,}342 & 10.04 & 6.90 & 0.94 & -0.044 & 0.071 \\
14+Human-Computer Interaction & Perception \\& interaction & 174{,}138 & 196{,}794 & 6.53 & 6.40 & 1.26 & 0.004 & 0.197 \\
15+Cybersecurity & Security \\& privacy & 154{,}197 & 195{,}498 & 5.78 & 6.36 & 1.26 & 0.017 & 0.272 \\
16+Natural Language Processing & Perception \\& interaction & 134{,}844 & 152{,}754 & 5.05 & 4.97 & 1.75 & 0.062 & 0.124 \\
17+Large Language Models & AI \\& ML & 11{,}813 & 132{,}127 & 0.44 & 4.30 & 3.49 & 0.189 & 1.085 \\
18+Software Engineering & Software \\& PL & 132{,}205 & 129{,}300 & 4.95 & 4.20 & 1.32 & -0.049 & 0.177 \\
19+Algorithms and Data Structures & Theory & 144{,}105 & 121{,}771 & 5.40 & 3.96 & 0.99 & -0.008 & 0.088 \\
20+Robotics & Perception \\& interaction & 101{,}643 & 118{,}241 & 3.81 & 3.84 & 1.28 & 0.072 & 0.140 \\
21+Deep Learning Architectures & AI \\& ML & 87{,}481 & 116{,}113 & 3.28 & 3.77 & 1.57 & 0.167 & 0.155 \\
22+Machine Learning Theory & AI \\& ML & 88{,}923 & 112{,}830 & 3.33 & 3.67 & 1.51 & 0.104 & 0.197 \\
23+Computational Biology & Computational science & 82{,}846 & 97{,}856 & 3.10 & 3.18 & 1.31 & 0.108 & 0.170 \\
24+Scientific Computing & Computational science & 93{,}944 & 84{,}877 & 3.52 & 2.76 & 1.10 & 0.042 & 0.066 \\
25+Computer Architecture & Hardware \\& emerging & 86{,}471 & 70{,}512 & 3.24 & 2.29 & 0.98 & -0.061 & 0.111 \\
26+Speech and Audio Computing & Perception \\& interaction & 62{,}133 & 65{,}819 & 2.33 & 2.14 & 1.14 & 0.012 & 0.118 \\
27+Blockchain Systems & Security \\& privacy & 46{,}516 & 62{,}087 & 1.74 & 2.02 & 1.40 & 0.163 & 0.120 \\
28+Information Retrieval & Data \\& distributed & 58{,}198 & 61{,}738 & 2.18 & 2.01 & 1.34 & -0.018 & 0.176 \\
29+Database Systems & Data \\& distributed & 74{,}552 & 61{,}582 & 2.79 & 2.00 & 1.02 & -0.043 & 0.073 \\
30+Quantum Computing & Hardware \\& emerging & 39{,}846 & 60{,}417 & 1.49 & 1.96 & 1.59 & 0.103 & 0.287 \\
31+Applied Cryptography & Security \\& privacy & 53{,}877 & 51{,}175 & 2.02 & 1.66 & 1.13 & -0.011 & 0.141 \\
32+Reinforcement Learning & AI \\& ML & 26{,}913 & 44{,}477 & 1.01 & 1.45 & 2.08 & 0.208 & 0.252 \\
33+Distributed Computing & Data \\& distributed & 52{,}661 & 43{,}642 & 1.97 & 1.42 & 1.22 & -0.050 & 0.132 \\
34+Computer Graphics & Perception \\& interaction & 46{,}021 & 43{,}270 & 1.72 & 1.41 & 1.20 & -0.051 & 0.157 \\
35+Edge Computing & Data \\& distributed & 34{,}568 & 40{,}445 & 1.30 & 1.31 & 1.28 & 0.133 & 0.126 \\
36+Generative AI & AI \\& ML & 7{,}503 & 38{,}194 & 0.28 & 1.24 & 3.44 & 0.296 & 0.870 \\
37+Cloud Computing & Data \\& distributed & 40{,}415 & 36{,}757 & 1.51 & 1.19 & 1.02 & -0.068 & 0.198 \\
38+AI Safety and Alignment & AI \\& ML & 9{,}997 & 33{,}792 & 0.37 & 1.10 & 3.71 & 0.294 & 0.647 \\
39+AI Agents and Autonomous Systems & AI \\& ML & 25{,}167 & 32{,}219 & 0.94 & 1.05 & 2.00 & -0.045 & 0.329 \\
40+Privacy-Preserving Computation & Security \\& privacy & 18{,}816 & 28{,}450 & 0.71 & 0.92 & 1.99 & 0.090 & 0.327 \\
41+Multimodal AI & AI \\& ML & 4{,}007 & 20{,}970 & 0.15 & 0.68 & 3.75 & 0.284 & 0.869 \\
42+Formal Verification & Software \\& PL & 22{,}992 & 18{,}225 & 0.86 & 0.59 & 0.91 & -0.069 & 0.027 \\
43+Programming Languages & Software \\& PL & 20{,}392 & 15{,}837 & 0.76 & 0.51 & 0.98 & -0.044 & 0.040 \\
44+Computational Complexity & Theory & 18{,}778 & 15{,}760 & 0.70 & 0.51 & 1.09 & -0.020 & 0.097 \\
45+Federated Learning & Security \\& privacy & 2{,}229 & 9{,}011 & 0.08 & 0.29 & 3.69 & 0.615 & 0.304 \\
46+Accelerators and AI Hardware & Hardware \\& emerging & 5{,}520 & 6{,}917 & 0.21 & 0.22 & 1.51 & 0.148 & 0.092 \\
47+Foundation Models & AI \\& ML & 838 & 6{,}186 & 0.03 & 0.20 & 3.12 & 0.661 & 0.802 \\
48+Sustainable Computing & Hardware \\& emerging & 3{,}356 & 5{,}389 & 0.13 & 0.18 & 1.37 & 0.092 & 0.233 \\
49+Operating Systems & Hardware \\& emerging & 6{,}877 & 5{,}158 & 0.26 & 0.17 & 0.96 & -0.087 & 0.078 \\
50+Compilers & Software \\& PL & 3{,}450 & 2{,}900 & 0.13 & 0.09 & 1.20 & -0.030 & 0.053 \\
51+Autonomous Scientific Discovery & Computational science & 841 & 2{,}539 & 0.03 & 0.08 & 2.50 & 0.287 & 0.382 \\
52+\bottomrule
53+\multicolumn{9}{p{0.95\linewidth}}{\footnotesize Primary counts and shares use the primary category; the multi-label ratio is the post-period count of all assignments to the category divided by the baseline count (periods split at the reference date). CAGR base: 2018 to 2022 (four years); CAGR post: 2023 to 2025 (two years), both on primary counts of complete calendar years (equation~\eqref{eq:cagr}). Sorted by post-period share.}\\
54+\end{longtable}
added manuscript/tables/tab_p9_classification.tex +20 −0
@@ -0,0 +1,20 @@
1+% generated by tables/make_tables.py from ./results/graph/category_counts_yearly.csv, ./data/metadata/prisma_counts.json
2+\begin{table}[htbp]\centering\caption{Classification coverage by publication year: eligible works, works with a primary category, and multi-label assignments.}\label{tab:p9-classification}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrrrrr}
5+\toprule
6+\textbf{Year} & \textbf{Eligible works} & \textbf{With primary category} & \textbf{Coverage (\%)} & \textbf{Category assignments} & \textbf{Labels per classified work} \\
7+\midrule
8+2018 & 642{,}849 & 499{,}561 & 77.7 & 1{,}014{,}269 & 2.03 \\
9+2019 & 695{,}369 & 539{,}741 & 77.6 & 1{,}127{,}407 & 2.09 \\
10+2020 & 709{,}083 & 549{,}101 & 77.4 & 1{,}183{,}991 & 2.16 \\
11+2021 & 711{,}244 & 558{,}970 & 78.6 & 1{,}247{,}143 & 2.23 \\
12+2022 & 699{,}548 & 560{,}433 & 80.1 & 1{,}285{,}941 & 2.29 \\
13+2023 & 785{,}194 & 634{,}466 & 80.8 & 1{,}506{,}835 & 2.37 \\
14+2024 & 901{,}654 & 739{,}619 & 82.0 & 1{,}834{,}511 & 2.48 \\
15+2025 & 1{,}065{,}828 & 892{,}840 & 83.8 & 2{,}391{,}547 & 2.68 \\
16+2026 & 901{,}651 & 771{,}301 & 85.5 & 2{,}153{,}220 & 2.79 \\
17+\bottomrule
18+\end{tabular}
19+\begin{tablenotes}\footnotesize\item A work receives a primary category when at least one category score passes the threshold of Part IV; works without any assignment (no abstract, unmapped venue and no arXiv category) have no primary category and enter no category series. Assignments count every (work, category) pair. 2026 is incomplete.\end{tablenotes}
20+\end{threeparttable}\end{table}
added manuscript/tables/tab_p9_cooccurrence.tex +26 −0
@@ -0,0 +1,26 @@
1+% generated by tables/make_tables.py from ./results/graph/category_cooccurrence_yearly.csv
2+\begin{table}[htbp]\centering\caption{The fifteen most frequent co-classification pairs (works carrying both categories in their label set), 2018--2022 and 2023--2026.}\label{tab:p9-cooccurrence}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrrrr}
5+\toprule
6+\textbf{Rank} & \textbf{Pair, 2018--2022} & \textbf{Works} & \textbf{Pair, 2023--2026} & \textbf{Works } \\
7+\midrule
8+1 & Computer Vision / Deep Learning Architectures & 195{,}746 & Computer Vision / Deep Learning Architectures & 298{,}635 \\
9+2 & Computer Networks / Cybersecurity & 105{,}619 & Large Language Models / Natural Language Processing & 233{,}717 \\
10+3 & Large Language Models / Natural Language Processing & 96{,}346 & Computer Networks / Cybersecurity & 126{,}739 \\
11+4 & Computer Networks / Edge Computing & 92{,}287 & Deep Learning Architectures / Machine Learning Theory & 117{,}534 \\
12+5 & Applied Cryptography / Cybersecurity & 88{,}353 & Applied Cryptography / Cybersecurity & 101{,}827 \\
13+6 & Deep Learning Architectures / Machine Learning Theory & 78{,}318 & Computer Networks / Edge Computing & 100{,}861 \\
14+7 & Computer Architecture / Distributed Computing & 64{,}532 & Computer Vision / Foundation Models & 96{,}417 \\
15+8 & Computer Vision / Human-Computer Interaction & 58{,}842 & Computer Vision / Multimodal AI & 94{,}608 \\
16+9 & Computer Networks / Distributed Computing & 54{,}996 & Deep Learning Architectures / Foundation Models & 93{,}935 \\
17+10 & Cloud Computing / Distributed Computing & 53{,}634 & Computer Vision / Generative AI & 89{,}552 \\
18+11 & Computer Vision / Cybersecurity & 53{,}447 & Foundation Models / Large Language Models & 87{,}013 \\
19+12 & Computer Vision / Machine Learning Theory & 44{,}326 & Deep Learning Architectures / Natural Language Processing & 75{,}917 \\
20+13 & Applied Cryptography / Computer Networks & 43{,}265 & Deep Learning Architectures / Large Language Models & 73{,}479 \\
21+14 & Cloud Computing / Computer Networks & 43{,}168 & Computer Vision / Human-Computer Interaction & 70{,}536 \\
22+15 & Algorithms and Data Structures / Computational Complexity & 42{,}821 & Foundation Models / Natural Language Processing & 70{,}076 \\
23+\bottomrule
24+\end{tabular}
25+\begin{tablenotes}\footnotesize\item Counts of works with both labels, summed over the calendar years of each span (2022 in the first); the post span is incomplete. Co-classification depends on abstract availability and vocabulary (Part IV), so the pairs describe the classifier's view of the literature.\end{tablenotes}
26+\end{threeparttable}\end{table}
added manuscript/tables/tab_p9_cross_area.tex +22 −0
@@ -0,0 +1,22 @@
1+% generated by tables/make_tables.py from ./results/graph/cross_area_share_yearly.csv
2+\begin{table}[htbp]\centering\caption{Cross-area share of internal references by citing area and citing year (per cent), with the aggregate share and its composition-adjusted counterpart.}\label{tab:p9-cross-area}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrrrrrrrrr}
5+\toprule
6+\textbf{Citing area} & \textbf{2018} & \textbf{2019} & \textbf{2020} & \textbf{2021} & \textbf{2022} & \textbf{2023} & \textbf{2024} & \textbf{2025} & \textbf{2026} \\
7+\midrule
8+AI \\& ML & 46.6 & 48.1 & 49.3 & 51.0 & 55.1 & 55.3 & 54.9 & 55.4 & 53.9 \\
9+Computational science & 31.9 & 33.4 & 31.6 & 31.4 & 32.3 & 33.8 & 34.8 & 34.6 & 31.3 \\
10+Data \\& distributed & 25.8 & 24.2 & 25.1 & 26.7 & 27.1 & 28.1 & 28.9 & 29.7 & 30.8 \\
11+Hardware \\& emerging & 30.2 & 29.0 & 29.2 & 28.5 & 28.0 & 28.8 & 28.7 & 28.1 & 29.6 \\
12+Perception \\& interaction & 20.5 & 16.2 & 15.7 & 15.8 & 15.2 & 15.3 & 15.2 & 15.6 & 16.3 \\
13+Security \\& privacy & 32.6 & 34.4 & 35.8 & 36.7 & 36.7 & 37.3 & 38.8 & 39.6 & 39.9 \\
14+Software \\& PL & 29.9 & 30.2 & 32.8 & 33.5 & 33.8 & 35.3 & 35.8 & 36.3 & 35.5 \\
15+Theory & 30.0 & 35.1 & 36.1 & 35.9 & 34.7 & 35.4 & 35.8 & 36.5 & 37.8 \\
16+All areas, raw & 27.8 & 25.9 & 26.3 & 26.9 & 26.8 & 27.4 & 28.2 & 29.5 & 30.5 \\
17+All areas, baseline area weights & 27.4 & 25.9 & 26.1 & 26.8 & 27.1 & 27.6 & 27.8 & 28.3 & 28.6 \\
18+Internal references (count) & 213{,}358.0 & 975{,}445.0 & 2{,}131{,}658.0 & 3{,}556{,}446.0 & 4{,}288{,}984.0 & 5{,}706{,}569.0 & 6{,}799{,}414.0 & 7{,}477{,}998.0 & 4{,}558{,}848.0 \\
19+\bottomrule
20+\end{tabular}
21+\begin{tablenotes}\footnotesize\item Equation~\eqref{eq:cross-area}: the share of an area's resolved internal references whose cited work has a primary category in another area. The composition-adjusted aggregate fixes each area's weight at its 2018--2022 share of internal references, so that it moves only when the areas' own cross-area rates move. The last row is the number of internal references with both ends classified. 2026 is incomplete.\end{tablenotes}
22+\end{threeparttable}\end{table}
added manuscript/tables/tab_p9_emerging_cutoffs.tex +19 −0
@@ -0,0 +1,19 @@
1+% generated by tables/make_tables.py from ./results/emerging/metrics_h12.json, ./results/emerging/metrics_h24.json
2+\begin{table}[htbp]\centering\caption{AUC of the three detectors at each test cutoff, by horizon.}\label{tab:p9-emerging-cutoffs}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrrrrrr}
5+\toprule
6+\textbf{Test cutoff} & \textbf{h12: z-score} & \textbf{h12: logistic} & \textbf{h12: GBM} & \textbf{h24: z-score} & \textbf{h24: logistic} & \textbf{h24: GBM} \\
7+\midrule
8+2023-01 & 0.558 & 0.463 & 0.639 & 0.773 & 0.596 & 0.525 \\
9+2023-04 & 0.722 & 0.487 & 0.623 & 0.835 & 0.589 & 0.534 \\
10+2023-07 & 0.679 & 0.401 & 0.800 & 0.870 & 0.715 & 0.727 \\
11+2023-10 & 0.687 & 0.487 & 0.656 & 0.832 & 0.696 & 0.743 \\
12+2024-01 & 0.997 & 0.025 & 0.996 & 0.764 & 0.710 & 0.680 \\
13+2024-04 & 0.618 & 0.568 & 0.843 & 0.894 & 0.738 & 0.728 \\
14+2024-07 & 0.646 & 0.538 & 0.756 & & & \\
15+2024-10 & 0.709 & 0.516 & 0.772 & & & \\
16+\bottomrule
17+\end{tabular}
18+\begin{tablenotes}\footnotesize\item Within-cutoff AUC on the eligible topics of the cutoff; blank where the label window of the cutoff extends beyond the panel. Pooled values are in Tables~\ref{tab:emerging_h12} and~\ref{tab:emerging_h24}.\end{tablenotes}
19+\end{threeparttable}\end{table}
added manuscript/tables/tab_p9_emerging_setup.tex +13 −0
@@ -0,0 +1,13 @@
1+% generated by tables/make_tables.py from ./results/emerging/metrics_h12.json, ./results/emerging/metrics_h24.json
2+\begin{table}[htbp]\centering\caption{Out-of-time evaluation design of the emerging-topic detector: cutoffs, panel sizes and positive rates.}\label{tab:p9-emerging-setup}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrrrrrr}
5+\toprule
6+\textbf{Horizon (months)} & \textbf{Training cutoffs} & \textbf{Test cutoffs} & \textbf{Train rows} & \textbf{Test rows} & \textbf{Positive rate, train} & \textbf{Positive rate, test} \\
7+\midrule
8+12 & 2020-01 to 2021-10 (8) & 2023-01 to 2024-10 (8) & 9{,}202 & 9{,}626 & 0.0083 & 0.0047 \\
9+24 & 2020-01 to 2020-10 (4) & 2023-01 to 2024-04 (6) & 4{,}487 & 7{,}177 & 0.0254 & 0.0145 \\
10+\bottomrule
11+\end{tabular}
12+\begin{tablenotes}\footnotesize\item Rows are eligible (topic, cutoff) pairs; the label of equation~\eqref{eq:label} requires top-decile future growth and a doubling of volume, so the positive rate is below ten per cent.\end{tablenotes}
13+\end{threeparttable}\end{table}
added manuscript/tables/tab_p9_h3.tex +19 −0
@@ -0,0 +1,19 @@
1+% generated by tables/make_tables.py from ./results/econometrics/its_results.json
2+\begin{table}[htbp]\centering\caption{Hypothesis H3: break dates and ITS estimates for the eight categories named in the protocol.}\label{tab:p9-h3}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrrrrrrrr}
5+\toprule
6+\textbf{Category} & \textbf{H3 prediction} & \textbf{Bai--Perron (BIC)} & \textbf{PELT} & \textbf{In window} & \textbf{Level change (p)} & \textbf{Slope change (p)} & \textbf{Placebo p (slope)} & \textbf{BOCPD P} \\
7+\midrule
8+Large Language Models & break expected & 2022-07 & 2023-10 & no & +0.106 (0.075) & +0.0336 (1.1e-60) & 0.200 & 0.093 \\
9+Generative AI & break expected & 2021-11 & 2020-12, 2023-12 & no & +0.036 (0.556) & +0.0194 (1e-15) & 0.308 & 0.020 \\
10+AI Agents and Autonomous Systems & break expected & 2022-07, 2024-07 & 2023-12, 2025-08 & no & -0.210 (0.028) & +0.0401 (3.6e-18) & 0.400 & 0.123 \\
11+AI Safety and Alignment & break expected & 2021-06, 2022-09 & 2020-02, 2024-02 & no & -0.206 (0.002) & +0.0164 (3e-13) & 0.308 & 0.017 \\
12+Formal Verification & no break expected & 2022-02, 2024-07 & 2024-03, 2025-09 & no & -0.109 (0.029) & +0.0177 (2.5e-15) & 0.246 & 0.048 \\
13+Computational Complexity & no break expected & 2023-04 & 2023-10, 2025-01 & yes & -0.068 (0.000) & +0.0142 (6.6e-55) & 0.062 & 0.061 \\
14+Operating Systems & no break expected & 2023-09 & 2024-11, 2026-01 & no & -0.033 (0.417) & +0.0158 (8.2e-15) & 0.477 & 0.292 \\
15+Computer Architecture & no break expected & 2023-05 & 2025-01 & yes & -0.009 (0.771) & +0.0158 (2.4e-38) & 0.062 & 0.065 \\
16+\bottomrule
17+\end{tabular}
18+\begin{tablenotes}\footnotesize\item In window: a Bai--Perron (BIC) break date between October 2022 and June 2023. The protocol predicted a break in that window for the first four categories and none for the last four.\end{tablenotes}
19+\end{threeparttable}\end{table}
added manuscript/tables/tab_p9_its_levels.tex +54 −0
@@ -0,0 +1,54 @@
1+% generated by tables/make_tables.py from ./results/econometrics/its_results.json
2+\begin{longtable}{lrrrrrrrr}
3+\caption{ITS level change at December 2022 by category: full-sample and donut estimates, HAC and placebo $p$-values, and the BOCPD change probability over December 2022 to March 2023.}\label{tab:p9-its-levels}\\
4+\toprule
5+\textbf{Category} & \textbf{Level change} & \textbf{SE} & \textbf{HAC p} & \textbf{Placebo p (all)} & \textbf{Placebo p (base)} & \textbf{Donut level} & \textbf{Donut p} & \textbf{BOCPD P} \\
6+\midrule
7+\endfirsthead
8+\toprule
9+\textbf{Category} & \textbf{Level change} & \textbf{SE} & \textbf{HAC p} & \textbf{Placebo p (all)} & \textbf{Placebo p (base)} & \textbf{Donut level} & \textbf{Donut p} & \textbf{BOCPD P} \\
10+\midrule
11+\endhead
12+AI Agents and Autonomous Systems & -0.210 & 0.095 & 0.028 & 0.631 & 0.882 & -0.105 & 0.282 & 0.123 \\
13+AI Safety and Alignment & -0.206 & 0.067 & 0.002 & 0.446 & 0.676 & +0.022 & 0.752 & 0.017 \\
14+Accelerators and AI Hardware & -0.113 & 0.051 & 0.027 & 0.477 & 0.618 & -0.017 & 0.749 & 0.025 \\
15+Algorithms and Data Structures & -0.045 & 0.022 & 0.039 & 0.508 & 0.676 & -0.003 & 0.906 & 0.036 \\
16+Applied Cryptography & +0.002 & 0.021 & 0.922 & 1.000 & 1.000 & +0.066 & 0.001 & 0.116 \\
17+Autonomous Scientific Discovery & -0.155 & 0.075 & 0.038 & 0.538 & 0.824 & +0.003 & 0.972 & 0.019 \\
18+Blockchain Systems & -0.008 & 0.044 & 0.858 & 0.938 & 1.000 & +0.068 & 0.163 & 0.023 \\
19+Cloud Computing & +0.003 & 0.037 & 0.936 & 0.985 & 1.000 & +0.039 & 0.255 & 0.405 \\
20+Compilers & -0.028 & 0.039 & 0.471 & 0.892 & 0.971 & +0.027 & 0.516 & 0.079 \\
21+Computational Biology & -0.116 & 0.032 & 0.000 & 0.292 & 0.500 & -0.045 & 0.157 & 0.018 \\
22+Computational Complexity & -0.068 & 0.019 & 0.000 & 0.354 & 0.353 & -0.005 & 0.827 & 0.061 \\
23+Computer Architecture & -0.009 & 0.031 & 0.771 & 0.954 & 1.000 & +0.030 & 0.335 & 0.065 \\
24+Computer Graphics & +0.043 & 0.025 & 0.090 & 0.846 & 0.794 & +0.136 & 0.000 & 0.084 \\
25+Computer Networks & +0.043 & 0.019 & 0.021 & 0.538 & 0.324 & +0.067 & 0.007 & 0.179 \\
26+Computer Vision & +0.024 & 0.023 & 0.305 & 0.831 & 0.706 & +0.109 & 0.000 & 0.019 \\
27+Cybersecurity & -0.021 & 0.025 & 0.408 & 0.923 & 0.971 & +0.051 & 0.049 & 0.087 \\
28+Database Systems & -0.013 & 0.025 & 0.615 & 0.954 & 1.000 & +0.046 & 0.047 & 0.150 \\
29+Deep Learning Architectures & -0.071 & 0.046 & 0.126 & 0.400 & 0.735 & +0.027 & 0.551 & 0.017 \\
30+Distributed Computing & -0.034 & 0.032 & 0.282 & 0.877 & 0.971 & +0.027 & 0.411 & 0.144 \\
31+Edge Computing & -0.044 & 0.030 & 0.139 & 0.600 & 0.794 & +0.019 & 0.519 & 0.033 \\
32+Federated Learning & -0.126 & 0.047 & 0.007 & 0.246 & 0.294 & +0.103 & 0.032 & 0.017 \\
33+Formal Verification & -0.109 & 0.050 & 0.029 & 0.415 & 0.618 & -0.085 & 0.123 & 0.048 \\
34+Foundation Models & -0.075 & 0.078 & 0.340 & 0.692 & 0.941 & +0.147 & 0.049 & 0.018 \\
35+Generative AI & +0.036 & 0.062 & 0.556 & 0.969 & 0.941 & +0.260 & 0.000 & 0.020 \\
36+Human-Computer Interaction & +0.049 & 0.022 & 0.027 & 0.877 & 0.765 & +0.151 & 0.000 & 0.043 \\
37+Information Retrieval & -0.094 & 0.040 & 0.019 & 0.585 & 0.765 & -0.003 & 0.949 & 0.104 \\
38+Large Language Models & +0.106 & 0.060 & 0.075 & 0.938 & 0.882 & +0.347 & 0.000 & 0.093 \\
39+Machine Learning Theory & -0.080 & 0.047 & 0.090 & 0.662 & 1.000 & +0.005 & 0.926 & 0.020 \\
40+Multimodal AI & +0.025 & 0.048 & 0.600 & 0.969 & 0.941 & +0.284 & 0.000 & 0.023 \\
41+Natural Language Processing & +0.016 & 0.041 & 0.694 & 0.969 & 0.941 & +0.150 & 0.001 & 0.045 \\
42+Operating Systems & -0.033 & 0.041 & 0.417 & 0.846 & 0.971 & -0.020 & 0.549 & 0.292 \\
43+Privacy-Preserving Computation & -0.001 & 0.028 & 0.965 & 0.985 & 0.971 & +0.142 & 0.000 & 0.020 \\
44+Programming Languages & -0.040 & 0.045 & 0.381 & 0.662 & 0.941 & -0.007 & 0.895 & 0.241 \\
45+Quantum Computing & -0.087 & 0.042 & 0.039 & 0.738 & 0.941 & +0.028 & 0.571 & 0.018 \\
46+Reinforcement Learning & -0.164 & 0.066 & 0.013 & 0.615 & 0.824 & -0.062 & 0.340 & 0.019 \\
47+Robotics & +0.012 & 0.027 & 0.660 & 0.908 & 0.912 & +0.085 & 0.006 & 0.020 \\
48+Scientific Computing & -0.047 & 0.024 & 0.054 & 0.523 & 0.676 & -0.003 & 0.890 & 0.025 \\
49+Software Engineering & +0.048 & 0.031 & 0.124 & 0.862 & 0.765 & +0.156 & 0.000 & 0.131 \\
50+Speech and Audio Computing & +0.036 & 0.039 & 0.354 & 0.815 & 0.824 & +0.109 & 0.024 & 0.032 \\
51+Sustainable Computing & +0.049 & 0.025 & 0.048 & 0.754 & 0.647 & +0.121 & 0.000 & 0.193 \\
52+\bottomrule
53+\multicolumn{9}{p{0.95\linewidth}}{\footnotesize Donut: the event month and three months on each side removed. BOCPD P: maximum posterior probability that a changepoint occurred within the last four months, taken over December 2022 to March 2023.}\\
54+\end{longtable}
added manuscript/tables/tab_p9_its_slopes.tex +54 −0
@@ -0,0 +1,54 @@
1+% generated by tables/make_tables.py from ./results/econometrics/its_results.json
2+\begin{longtable}{lrrrrrrrr}
3+\caption{ITS slope change at December 2022 by category: pre-period slope, change, HAC and placebo $p$-values, and implied post-period slope (monthly log growth).}\label{tab:p9-its-slopes}\\
4+\toprule
5+\textbf{Category} & \textbf{Works} & \textbf{Pre-slope} & \textbf{Slope change} & \textbf{SE} & \textbf{HAC p} & \textbf{Placebo p (all)} & \textbf{Placebo p (base)} & \textbf{Post slope} \\
6+\midrule
7+\endfirsthead
8+\toprule
9+\textbf{Category} & \textbf{Works} & \textbf{Pre-slope} & \textbf{Slope change} & \textbf{SE} & \textbf{HAC p} & \textbf{Placebo p (all)} & \textbf{Placebo p (base)} & \textbf{Post slope} \\
10+\midrule
11+\endhead
12+AI Agents and Autonomous Systems & 116{,}719 & 0.0032 & +0.0401 & 0.0046 & 3.61e-18 & 0.400 & 0.000 & 0.0433 \\
13+Large Language Models & 346{,}928 & 0.0132 & +0.0336 & 0.0020 & 1.07e-60 & 0.200 & 0.353 & 0.0469 \\
14+Multimodal AI & 133{,}911 & 0.0192 & +0.0238 & 0.0018 & 6.35e-39 & 0.200 & 0.353 & 0.0430 \\
15+Information Retrieval & 206{,}216 & 0.0031 & +0.0233 & 0.0019 & 7.12e-35 & 0.092 & 0.000 & 0.0264 \\
16+Distributed Computing & 359{,}867 & 0.0013 & +0.0200 & 0.0015 & 4.32e-43 & 0.077 & 0.000 & 0.0213 \\
17+Generative AI & 175{,}254 & 0.0190 & +0.0194 & 0.0024 & 1.05e-15 & 0.308 & 0.324 & 0.0384 \\
18+Cloud Computing & 202{,}261 & -0.0013 & +0.0187 & 0.0018 & 1.94e-26 & 0.231 & 0.265 & 0.0174 \\
19+Natural Language Processing & 434{,}211 & 0.0094 & +0.0184 & 0.0015 & 1.21e-33 & 0.154 & 0.265 & 0.0278 \\
20+Quantum Computing & 125{,}359 & 0.0074 & +0.0181 & 0.0016 & 5.22e-31 & 0.169 & 0.000 & 0.0255 \\
21+Compilers & 48{,}920 & 0.0018 & +0.0180 & 0.0021 & 5.63e-18 & 0.215 & 0.176 & 0.0198 \\
22+Accelerators and AI Hardware & 104{,}043 & 0.0071 & +0.0180 & 0.0021 & 4.32e-18 & 0.169 & 0.029 & 0.0252 \\
23+Formal Verification & 77{,}664 & -0.0009 & +0.0177 & 0.0022 & 2.53e-15 & 0.246 & 0.029 & 0.0168 \\
24+AI Safety and Alignment & 139{,}500 & 0.0266 & +0.0164 & 0.0022 & 3.04e-13 & 0.308 & 0.029 & 0.0430 \\
25+Software Engineering & 388{,}733 & 0.0042 & +0.0162 & 0.0012 & 1.09e-44 & 0.154 & 0.265 & 0.0204 \\
26+Operating Systems & 101{,}411 & -0.0011 & +0.0158 & 0.0020 & 8.2e-15 & 0.477 & 0.294 & 0.0146 \\
27+Computer Architecture & 304{,}720 & -0.0011 & +0.0158 & 0.0012 & 2.36e-38 & 0.062 & 0.088 & 0.0146 \\
28+Sustainable Computing & 67{,}460 & 0.0039 & +0.0156 & 0.0012 & 1.23e-38 & 0.169 & 0.294 & 0.0196 \\
29+Human-Computer Interaction & 451{,}350 & 0.0038 & +0.0149 & 0.0009 & 7.23e-69 & 0.154 & 0.265 & 0.0188 \\
30+Cybersecurity & 630{,}599 & 0.0045 & +0.0146 & 0.0011 & 6.43e-37 & 0.062 & 0.000 & 0.0191 \\
31+Computational Complexity & 150{,}983 & 0.0024 & +0.0142 & 0.0009 & 6.63e-55 & 0.062 & 0.000 & 0.0166 \\
32+Computer Graphics & 192{,}097 & 0.0030 & +0.0141 & 0.0010 & 2.69e-42 & 0.185 & 0.324 & 0.0171 \\
33+Database Systems & 261{,}915 & 0.0013 & +0.0136 & 0.0012 & 5.81e-31 & 0.092 & 0.000 & 0.0148 \\
34+Applied Cryptography & 252{,}420 & 0.0026 & +0.0133 & 0.0009 & 9.67e-46 & 0.046 & 0.059 & 0.0159 \\
35+Privacy-Preserving Computation & 181{,}159 & 0.0140 & +0.0125 & 0.0013 & 6.58e-21 & 0.138 & 0.235 & 0.0265 \\
36+Programming Languages & 116{,}536 & 0.0014 & +0.0123 & 0.0023 & 1.34e-07 & 0.431 & 0.324 & 0.0137 \\
37+Speech and Audio Computing & 146{,}061 & 0.0029 & +0.0118 & 0.0016 & 1.76e-13 & 0.169 & 0.294 & 0.0147 \\
38+Reinforcement Learning & 171{,}820 & 0.0173 & +0.0116 & 0.0022 & 1.55e-07 & 0.385 & 0.000 & 0.0289 \\
39+Autonomous Scientific Discovery & 24{,}409 & 0.0207 & +0.0112 & 0.0021 & 1.31e-07 & 0.431 & 0.000 & 0.0319 \\
40+Algorithms and Data Structures & 500{,}703 & 0.0022 & +0.0098 & 0.0010 & 4.01e-21 & 0.062 & 0.000 & 0.0120 \\
41+Machine Learning Theory & 424{,}417 & 0.0106 & +0.0091 & 0.0016 & 1.07e-08 & 0.215 & 0.000 & 0.0197 \\
42+Computer Networks & 654{,}017 & 0.0002 & +0.0086 & 0.0010 & 4.81e-18 & 0.200 & 0.353 & 0.0088 \\
43+Scientific Computing & 388{,}897 & 0.0056 & +0.0068 & 0.0009 & 4.43e-14 & 0.154 & 0.000 & 0.0124 \\
44+Computer Vision & 1{,}027{,}433 & 0.0076 & +0.0065 & 0.0009 & 3.43e-13 & 0.169 & 0.294 & 0.0141 \\
45+Foundation Models & 193{,}776 & 0.0255 & +0.0064 & 0.0028 & 0.0208 & 0.246 & 0.412 & 0.0320 \\
46+Computational Biology & 267{,}762 & 0.0107 & +0.0055 & 0.0008 & 3.95e-11 & 0.308 & 0.000 & 0.0162 \\
47+Robotics & 318{,}268 & 0.0083 & +0.0051 & 0.0012 & 8.89e-06 & 0.108 & 0.176 & 0.0134 \\
48+Edge Computing & 231{,}512 & 0.0103 & +0.0039 & 0.0014 & 0.00561 & 0.077 & 0.000 & 0.0142 \\
49+Blockchain Systems & 138{,}361 & 0.0148 & -0.0021 & 0.0016 & 0.2 & 0.508 & 0.735 & 0.0127 \\
50+Deep Learning Architectures & 772{,}532 & 0.0174 & -0.0025 & 0.0017 & 0.136 & 0.708 & 1.000 & 0.0149 \\
51+Federated Learning & 58{,}126 & 0.0369 & -0.0063 & 0.0019 & 0.001 & 0.385 & 0.676 & 0.0305 \\
52+\bottomrule
53+\multicolumn{9}{p{0.95\linewidth}}{\footnotesize Multi-label monthly series of month-precision works (Part IV); Works: series total over the window. Placebo $p$ (all): rank of the slope $t$-statistic among all candidate event months; (base): among baseline months only, where zero is attainable. Sorted by slope change.}\\
54+\end{longtable}
added manuscript/tables/tab_p9_its_total_events.tex +16 −0
@@ -0,0 +1,16 @@
1+% generated by tables/make_tables.py from ./results/econometrics/its_results.json
2+\begin{table}[htbp]\centering\caption{Interrupted time series of the total monthly series at the reference date and at four alternative event dates (log counts, month-of-year dummies, Newey--West standard errors).}\label{tab:p9-its-total-events}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrrrrrr}
5+\toprule
6+\textbf{Event (month)} & \textbf{Pre-slope (SE)} & \textbf{Level change (SE)} & \textbf{HAC p} & \textbf{Slope change (SE)} & \textbf{HAC p } & \textbf{R2} \\
7+\midrule
8+GPT-3 (2020-06) & 0.0034 (0.0012) & -0.091 (0.059) & 0.123 & +0.0059 (0.0015) & 8.51e-05 & 0.926 \\
9+DALL$\cdot$E 2 (2022-05) & 0.0034 (0.0007) & -0.075 (0.028) & 0.006 & +0.0100 (0.0009) & 1.39e-28 & 0.965 \\
10+ChatGPT (2022-12) & 0.0031 (0.0005) & -0.010 (0.024) & 0.685 & +0.0108 (0.0009) & 6.56e-34 & 0.966 \\
11+GPT-4 (2023-04) & 0.0032 (0.0005) & +0.025 (0.024) & 0.307 & +0.0109 (0.0010) & 6.36e-30 & 0.966 \\
12+o1 (2024-10) & 0.0049 (0.0006) & +0.103 (0.039) & 0.008 & +0.0115 (0.0018) & 2.88e-10 & 0.951 \\
13+\bottomrule
14+\end{tabular}
15+\begin{tablenotes}\footnotesize\item Equation~\eqref{eq:its}; each event is coded as the first full month after the public release. Slopes are monthly log growth rates. A HAC $p$ printed as 0.000 is below 0.0005.\end{tablenotes}
16+\end{threeparttable}\end{table}
added manuscript/tables/tab_p9_landmarks.tex +28 −0
@@ -0,0 +1,28 @@
1+% generated by tables/make_tables.py from ./literature_synthesis/landmarks/selection_report.json
2+\begin{table}[htbp]\centering\caption{Composition of the landmark set by area and by publication year.}\label{tab:p9-landmarks}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrr}
5+\toprule
6+\textbf{Group} & \textbf{Level} & \textbf{Landmarks} \\
7+\midrule
8+Area & Artificial Intelligence and Machine Learning & 234 \\
9+Area & Perception, Language, Robotics and Interaction & 207 \\
10+Area & Data, Information and Distributed Systems & 152 \\
11+Area & Security, Privacy and Trust & 125 \\
12+Area & Hardware, Architecture and Emerging Computing & 125 \\
13+Area & Software, Programming Languages and Formal Methods & 100 \\
14+Area & Computational Science and Discovery & 100 \\
15+Area & Theory of Computation & 50 \\
16+Year & 2018 & 108 \\
17+Year & 2019 & 107 \\
18+Year & 2020 & 106 \\
19+Year & 2021 & 87 \\
20+Year & 2022 & 92 \\
21+Year & 2023 & 128 \\
22+Year & 2024 & 135 \\
23+Year & 2025 & 111 \\
24+Year & 2026 & 219 \\
25+\bottomrule
26+\end{tabular}
27+\begin{tablenotes}\footnotesize\item Selection by field-normalised impact within category and year with per-area quotas (Part IV); the specialist review of the set (checkpoint H-3) is pending, so the composition is reported without any assessment of the works.\end{tablenotes}
28+\end{threeparttable}\end{table}
added manuscript/tables/tab_p9_llm_partners.tex +26 −0
@@ -0,0 +1,26 @@
1+% generated by tables/make_tables.py from ./results/graph/category_cooccurrence_yearly.csv
2+\begin{table}[htbp]\centering\caption{Categories most often co-assigned with Large Language Models, by span.}\label{tab:p9-llm-partners}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrrr}
5+\toprule
6+\textbf{Partner category} & \textbf{Works, 2018--2022} & \textbf{Works, 2023--2026} & \textbf{Ratio} \\
7+\midrule
8+Natural Language Processing & 96{,}346 & 233{,}717 & 2.43 \\
9+Foundation Models & 17{,}107 & 87{,}013 & 5.09 \\
10+Deep Learning Architectures & 29{,}989 & 73{,}479 & 2.45 \\
11+Computer Vision & 12{,}171 & 63{,}271 & 5.20 \\
12+Multimodal AI & 9{,}607 & 62{,}476 & 6.50 \\
13+Information Retrieval & 20{,}069 & 59{,}453 & 2.96 \\
14+Software Engineering & 4{,}793 & 55{,}862 & 11.65 \\
15+AI Safety and Alignment & 2{,}505 & 46{,}899 & 18.72 \\
16+Machine Learning Theory & 10{,}228 & 37{,}210 & 3.64 \\
17+AI Agents and Autonomous Systems & 1{,}097 & 35{,}482 & 32.34 \\
18+Generative AI & 3{,}218 & 34{,}832 & 10.82 \\
19+Cybersecurity & 4{,}824 & 32{,}240 & 6.68 \\
20+Distributed Computing & 3{,}005 & 29{,}709 & 9.89 \\
21+Human-Computer Interaction & 2{,}544 & 26{,}526 & 10.43 \\
22+Reinforcement Learning & 1{,}723 & 22{,}085 & 12.82 \\
23+\bottomrule
24+\end{tabular}
25+\begin{tablenotes}\footnotesize\item Works carrying both the LLM label and the partner label; spans split by calendar year; the second span is incomplete.\end{tablenotes}
26+\end{threeparttable}\end{table}
added manuscript/tables/tab_p9_modularity.tex +20 −0
@@ -0,0 +1,20 @@
1+% generated by tables/make_tables.py from ./results/graph/modularity_yearly.csv
2+\begin{table}[htbp]\centering\caption{Modularity of the yearly category citation-flow network at the category and area partitions, and the within-category self-share of citations.}\label{tab:p9-modularity}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrrrr}
5+\toprule
6+\textbf{Year} & \textbf{Internal citations} & \textbf{Q (categories)} & \textbf{Q (areas)} & \textbf{Self-share (\%)} \\
7+\midrule
8+2018 & 213{,}358 & 0.5428 & 0.5083 & 62.09 \\
9+2019 & 975{,}445 & 0.5510 & 0.5007 & 64.15 \\
10+2020 & 2{,}131{,}658 & 0.5450 & 0.4914 & 63.74 \\
11+2021 & 3{,}556{,}446 & 0.5372 & 0.4789 & 63.36 \\
12+2022 & 4{,}288{,}984 & 0.5356 & 0.4750 & 63.58 \\
13+2023 & 5{,}706{,}569 & 0.5263 & 0.4605 & 62.97 \\
14+2024 & 6{,}799{,}414 & 0.5123 & 0.4595 & 61.83 \\
15+2025 & 7{,}477{,}998 & 0.5059 & 0.4593 & 60.19 \\
16+2026 & 4{,}558{,}848 & 0.5004 & 0.4615 & 58.93 \\
17+\bottomrule
18+\end{tabular}
19+\begin{tablenotes}\footnotesize\item Equation~\eqref{eq:modularity} at the fixed partitions; the citation counts are internal references of works of the citing year whose two ends have a primary category. The 2018 network is small because few cited works of 2018 or later existed; 2026 is incomplete.\end{tablenotes}
20+\end{threeparttable}\end{table}
added manuscript/tables/tab_p9_openness.tex +20 −0
@@ -0,0 +1,20 @@
1+% generated by tables/make_tables.py from ./results/bibliometrics/volume_yearly.csv
2+\begin{table}[htbp]\centering\caption{Openness and metadata indicators by publication year: arXiv linkage, preprint share, open access, abstract availability and document type (per cent of eligible works).}\label{tab:p9-openness}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrrrrrrrrr}
5+\toprule
6+\textbf{Year} & \textbf{Works} & \textbf{DBLP (\%)} & \textbf{arXiv-linked (\%)} & \textbf{Preprint type (\%)} & \textbf{Open access (\%)} & \textbf{Abstract (\%)} & \textbf{Conference (\%)} & \textbf{Article (\%)} & \textbf{Mean refs.} \\
7+\midrule
8+2018 & 642{,}849 & 50.4 & 5.6 & 4.0 & 39.5 & 70.0 & 32.4 & 50.9 & 18.1 \\
9+2019 & 695{,}369 & 52.8 & 7.2 & 4.7 & 43.0 & 71.3 & 32.9 & 50.9 & 19.8 \\
10+2020 & 709{,}083 & 52.8 & 8.9 & 5.7 & 47.3 & 72.0 & 28.9 & 54.3 & 22.1 \\
11+2021 & 711{,}244 & 56.2 & 10.0 & 6.0 & 50.6 & 76.3 & 30.9 & 53.7 & 25.5 \\
12+2022 & 699{,}548 & 57.3 & 10.4 & 6.1 & 54.2 & 78.4 & 31.4 & 53.1 & 24.9 \\
13+2023 & 785{,}194 & 55.1 & 11.1 & 7.1 & 58.1 & 80.1 & 33.2 & 51.3 & 24.2 \\
14+2024 & 901{,}654 & 51.9 & 12.1 & 8.1 & 57.7 & 80.6 & 32.2 & 51.6 & 21.9 \\
15+2025 & 1{,}065{,}828 & 46.1 & 13.4 & 10.8 & 59.4 & 79.2 & 29.3 & 51.1 & 17.4 \\
16+2026 & 901{,}651 & 38.8 & 19.1 & 21.3 & 66.4 & 70.8 & 21.6 & 47.2 & 10.9 \\
17+\bottomrule
18+\end{tabular}
19+\begin{tablenotes}\footnotesize\item arXiv-linked: the work is linked to an arXiv e-print (any version); preprint type: the surviving record's OpenAlex type is preprint; open access: any OA location in OpenAlex. Mean refs.: declared references per work, which falls in recent years because references are indexed with a lag. 2026 is incomplete.\end{tablenotes}
20+\end{threeparttable}\end{table}
added manuscript/tables/tab_p9_periods.tex +24 −0
@@ -0,0 +1,24 @@
1+% generated by tables/make_tables.py from ./results/bibliometrics/volume_monthly.csv
2+\begin{table}[htbp]\centering\caption{The eligible corpus by protocol period: completeness, source membership and document type.}\label{tab:p9-periods}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrr}
5+\toprule
6+\textbf{Indicator} & \textbf{Baseline} & \textbf{Post} \\
7+\midrule
8+Eligible works (count) & 3{,}405{,}453.00 & 3{,}705{,}041.00 \\
9+Month-precision first date (\\%) & 77.43 & 82.78 \\
10+DBLP-indexed (\\%) & 54.16 & 47.56 \\
11+OpenAlex primary field CS (\\%) & 72.33 & 77.15 \\
12+arXiv-linked (\\%) & 8.70 & 13.66 \\
13+With abstract (\\%) & 73.42 & 77.88 \\
14+With at least one reference (\\%) & 71.63 & 58.71 \\
15+Open access (\\%) & 47.04 & 60.23 \\
16+Type: article (\\%) & 52.63 & 50.31 \\
17+Type: conference paper (\\%) & 31.24 & 29.01 \\
18+Type: preprint (\\%) & 5.29 & 11.85 \\
19+Type: book chapter (\\%) & 5.41 & 4.29 \\
20+Type: dissertation (\\%) & 3.99 & 2.74 \\
21+\bottomrule
22+\end{tabular}
23+\begin{tablenotes}\footnotesize\item Baseline: first date 2018-01-01 to 2022-11-29; post: 2022-11-30 to 2026-11-02 (right-truncated, see Part III). Percentages are over all eligible works of the period. The count row is an integer written with two decimals by the table generator.\end{tablenotes}
24+\end{threeparttable}\end{table}
added manuscript/tables/tab_p9_total_robustness.tex +23 −0
@@ -0,0 +1,23 @@
1+% generated by tables/make_tables.py from ./results/econometrics/its_results.json
2+\begin{table}[htbp]\centering\caption{Robustness diagnostics of the structural-change analysis of the total monthly series.}\label{tab:p9-total-robustness}
3+\begin{threeparttable}\small
4+\begin{tabular}{lr}
5+\toprule
6+\textbf{Diagnostic} & \textbf{Total series} \\
7+\midrule
8+Observations (months), first to last complete month & 101 (2018-02 to 2026-06) \\
9+Placebo rank of the level $t$-statistic (all candidate months) & rank 63 of 65, placebo $p$ = 0.969 \\
10+Placebo rank of the slope $t$-statistic (all candidate months) & rank 1 of 65, placebo $p$ = 0.015 \\
11+Placebo $p$, baseline candidate months only (level; slope) & 1.000; 0.000 (n = 34; zero is attainable) \\
12+Donut ITS, $\pm$3 months removed: level change (HAC $p$) & +0.049 (0.057), n = 94 \\
13+Donut ITS: slope change (HAC $p$) & +0.0110 (7e-24) \\
14+Bai--Perron break dates, BIC (LWZ) & 2022-04-01 (2022-04-01); m = 1, h = 15 \\
15+Bai--Perron BIC by number of breaks m = 0, 1, 2, 3 & -463.9, -555.3, -551.6, -546.0 \\
16+PELT changepoints (piecewise-constant mean) & 2023-05, 2024-12 \\
17+BOCPD: max P(run length $\le$ 3) over 2022-12 to 2023-03; MAP resets & 0.033; none \\
18+Last complete month 2026-03: level (p); slope (p); Bai--Perron & -0.009 (0.723); +0.0108 (1.1e-30); 2022-04-01 \\
19+Last complete month 2026-08: level (p); slope (p); Bai--Perron & +0.004 (0.870); +0.0098 (7.4e-18); 2022-02-01 \\
20+\bottomrule
21+\end{tabular}
22+\begin{tablenotes}\footnotesize\item Placebo ranks follow equation~\eqref{eq:placebo}; the baseline-only rank excludes the true event from the candidate set, so a value of zero means that no baseline month yields a larger $t$-statistic. Bai--Perron on the seasonally adjusted log series with trimming $h$; PELT with the $\ell_2$ cost; BOCPD with hazard 1/60 (Part IV).\end{tablenotes}
23+\end{threeparttable}\end{table}
added manuscript/tables/tab_p9_unknown_breaks.tex +54 −0
@@ -0,0 +1,54 @@
1+% generated by tables/make_tables.py from ./results/econometrics/its_results.json
2+\begin{longtable}{lrrrrrr}
3+\caption{Data-driven break and changepoint dates by category, and whether any falls in the H3 window (October 2022 to June 2023).}\label{tab:p9-unknown-breaks}\\
4+\toprule
5+\textbf{Category} & \textbf{Bai--Perron (BIC)} & \textbf{Bai--Perron (LWZ)} & \textbf{PELT} & \textbf{BOCPD resets} & \textbf{BIC break in window} & \textbf{PELT in window} \\
6+\midrule
7+\endfirsthead
8+\toprule
9+\textbf{Category} & \textbf{Bai--Perron (BIC)} & \textbf{Bai--Perron (LWZ)} & \textbf{PELT} & \textbf{BOCPD resets} & \textbf{BIC break in window} & \textbf{PELT in window} \\
10+\midrule
11+\endhead
12+AI Agents and Autonomous Systems & 2022-07, 2024-07 & 2022-07, 2024-07 & 2023-12, 2025-08 & none & no & no \\
13+AI Safety and Alignment & 2021-06, 2022-09 & 2022-07 & 2020-02, 2024-02 & none & no & no \\
14+Accelerators and AI Hardware & 2024-11 & 2024-11 & 2023-04, 2025-01 & none & no & yes \\
15+Algorithms and Data Structures & 2023-04 & 2023-04 & 2023-08, 2025-08 & none & yes & no \\
16+Applied Cryptography & 2022-07 & 2022-07 & 2022-11, 2024-12 & none & no & yes \\
17+Autonomous Scientific Discovery & 2020-02, 2022-06, 2024-07 & 2022-05 & 2020-02, 2023-12 & none & no & no \\
18+Blockchain Systems & 2020-04 & none & 2019-02, 2022-12 & 2019-03 & no & yes \\
19+Cloud Computing & 2022-11, 2025-01 & 2023-05 & 2025-01 & 2025-02 & yes & no \\
20+Compilers & 2023-04 & 2023-04 & 2023-10, 2025-10 & 2021-04 & yes & no \\
21+Computational Biology & 2020-02, 2023-12 & 2022-07 & 2020-02, 2023-12 & none & no & no \\
22+Computational Complexity & 2023-04 & 2023-04 & 2023-10, 2025-01 & none & yes & no \\
23+Computer Architecture & 2023-05 & 2023-05 & 2025-01 & none & yes & no \\
24+Computer Graphics & 2022-08, 2024-07 & 2022-08 & 2023-02, 2024-12 & none & no & yes \\
25+Computer Networks & 2022-02 & 2022-02 & 2023-05, 2024-12 & none & no & yes \\
26+Computer Vision & 2022-04 & 2022-04 & 2020-12, 2023-12 & none & no & no \\
27+Cybersecurity & 2022-07 & 2022-07 & 2022-11, 2024-12 & none & no & yes \\
28+Database Systems & 2023-11 & 2023-11 & 2023-11, 2025-01 & none & no & no \\
29+Deep Learning Architectures & 2021-07 & none & 2020-02, 2023-12 & 2019-03 & no & no \\
30+Distributed Computing & 2022-07, 2024-07 & 2023-04 & 2023-10, 2025-01 & none & no & no \\
31+Edge Computing & 2025-01 & none & 2020-06, 2023-12 & none & no & no \\
32+Federated Learning & 2020-02 & 2020-02 & 2020-02, 2023-05 & none & no & yes \\
33+Formal Verification & 2022-02, 2024-07 & 2024-09 & 2024-03, 2025-09 & 2024-03 & no & no \\
34+Foundation Models & 2021-10 & none & 2020-02, 2023-10 & none & no & no \\
35+Generative AI & 2021-11 & 2021-11 & 2020-12, 2023-12 & none & no & no \\
36+Human-Computer Interaction & 2022-11 & 2022-11 & 2023-12 & 2021-05 & yes & no \\
37+Information Retrieval & 2022-07, 2024-07 & 2023-07 & 2023-02, 2024-12 & none & no & yes \\
38+Large Language Models & 2022-07 & 2022-07 & 2023-10 & none & no & no \\
39+Machine Learning Theory & 2020-05, 2022-07 & 2021-07 & 2020-02, 2023-12 & none & no & no \\
40+Multimodal AI & 2022-04 & 2022-04 & 2020-12, 2023-12 & none & no & no \\
41+Natural Language Processing & 2022-07 & 2022-07 & 2023-10 & none & no & no \\
42+Operating Systems & 2023-09 & 2023-09 & 2024-11, 2026-01 & 2024-01 & no & no \\
43+Privacy-Preserving Computation & 2022-07 & 2022-07 & 2020-12, 2023-12 & none & no & no \\
44+Programming Languages & 2022-04, 2024-06 & 2024-08 & 2023-10, 2025-10 & none & no & no \\
45+Quantum Computing & 2022-01, 2025-04 & 2024-04 & 2023-05, 2025-04 & none & no & yes \\
46+Reinforcement Learning & 2021-07, 2024-12 & 2024-12 & 2021-01, 2024-12 & none & no & no \\
47+Robotics & 2021-10 & none & 2020-09, 2023-12 & none & no & no \\
48+Scientific Computing & 2025-01 & 2025-01 & 2020-09, 2024-01 & none & no & no \\
49+Software Engineering & 2022-07 & 2022-07 & 2023-03, 2025-01 & none & no & yes \\
50+Speech and Audio Computing & 2023-05 & 2023-05 & 2023-05, 2025-12 & none & yes & yes \\
51+Sustainable Computing & 2022-07 & 2022-07 & 2022-11, 2024-12 & none & no & yes \\
52+\bottomrule
53+\multicolumn{7}{p{0.95\linewidth}}{\footnotesize Bai--Perron on the seasonally adjusted log series with trimming $h = 15$ months (no break possible after April 2025); PELT with the $\ell_2$ cost and a six-month minimum segment; BOCPD resets as defined in Part IV. The window test is the one pre-specified in H3; Bai--Perron dates carry no $p$-value.}\\
54+\end{longtable}
added manuscript/tables/tab_security_counts.tex +23 −0
@@ -0,0 +1,23 @@
1+% generated by tables/make_tables.py from ./results/graph/category_counts_yearly.csv
2+\begin{longtable}{lrrrrrrrrrr}
3+\caption{Eligible works in the five security-and-privacy categories by publication year: works carrying the category (all) and works whose primary category it is (prim.).}\label{tab:security_counts}\\
4+\toprule
5+\textbf{Year} & \textbf{Cyber. (all)} & \textbf{Crypto. (all)} & \textbf{Blockch. (all)} & \textbf{Privacy (all)} & \textbf{Fed. learn. (all)} & \textbf{Cyber. (prim.)} & \textbf{Crypto. (prim.)} & \textbf{Blockch. (prim.)} & \textbf{Privacy (prim.)} & \textbf{Fed. learn. (prim.)} \\
6+\midrule
7+\endfirsthead
8+\toprule
9+\textbf{Year} & \textbf{Cyber. (all)} & \textbf{Crypto. (all)} & \textbf{Blockch. (all)} & \textbf{Privacy (all)} & \textbf{Fed. learn. (all)} & \textbf{Cyber. (prim.)} & \textbf{Crypto. (prim.)} & \textbf{Blockch. (prim.)} & \textbf{Privacy (prim.)} & \textbf{Fed. learn. (prim.)} \\
10+\midrule
11+\endhead
12+2018 & 67{,}575 & 29{,}699 & 10{,}872 & 11{,}378 & 1{,}040 & 29{,}440 & 10{,}782 & 6{,}441 & 3{,}105 & 143 \\
13+2019 & 73{,}397 & 31{,}450 & 15{,}314 & 13{,}062 & 1{,}578 & 31{,}561 & 11{,}355 & 9{,}047 & 3{,}256 & 216 \\
14+2020 & 76{,}566 & 31{,}778 & 16{,}707 & 15{,}779 & 2{,}684 & 32{,}705 & 11{,}279 & 9{,}691 & 4{,}157 & 304 \\
15+2021 & 77{,}276 & 32{,}097 & 18{,}128 & 17{,}795 & 4{,}246 & 31{,}779 & 10{,}924 & 10{,}647 & 4{,}241 & 651 \\
16+2022 & 78{,}082 & 32{,}075 & 20{,}159 & 20{,}053 & 5{,}888 & 31{,}554 & 10{,}299 & 11{,}770 & 4{,}376 & 974 \\
17+2023 & 89{,}483 & 35{,}724 & 24{,}314 & 25{,}211 & 8{,}515 & 36{,}481 & 11{,}219 & 14{,}349 & 5{,}010 & 1{,}497 \\
18+2024 & 107{,}542 & 41{,}181 & 26{,}919 & 34{,}579 & 11{,}923 & 45{,}548 & 12{,}426 & 15{,}376 & 6{,}686 & 2{,}070 \\
19+2025 & 139{,}134 & 51{,}759 & 33{,}426 & 49{,}266 & 18{,}093 & 59{,}044 & 14{,}612 & 18{,}012 & 8{,}817 & 2{,}544 \\
20+2026 & 120{,}483 & 43{,}945 & 24{,}814 & 41{,}713 & 17{,}070 & 51{,}618 & 12{,}178 & 13{,}276 & 7{,}625 & 2{,}841 \\
21+\bottomrule
22+\multicolumn{11}{p{0.95\linewidth}}{\footnotesize Counts of eligible works (Part III) per publication year, summed over the pre/baseline/post period labels of the result file. `all' counts every work to which the multi-label classifier assigns the category (a work can carry several categories); `prim.' counts works for which the category is the argmax of the assignment score (Part IV). Cyber. = cybersecurity; Crypto. = applied cryptography; Blockch. = blockchain systems; Privacy = privacy-preserving computation; Fed. learn. = federated learning. The 2026 row is incomplete (the corpus ends on 2026-11-02 and recent months are under-indexed).}\\
23+\end{longtable}
added manuscript/tables/tab_security_landmarks.tex +16 −0
@@ -0,0 +1,16 @@
1+% generated by tables/make_tables.py from ./literature_synthesis/landmarks/security_privacy_landmarks.csv
2+\begin{table}[htbp]\centering\caption{Composition of the automatically selected landmark set of the security-and-privacy area, by category.}\label{tab:security_landmarks}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrrrrrrrrr}
5+\toprule
6+\textbf{Category} & \textbf{Works} & \textbf{Post-period} & \textbf{Dated 2026} & \textbf{Articles} & \textbf{Conf.} & \textbf{Preprints} & \textbf{Grey} & \textbf{Abstract} & \textbf{No DOI} \\
7+\midrule
8+Cybersecurity & 25 & 16 & 9 & 13 & 1 & 1 & 10 & 20 & 5 \\
9+Applied cryptography & 25 & 15 & 8 & 11 & 2 & 0 & 12 & 18 & 1 \\
10+Blockchain systems & 25 & 6 & 2 & 14 & 1 & 5 & 5 & 19 & 0 \\
11+Privacy-preserving computation & 25 & 14 & 8 & 16 & 3 & 1 & 5 & 21 & 1 \\
12+Federated learning & 25 & 15 & 6 & 18 & 4 & 0 & 3 & 14 & 2 \\
13+\bottomrule
14+\end{tabular}
15+\begin{tablenotes}\footnotesize\item Landmarks are the works with the highest field-normalised 36-month citation score (NCS36; 2025--2026 works ranked by 12-month citations) within each primary category, with a floor of fifteen per category, plus the 200 most cited works of the corpus overall (literature_synthesis/select_landmarks.py). Grey = dissertations, reports, book chapters and books as typed by OpenAlex. Abstract = works whose abstract is available in the metadata; the synthesis of Part VII reads these abstracts only, never full texts. The set is flagged for principal-investigator review; Part VII documents the records that are duplicates, misdated or misclassified.\end{tablenotes}
16+\end{threeparttable}\end{table}
added manuscript/tables/tab_security_llm_cooccurrence.tex +23 −0
@@ -0,0 +1,23 @@
1+% generated by tables/make_tables.py from ./results/graph/category_cooccurrence_yearly.csv, ./results/graph/category_counts_yearly.csv
2+\begin{longtable}{lrrrrrrrrrrrrrr}
3+\caption{Co-classification of security-and-privacy works with the LLM, AI-safety and AI-agents categories, by publication year: number of works carrying both categories and share of the security category's works.}\label{tab:security_llm_cooccurrence}\\
4+\toprule
5+\textbf{Year} & \textbf{Cyber. x LLM} & \textbf{Cyber. x LLM (\%)} & \textbf{Cyber. x AI safety} & \textbf{Cyber. x AI safety (\%)} & \textbf{Cyber. x AI agents} & \textbf{Cyber. x AI agents (\%)} & \textbf{Crypto. x LLM} & \textbf{Crypto. x LLM (\%)} & \textbf{Blockch. x LLM} & \textbf{Blockch. x LLM (\%)} & \textbf{Privacy x LLM} & \textbf{Privacy x LLM (\%)} & \textbf{Fed. learn. x LLM} & \textbf{Fed. learn. x LLM (\%)} \\
6+\midrule
7+\endfirsthead
8+\toprule
9+\textbf{Year} & \textbf{Cyber. x LLM} & \textbf{Cyber. x LLM (\%)} & \textbf{Cyber. x AI safety} & \textbf{Cyber. x AI safety (\%)} & \textbf{Cyber. x AI agents} & \textbf{Cyber. x AI agents (\%)} & \textbf{Crypto. x LLM} & \textbf{Crypto. x LLM (\%)} & \textbf{Blockch. x LLM} & \textbf{Blockch. x LLM (\%)} & \textbf{Privacy x LLM} & \textbf{Privacy x LLM (\%)} & \textbf{Fed. learn. x LLM} & \textbf{Fed. learn. x LLM (\%)} \\
10+\midrule
11+\endhead
12+2018 & 536 & 0.8 & 929 & 1.4 & 571 & 0.8 & 66 & 0.2 & 263 & 2.4 & 58 & 0.5 & 13 & 1.2 \\
13+2019 & 735 & 1.0 & 1{,}726 & 2.4 & 621 & 0.8 & 69 & 0.2 & 272 & 1.8 & 85 & 0.7 & 15 & 1.0 \\
14+2020 & 985 & 1.3 & 2{,}648 & 3.5 & 639 & 0.8 & 108 & 0.3 & 281 & 1.7 & 130 & 0.8 & 36 & 1.3 \\
15+2021 & 1{,}231 & 1.6 & 3{,}438 & 4.4 & 645 & 0.8 & 136 & 0.4 & 283 & 1.6 & 233 & 1.3 & 66 & 1.6 \\
16+2022 & 1{,}337 & 1.7 & 4{,}056 & 5.2 & 596 & 0.8 & 198 & 0.6 & 275 & 1.4 & 278 & 1.4 & 91 & 1.5 \\
17+2023 & 2{,}304 & 2.6 & 5{,}288 & 5.9 & 807 & 0.9 & 403 & 1.1 & 357 & 1.5 & 717 & 2.8 & 226 & 2.7 \\
18+2024 & 4{,}977 & 4.6 & 7{,}850 & 7.3 & 1{,}233 & 1.1 & 839 & 2.0 & 508 & 1.9 & 2{,}022 & 5.8 & 601 & 5.0 \\
19+2025 & 10{,}549 & 7.6 & 12{,}933 & 9.3 & 3{,}207 & 2.3 & 1{,}745 & 3.4 & 1{,}033 & 3.1 & 4{,}584 & 9.3 & 1{,}374 & 7.6 \\
20+2026 & 14{,}410 & 12.0 & 14{,}029 & 11.6 & 6{,}329 & 5.3 & 2{,}423 & 5.5 & 1{,}067 & 4.3 & 5{,}665 & 13.6 & 1{,}990 & 11.7 \\
21+\bottomrule
22+\multicolumn{15}{p{0.95\linewidth}}{\footnotesize A work is co-classified when the multi-label classifier assigns both categories (Part IV). Shares are relative to all eligible works carrying the security category in the year (multi-label count). Non-zero counts before 2022 reflect the breadth of the keyword patterns of the AI categories (for example `language model' matches pre-transformer work), not the use of LLMs; the table should be read for its trend, not its level. The 2026 row is incomplete.}\\
23+\end{longtable}
added manuscript/tables/tab_security_summary.tex +16 −0
@@ -0,0 +1,16 @@
1+% generated by tables/make_tables.py from ./results/graph/category_counts_yearly.csv, ./results/econometrics/breaks.csv
2+\begin{table}[htbp]\centering\caption{Growth and structural-change summary for the five security-and-privacy categories (multi-label counts).}\label{tab:security_summary}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrrrrrrrrrrr}
5+\toprule
6+\textbf{Category} & \textbf{Works 2018} & \textbf{Works 2025} & \textbf{CAGR 18--22 (\%)} & \textbf{CAGR 23--25 (\%)} & \textbf{Level 2022-12} & \textbf{p} & \textbf{Slope change} & \textbf{p } & \textbf{Placebo p (slope)} & \textbf{Bai--Perron} & \textbf{PELT} \\
7+\midrule
8+Cybersecurity & 67{,}575 & 139{,}134 & 3.7 & 24.7 & -0.021 & 0.408 & 0.015 & 0.000 & 0.062 & 2022-07-01 & 2022-11, 2024-12 \\
9+Applied cryptography & 29{,}699 & 51{,}759 & 1.9 & 20.4 & 0.002 & 0.922 & 0.013 & 0.000 & 0.046 & 2022-07-01 & 2022-11, 2024-12 \\
10+Blockchain systems & 10{,}872 & 33{,}426 & 16.7 & 17.3 & -0.008 & 0.858 & -0.002 & 0.200 & 0.508 & 2020-04-01 & 2019-02, 2022-12 \\
11+Privacy-preserving computation & 11{,}378 & 49{,}266 & 15.2 & 39.8 & -0.001 & 0.965 & 0.012 & 0.000 & 0.138 & 2022-07-01 & 2020-12, 2023-12 \\
12+Federated learning & 1{,}040 & 18{,}093 & 54.3 & 45.8 & -0.126 & 0.007 & -0.006 & 0.001 & 0.385 & 2020-02-01 & 2020-02, 2023-05 \\
13+\bottomrule
14+\end{tabular}
15+\begin{tablenotes}\footnotesize\item Works: eligible works carrying the category (multi-label). CAGR: compound annual growth of the multi-label count between the years shown (2026 excluded as incomplete). Level and slope change: interrupted-time-series estimates on the log monthly series at the 2022-12 reference month with HAC p-values (Part IV, Section~\ref{sec:m-its}); Placebo p: share of placebo event months whose slope t-statistic exceeds the one at 2022-12 (Section~\ref{sec:m-placebo}); Bai--Perron: BIC-selected break months (Section~\ref{sec:m-bp}); PELT: penalised change-point months. The ITS, placebo and break columns are those of Table~\ref{tab:breaks}.\end{tablenotes}
16+\end{threeparttable}\end{table}
added manuscript/tables/tab_systems_categories.tex +22 −0
@@ -0,0 +1,22 @@
1+% generated by tables/make_tables.py from ./results/bibliometrics/cagr_by_category.csv, ./results/econometrics/breaks.csv
2+\begin{table}[htbp]\centering\caption{The systems and infrastructure categories of Part VI: primary-category output, compound annual growth, and structural-change estimates of the monthly log-volume series.}\label{tab:systems_categories}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrrrrrrrrr}
5+\toprule
6+\textbf{Category} & \textbf{n 2018} & \textbf{n 2022} & \textbf{n 2025} & \textbf{CAGR base} & \textbf{CAGR post} & \textbf{Slope ch.} & \textbf{p} & \textbf{Placebo p} & \textbf{BP breaks} \\
7+\midrule
8+Computer architecture & 19{,}554 & 15{,}174 & 20{,}209 & -0.061 & 0.111 & 0.016 & 0.000 & 0.062 & 2023-05-01 \\
9+Accelerators and AI hardware & 818 & 1{,}421 & 1{,}906 & 0.148 & 0.092 & 0.018 & 0.000 & 0.169 & 2024-11-01 \\
10+Operating systems & 1{,}556 & 1{,}083 & 1{,}363 & -0.087 & 0.078 & 0.016 & 0.000 & 0.477 & 2023-09-01 \\
11+Distributed computing & 11{,}519 & 9{,}379 & 12{,}550 & -0.050 & 0.132 & 0.020 & 0.000 & 0.077 & 2022-07-01;2024-07-01 \\
12+Cloud computing & 9{,}396 & 7{,}103 & 10{,}974 & -0.068 & 0.198 & 0.019 & 0.000 & 0.231 & 2022-11-01;2025-01-01 \\
13+Edge computing & 5{,}185 & 8{,}554 & 11{,}554 & 0.133 & 0.126 & 0.004 & 0.006 & 0.077 & 2025-01-01 \\
14+Computer networks & 58{,}744 & 49{,}141 & 59{,}555 & -0.044 & 0.071 & 0.009 & 0.000 & 0.200 & 2022-02-01 \\
15+Database systems & 16{,}265 & 13{,}625 & 16{,}681 & -0.043 & 0.073 & 0.014 & 0.000 & 0.092 & 2023-11-01 \\
16+Information retrieval & 12{,}095 & 11{,}243 & 17{,}379 & -0.018 & 0.176 & 0.023 & 0.000 & 0.092 & 2022-07-01;2024-07-01 \\
17+Quantum computing & 6{,}224 & 9{,}216 & 17{,}471 & 0.103 & 0.287 & 0.018 & 0.000 & 0.169 & 2022-01-01;2025-04-01 \\
18+Sustainable computing & 583 & 830 & 1{,}639 & 0.092 & 0.233 & 0.016 & 0.000 & 0.169 & 2022-07-01 \\
19+\bottomrule
20+\end{tabular}
21+\begin{tablenotes}\footnotesize\item Counts are eligible works whose primary category is the row category (multi-label memberships are larger). CAGR base: 2018 to 2022; CAGR post: 2023 to 2025 (complete years). Slope ch.: change in the monthly trend of the log-volume series at 2022-12 in the interrupted time-series model of Part IV, with its HAC p-value; Placebo p: share of placebo event dates whose slope statistic is at least as extreme; BP breaks: Bai--Perron break dates selected by BIC. 2026 is incomplete and excluded from the growth columns.\end{tablenotes}
22+\end{threeparttable}\end{table}
modified results/bibliometrics/cagr_by_category.csv +40 −40
@@ -1,41 +1,41 @@
1 category,n_2018,n_2022,n_2023,n_2025,cagr_baseline,cagr_post,share_2018,share_2022,share_2025 1 category,n_2018,n_2022,n_2023,n_2025,cagr_baseline,cagr_post,share_2018,share_2022,share_2025
2−ai_accelerators,409,710,799,953,0.14804838919248775,0.09178523385922643,0.002409752131363896,0.0027899579837432427,0.002218169672523209 2+ai_accelerators,818,1421,1599,1906,0.14804838919248775,0.09178523385922643,0.001637437670274501,0.002535539484648477,0.0021347609874109584
3−ai_agents,2718,2260,2495,4409,-0.045032119445381436,0.3292782120448039,0.016013951816741003,0.00887642508409796,0.010263398920242486 3+ai_agents,5436,4521,4991,8819,-0.045032119445381436,0.3292782120448039,0.010881554004415877,0.008066976784022354,0.00987746964741723
4−ai_safety,526,1477,2007,5442,0.2944899640897396,0.6465368010192893,0.0030990944281110255,0.005799814133692849,0.01266777381186523 4+ai_safety,1052,2954,4015,10885,0.2944899640897396,0.6465368010192893,0.0021058489353652505,0.005270924445919495,0.01219143407553425
5−algorithms,9716,9400,9510,11265,-0.008214426291722088,0.08834786213446155,0.05724486970252229,0.03691409573423735,0.02622002241445325 5+algorithms,29148,28202,28531,33795,-0.008214426291722088,0.08834786213446155,0.058347228866945176,0.050321804747400674,0.03785112674163344
6−applied_cryptography,3594,5149,3739,7306,0.09407412599520226,0.3977303237525569,0.0211751813205913,0.020220814408565557,0.01700519163426502 6+applied_cryptography,10782,10299,11219,14612,-0.01139242765304549,0.14124203040296623,0.021582949829950696,0.01837686217621018,0.01636575422248107
7−autonomous_scientific_discovery,43,118,200,382,0.28707361667658393,0.38202749610852527,0.0002533480235908253,0.0004633568502205526,0.0008891299212002788 7+autonomous_scientific_discovery,86,236,400,764,0.28707361667658393,0.38202749610852527,0.00017215114870856613,0.00042110296859749515,0.000855696429371444
8−blockchain,2147,5885,7174,9006,0.2867038655679659,0.12039239265851642,0.012649725736034926,0.02310894121650807,0.020962052540130133 8+blockchain,6441,11770,14349,18012,0.1626681909797465,0.12039239265851642,0.012893320335254354,0.02100161839149372,0.02017382733748488
9−cloud_computing,3132,3551,3821,3658,0.03192353243026824,-0.0216259646860103,0.01845316302061546,0.013945863165748243,0.00851423364332623 9+cloud_computing,9396,7103,7643,10974,-0.06755182852386099,0.19825858205350233,0.01880851387518241,0.012674128754016984,0.012291115989426997
10−compilers,241,321,346,383,0.073549490456021,0.05279699035007379,0.0014238551713437854,0.001260487702718622,0.0008926212690583952 10+compilers,725,642,692,767,-0.029938529449611395,0.05279699035007379,0.0014512742187640748,0.001145542821354203,0.0008590564938846826
11−computational_biology,4322,9768,10288,14077,0.22606494630771068,0.1697406366696017,0.025468350154463738,0.03835652299113863,0.032765135865801896 11+computational_biology,12968,19536,20576,28154,0.10787474203944436,0.1697406366696017,0.025958791819217272,0.034858760993731634,0.03153308543524036
12−computational_complexity,1279,1770,1166,2104,0.08469177413199191,0.34326939885786634,0.007535630748201524,0.006952316129792274,0.004898361044937139 12+computational_complexity,3837,3541,3499,4209,-0.019870339595214892,0.0967748714322898,0.007680743692962421,0.006318328863575129,0.004714170512073831
13−computer_architecture,6518,7587,8184,6736,0.03869713083776305,-0.0927735404534441,0.03840284692476742,0.029792274767994346,0.015679255303260413 13+computer_architecture,19554,15174,16369,20209,-0.06143119143120024,0.1111209535203932,0.03914236699822444,0.027075493413128777,0.02263451458267999
14−computer_graphics,3399,4143,3099,6220,0.05073548076582268,0.4167212372990894,0.02002824143084679,0.016270500922786948,0.014477455784988833 14+computer_graphics,10198,8287,9297,12440,-0.050553314316049036,0.15674804638240403,0.020413923424766944,0.014786780935455265,0.013933067514896286
15−computer_networks,19581,16380,17293,29777,-0.04364282894058291,0.3122016357047006,0.11536958370402665,0.06432152253301744,0.06930907389670499 15+computer_networks,58744,49141,51881,59555,-0.04364282894058291,0.07140814904065751,0.1175912451132094,0.08768398720275215,0.06670288069530934
16−computer_vision,23246,49352,57130,75378,0.20708855155154882,0.1486553322088684,0.13696111991610055,0.19379311247529418,0.17544837634867858 16+computer_vision,69738,98704,114261,150757,0.09072763371648507,0.1486553322088684,0.13959856754230215,0.17612096361206425,0.16885108194077325
17−cybersecurity,9813,15777,12160,29522,0.12603520471872232,0.5581172943745198,0.057818339647394544,0.06195238157567507,0.06871438097820584 17+cybersecurity,29440,31554,36481,59044,0.017487667119059358,0.2721974435411547,0.05893174206953705,0.056302894369175266,0.06613054970655437
18−database_systems,5421,6812,7243,8340,0.05875000550434839,0.07305457648307523,0.03194345429228506,0.02675100459430097,0.01941305787374588 18+database_systems,16265,13625,14487,16681,-0.043311373704336575,0.07305457648307523,0.032558586438893346,0.02431155909805454,0.01868307871511133
19−deep_learning_architectures,3997,7423,12803,17070,0.16736629336390108,0.15467235481054153,0.023551546503109897,0.029149596198903123,0.039732702407982615 19+deep_learning_architectures,11992,22270,25607,34141,0.16736629336390108,0.15467235481054153,0.02400507645712936,0.03973713182485685,0.0382386541821603
20−distributed_computing,3839,4689,3266,4183,0.0512554351002068,0.13175688541453123,0.022622603749943536,0.018414508043298996,0.009736981248746508 20+distributed_computing,11519,9379,9798,12550,-0.050083482537713886,0.13175688541453123,0.023058245139232245,0.01673527433252503,0.014056269880381703
21−edge_computing,1728,4277,3040,5777,0.25423217055612524,0.3785241917965825,0.010183019397817279,0.016794722444011048,0.01344634438422516 21+edge_computing,5185,8554,9120,11554,0.1333267023064395,0.1255602893280665,0.010379112861092039,0.015263198277046497,0.01294072846198647
22−federated_learning,71,487,748,1272,0.6154954153196714,0.303609910055608,0.0004212647369010234,0.0019123286954017722,0.002960662983682604 22+federated_learning,143,974,1497,2544,0.6154954153196714,0.303609910055608,0.00028625132866656923,0.001737941912771018,0.002849334707226379
23−formal_verification,1769,1997,1461,2311,0.030836383835269654,0.2575428068501022,0.010422619854236509,0.007843689053521643,0.005380167049357185 23+formal_verification,5307,3995,4385,4623,-0.06853417817252339,0.026779402163364052,0.010623327281353028,0.0071284167777414965,0.0051778594149007665
24−foundation_models,23,175,285,927,0.6608390716251988,0.8019254150494273,0.00013551173354858095,0.0006871817693948874,0.00215765297631586 24+foundation_models,46,350,571,1854,0.6608390716251988,0.8019254150494273,9.208084698365164e-05,0.0006245171144454377,0.0020765198691814883
25−generative_ai,255,1080,1823,6371,0.43409875951610655,0.8695087811589706,0.0015043766362059857,0.004240893205408448,0.014830081918658578 25+generative_ai,766,2160,3646,12743,0.29585451247314687,0.8695087811589706,0.0015333462780321122,0.0038541627634347013,0.01427243403073339
26−hci,11408,17400,20293,29074,0.11130912772658119,0.19694422895345198,0.06721578377826352,0.06832746501917564,0.06767163175124845 26+hci,34225,34801,40587,58148,0.004181154443647861,0.19694422895345198,0.06851015191337995,0.06209662885661622,0.06512701043860042
27−information_retrieval,4031,5621,6284,8689,0.08665554196594227,0.1759241392022266,0.023753832134349083,0.022074241809447767,0.0202253781420676 27+information_retrieval,12095,11243,12568,17379,-0.018095875045832766,0.1759241392022266,0.024211257484071016,0.02006127405060016,0.019464853725191522
28−llm,545,1088,5194,22579,0.1888432761609986,1.0848771832846835,0.0032110389036511576,0.004274925064464195,0.052554095525604964 28+llm,1635,3266,10389,45158,0.1888432761609986,1.0848771832846835,0.003272873583005879,0.005827636845082284,0.05057793109627705
29−ml_theory,4511,10065,11446,16399,0.22217278554507214,0.19693985094098543,0.026579939157195573,0.039524731999109934,0.038169742350165896 29+ml_theory,13534,20131,22893,32798,0.10435777777595234,0.19693985094098543,0.02709178658862481,0.035920440088288874,0.03673446530173379
30−multimodal_ai,156,639,966,3374,0.421121021734181,0.8690294865336103,0.000923050938664247,0.0025091951465333316,0.007854368898142254 30+multimodal_ai,470,1278,1932,6749,0.2841278026110756,0.8690294865336103,0.000940826045267745,0.002280379635032198,0.007559025133282559
31−nlp,7790,9892,16982,21457,0.06153590511416507,0.12404497631921241,0.045901162847788904,0.038846058194478984,0.049942567327733986 31+nlp,23372,29678,33965,42914,0.06153590511416507,0.12404497631921241,0.04678507729786753,0.052955482635747717,0.04806460284037453
32−operating_systems,778,541,586,681,-0.0866132316020718,0.07795075704417309,0.004583831682643304,0.00212633673215618,0.0015862357102041625 32+operating_systems,1556,1083,1173,1363,-0.0866132316020718,0.07795075704417309,0.003114734737099173,0.0019324343855554546,0.001526589310514762
33−privacy_preserving_computation,1035,2188,2505,4408,0.20580406040260257,0.3266047789454696,0.006098028009686143,0.008591735493920076,0.010261071355003742 33+privacy_preserving_computation,3105,4376,5010,8817,0.08956696494067806,0.3266047789454696,0.006215457171396486,0.007808248265180673,0.009875229604408405
34−programming_languages,1463,1832,1938,2096,0.057780486211923554,0.04008910470076277,0.008621688554757541,0.007193811437322478,0.0048797405230271845 34+programming_languages,4390,3664,3876,4193,-0.044187433279510024,0.04008910470076277,0.008787715614309364,0.006537802020937383,0.004696250168003226
35−quantum_computing,3112,4608,5272,8735,0.10310847324827721,0.2872300901577727,0.018335326730573216,0.018094477676409378,0.020332446143049833 35+quantum_computing,6224,9216,10544,17471,0.10310847324827721,0.2872300901577727,0.012458938948396692,0.016444427790654727,0.01956789570359751
36−reinforcement_learning,1648,3508,4211,6605,0.20792805552654703,0.25240210895212,0.009709710299480931,0.013777012788125499,0.015373568401905345 36+reinforcement_learning,3296,7017,8422,13210,0.20792805552654703,0.25240210895212,0.006597792862132953,0.012520675977324676,0.014795484073294207
37−robotics,5832,11551,12879,16732,0.18632919081853316,0.1397889079167418,0.034361062176318444,0.04535988690951452,0.038944821574667714 37+robotics,17496,23103,25759,33464,0.07196943376447251,0.1397889079167418,0.035022749974477595,0.04122348255723699,0.03748039962367278
38−scientific_computing,5627,6644,6807,7730,0.04239362249738088,0.06561412676886369,0.033155204141552805,0.026089346719197895,0.017992855150572133 38+scientific_computing,16882,19932,20423,23191,0.04239362249738088,0.06561412676886369,0.033793670843000154,0.03556535750036133,0.02597441870883921
39−software_engineering,9672,11862,13122,18190,0.05236158017872561,0.17737287914735167,0.05698562986442935,0.04658110708255343,0.042339575475376104 39+software_engineering,29016,23725,26245,36381,-0.04908396762848111,0.17737287914735167,0.05808299687125296,0.04233333868633717,0.04074750235204516
40−speech_audio,4032,6349,7344,9186,0.12020163164106301,0.11842973684211522,0.023755796072516454,0.024930954593646512,0.02138217806572346 40+speech_audio,12096,12698,14688,18373,0.012216438797881812,0.11842973684211522,0.024213259241614136,0.02265748091208048,0.020578155100577932
41−sustainable_computing,291,276,539,819,-0.012971780149642509,0.23304832154514776,0.0017174639273657107,0.0010864016544719173,0.00190743971315086 41+sustainable_computing,583,830,1078,1639,0.09232628514238472,0.23304832154514776,0.0011670246476406284,0.001480997728542038,0.0018357152457327182
modified results/bibliometrics/category_share_ci_yearly.csv +320 −320
@@ -1,321 +1,321 @@
1 year,category,share,ci_low,ci_high 1 year,category,share,ci_low,ci_high
2−2018,ai_accelerators,0.002409752131363896,0.0021650945341636866,0.0026720851720704424 2+2018,ai_accelerators,0.001637437670274501,0.0015132286147237274,0.0017456326654802915
3−2018,ai_agents,0.016013951816741003,0.015468811679933069,0.016632592339462786 3+2018,ai_agents,0.010881554004415877,0.010604260540754784,0.011113757879418129
4−2018,ai_safety,0.0030990944281110255,0.0028395599992929823,0.003373211097821796 4+2018,ai_safety,0.0021058489353652505,0.0019826908025246167,0.0022380650210885155
5−2018,algorithms,0.05724486970252229,0.0561930335185327,0.05852059483759213 5+2018,algorithms,0.058347228866945176,0.05766232752356569,0.05901791773176849
6−2018,applied_cryptography,0.0211751813205913,0.020432518102600057,0.021796915045926693 6+2018,applied_cryptography,0.021582949829950696,0.021211573761762827,0.02194957172397365
7−2018,autonomous_scientific_discovery,0.0002533480235908253,0.00017675443506336646,0.0003358334266203963 7+2018,autonomous_scientific_discovery,0.00017215114870856613,0.00013907210530846083,0.00021123546473803995
8−2018,blockchain,0.012649725736034926,0.012084111543832154,0.0132153399282377 8+2018,blockchain,0.012893320335254354,0.01257243860109176,0.013200740249939447
9−2018,cloud_computing,0.01845316302061546,0.01784041431239579,0.019110247632963522 9+2018,cloud_computing,0.01880851387518241,0.01841101487105679,0.019190949653796033
10−2018,compilers,0.0014238551713437854,0.0012431728599456775,0.0016025735445745226 10+2018,compilers,0.0014512742187640748,0.001354138933984038,0.0015513620959202179
11−2018,computational_biology,0.025468350154463738,0.02472794546536497,0.02618042503549818 11+2018,computational_biology,0.025958791819217272,0.02551665362188001,0.026400329489291597
12−2018,computational_complexity,0.007535630748201524,0.007137785973946396,0.007889139618328257 12+2018,computational_complexity,0.007680743692962421,0.0074505415755032915,0.007914949325507795
13−2018,computer_architecture,0.03840284692476742,0.037570922717069175,0.03931607817259481 13+2018,computer_architecture,0.03914236699822444,0.038611801161419726,0.03963489944170982
14−2018,computer_graphics,0.02002824143084679,0.019395853340953412,0.020706929363035934 14+2018,computer_graphics,0.020413923424766944,0.02001547358580834,0.020802464563887093
15−2018,computer_networks,0.11536958370402665,0.113782427074066,0.11679712125943426 15+2018,computer_networks,0.1175912451132094,0.11667899415686973,0.11851590696631642
16−2018,computer_vision,0.13696111991610055,0.135278859580385,0.1385552976250096 16+2018,computer_vision,0.13959856754230215,0.13864142717305794,0.14057162188401418
17−2018,cybersecurity,0.057818339647394544,0.05672211846082238,0.05889516694456392 17+2018,cybersecurity,0.05893174206953705,0.05832616036880381,0.05954638172315293
18−2018,database_systems,0.03194345429228506,0.031152821884555787,0.0328440672373871 18+2018,database_systems,0.032558586438893346,0.03206815584082825,0.03305016804754574
19−2018,deep_learning_architectures,0.023551546503109897,0.02283947162207545,0.02420666717729059 19+2018,deep_learning_architectures,0.02400507645712936,0.023564589709765174,0.024445563204493544
20−2018,distributed_computing,0.022622603749943536,0.021893982689848993,0.023379014535106376 20+2018,distributed_computing,0.023058245139232245,0.022629368585618172,0.02350398650014713
21−2018,edge_computing,0.010183019397817279,0.009712508911369435,0.010628833361810437 21+2018,edge_computing,0.010379112861092039,0.010085805337085961,0.010655355402042994
22−2018,federated_learning,0.0004212647369010234,0.00032404979761617187,0.000518479676185875 22+2018,federated_learning,0.00028625132866656923,0.0002442144202609891,0.000341349705041026
23−2018,formal_verification,0.010422619854236509,0.00994228967695181,0.010988823227889492 23+2018,formal_verification,0.010623327281353028,0.010356042605407548,0.010894615472384754
24−2018,foundation_models,0.00013551173354858095,8.837721753168323e-05,0.0001885380640675909 24+2018,foundation_models,9.208084698365164e-05,6.605799892305444e-05,0.00011810369504424884
25−2018,generative_ai,0.0015043766362059857,0.0013225650603616397,0.0016763685212134779 25+2018,generative_ai,0.0015333462780321122,0.0014250511949491652,0.0016344850778983947
26−2018,hci,0.06721578377826352,0.06605593099506855,0.0683425442033383 26+2018,hci,0.06851015191337995,0.0678100872566113,0.06915541845740561
27−2018,information_retrieval,0.023753832134349083,0.023113293701061115,0.024454123386379305 27+2018,information_retrieval,0.024211257484071016,0.02380680237248304,0.02469873548976001
28−2018,llm,0.0032110389036511576,0.00295179906555822,0.0034584951127398706 28+2018,llm,0.003272873583005879,0.003112632891678894,0.0034310124289125853
29−2018,ml_theory,0.026579939157195573,0.025785084282406452,0.027376757970152068 29+2018,ml_theory,0.02709178658862481,0.026672268251524838,0.027592626325914152
30−2018,multimodal_ai,0.000923050938664247,0.0007836113287809246,0.0010812952565001442 30+2018,multimodal_ai,0.000940826045267745,0.0008627575010859535,0.0010269016196220282
31−2018,nlp,0.045901162847788904,0.04488826173796744,0.046904735251315346 31+2018,nlp,0.04678507729786753,0.04615212156273208,0.04735272569315859
32−2018,operating_systems,0.004583831682643304,0.0042625804969156356,0.004941022936833856 32+2018,operating_systems,0.003114734737099173,0.0029674554258639087,0.003269920990629773
33−2018,privacy_preserving_computation,0.006098028009686143,0.005709168252546737,0.006472305525932821 33+2018,privacy_preserving_computation,0.006215457171396486,0.005999267356739217,0.00644165577376937
34−2018,programming_languages,0.008621688554757541,0.008177838528931756,0.009052920277857972 34+2018,programming_languages,0.008787715614309364,0.00853634491083171,0.009009910701596001
35−2018,quantum_computing,0.018335326730573216,0.017710499802624215,0.018983426325805557 35+2018,quantum_computing,0.012458938948396692,0.012182596319568582,0.012739285092311049
36−2018,reinforcement_learning,0.009709710299480931,0.009264731009208906,0.010163380016143572 36+2018,reinforcement_learning,0.006597792862132953,0.006364538064420561,0.006824091552382992
37−2018,robotics,0.034361062176318444,0.03349776405639645,0.035212576667236206 37+2018,robotics,0.035022749974477595,0.03457415610906376,0.03550672490446612
38−2018,scientific_computing,0.033155204141552805,0.032280957066347725,0.03399105622558579 38+2018,scientific_computing,0.033793670843000154,0.033273113793911055,0.03426913830343041
39−2018,software_engineering,0.05698562986442935,0.055954267735834604,0.05796985747700719 39+2018,software_engineering,0.05808299687125296,0.05749533090053067,0.05869928797484191
40−2018,speech_audio,0.023755796072516454,0.023063360573155715,0.02449286206673069 40+2018,speech_audio,0.024213259241614136,0.023774674163915918,0.024626772306084743
41−2018,sustainable_computing,0.0017174639273657107,0.0015551444378325194,0.001938406971194919 41+2018,sustainable_computing,0.0011670246476406284,0.0010709402855707312,0.0012691142823398945
42−2019,ai_accelerators,0.0018345044017934501,0.0016421773274118785,0.002011251133851322 42+2019,ai_accelerators,0.0018379185572339326,0.0017257814396164086,0.0019417646611986118
43−2019,ai_agents,0.0101822391494705,0.009719637056619056,0.01066069897946446 43+2019,ai_agents,0.010201189088840759,0.009925038120135399,0.010464370874178541
44−2019,ai_safety,0.0027351129135610003,0.0025187449548817326,0.0029736724577458344 44+2019,ai_safety,0.0027402031715211554,0.00261041870082132,0.002881104085107487
45−2019,algorithms,0.05409568826740121,0.05305952615417049,0.055143454730552544 45+2019,algorithms,0.05419636455262802,0.05364415525224135,0.05481429982158109
46−2019,applied_cryptography,0.02099878778464176,0.020337755931394714,0.02167355068142037 46+2019,applied_cryptography,0.021037868162692847,0.020641289062717116,0.02140109793400909
47−2019,autonomous_scientific_discovery,0.0006324601869086289,0.0005270501557571907,0.0007434181144364586 47+2019,autonomous_scientific_discovery,0.00021121241484341565,0.00017601034570284636,0.0002464144839839849
48−2019,blockchain,0.01673060617240458,0.016130092622130006,0.01733773096586102 48+2019,blockchain,0.01676174313235422,0.01644112639210288,0.017132383865594795
49−2019,cloud_computing,0.016192460223894603,0.015625511446697198,0.01676310759867627 49+2019,cloud_computing,0.016222595652359187,0.01587770801180566,0.016573921566084473
50−2019,compilers,0.0019112503016668655,0.0017198478766813593,0.0021111132686263685 50+2019,compilers,0.0012765381914659067,0.0011829284786592088,0.0013776515032209893
51−2019,computational_biology,0.02751016883173323,0.026732122597414126,0.028275269974507414 51+2019,computational_biology,0.027561367396584657,0.02711017877092902,0.02794400462444024
52−2019,computational_complexity,0.007308428826499711,0.0069264099285153554,0.007689661772497413 52+2019,computational_complexity,0.007322030381238409,0.007091317872831599,0.007544451875992374
53−2019,computer_architecture,0.03485188503649129,0.03403634426916174,0.03573136530955875 53+2019,computer_architecture,0.03491674710648256,0.03438111983340157,0.035438525144467437
54−2019,computer_graphics,0.018119429565294578,0.017497787776319915,0.018713331872387287 54+2019,computer_graphics,0.01815315123364725,0.01780279800867453,0.01851276816102538
55−2019,computer_networks,0.10781227028241566,0.10634998349500828,0.10915527729773063 55+2019,computer_networks,0.10801291730663411,0.10721031013022912,0.10884544624180857
56−2019,computer_vision,0.15103223235329744,0.1493210761809391,0.15265619408760683 56+2019,computer_vision,0.1513133150900154,0.15043252226530873,0.1521450566104854
57−2019,cybersecurity,0.058365719178430524,0.05710713265297631,0.05944099396111479 57+2019,cybersecurity,0.058474342323447726,0.05786344746832277,0.05903998399232224
58−2019,database_systems,0.03065952467473146,0.029935894057370797,0.03136517086133865 58+2019,database_systems,0.030716584435868315,0.030259744951745375,0.031185420414606264
59−2019,deep_learning_architectures,0.028410777343500778,0.02762824655961076,0.029118272828900102 59+2019,deep_learning_architectures,0.028463652010871882,0.0280585966231952,0.028890755010273446
60−2019,distributed_computing,0.02125768961554003,0.0205742350144384,0.021947478065004702 60+2019,distributed_computing,0.02129725183004441,0.020941433020652496,0.021700361840215954
61−2019,edge_computing,0.012297836967667784,0.011822567178090248,0.012776805354829583 61+2019,edge_computing,0.012320724199199246,0.012038042320298068,0.012611789728777321
62−2019,federated_learning,0.00119834561730056,0.0010374566223852071,0.0013592346122159131 62+2019,federated_learning,0.00040019194391384015,0.00035016795092461013,0.0004623050685421341
63−2019,formal_verification,0.008880332799810632,0.008463177224775796,0.009334474350688078 63+2019,formal_verification,0.008896859790158614,0.008633770641844885,0.009142394222414084
64−2019,foundation_models,0.000153491799746831,9.986213477504667e-05,0.0002108200623028763 64+2019,foundation_models,0.00015377745992985526,0.00011945544251780021,0.00018897952907042452
65−2019,generative_ai,0.002341212270837205,0.002130392208534329,0.0025607702699328976 65+2019,generative_ai,0.0023455694490505633,0.0022156923413266735,0.002481792192922161
66−2019,hci,0.06881610665275995,0.06770916262826043,0.06997852964102336 66+2019,hci,0.06894417878204546,0.06825028115336801,0.0695973161201391
67−2019,information_retrieval,0.022753772338373595,0.02208035018321928,0.023462331170578395 67+2019,information_retrieval,0.02279611887924023,0.02239310150609274,0.023232670855095314
68−2019,llm,0.003434147856986327,0.003170484081698322,0.003681167943145158 68+2019,llm,0.0034405390733703756,0.003288521716897549,0.0035888509488810373
69−2019,ml_theory,0.031541640198577706,0.030632571144836158,0.03241108802569785 69+2019,ml_theory,0.03160034164534471,0.031083936554754964,0.03207376315677334
70−2019,multimodal_ai,0.001168756836626472,0.001017900287658427,0.0013314951303339557 70+2019,multimodal_ai,0.001170931984044199,0.0010801477004711518,0.0012654217485794112
71−2019,nlp,0.04759725231185465,0.04664934798347837,0.04857211017012624 71+2019,nlp,0.047685834502103785,0.0471130597823771,0.04822322187864179
72−2019,operating_systems,0.0029200427927740503,0.0026851818461734772,0.003167848830919536 72+2019,operating_systems,0.002925477219629415,0.0027892544757578174,0.003072816406387508
73−2019,privacy_preserving_computation,0.006021316867176888,0.005650393761945314,0.006388541374824201 73+2019,privacy_preserving_computation,0.006032523006404924,0.005860032867616135,0.006264393477612411
74−2019,programming_languages,0.00827376279599183,0.007886196001631082,0.008679822578273884 74+2019,programming_languages,0.008289160912363522,0.008050157390303868,0.008565311881068882
75−2019,quantum_computing,0.013836453562720355,0.013309403406963165,0.01446419803770905 75+2019,quantum_computing,0.013862204279459963,0.013575909556620675,0.014174622643082516
76−2019,reinforcement_learning,0.008767525573490672,0.008332662962521185,0.00917067271017512 76+2019,reinforcement_learning,0.008783842620812575,0.008566099295773342,0.009014555129219386
77−2019,robotics,0.037087687276177055,0.03629142544875547,0.03792569702383099 77+2019,robotics,0.03715671034811141,0.0366456782049168,0.03765695027800371
78−2019,scientific_computing,0.03388655106699917,0.033137585056186324,0.0346579860581364 78+2019,scientific_computing,0.03394961657535744,0.03344090035776419,0.034429105811861614
79−2019,software_engineering,0.05350946055029584,0.05239849430092345,0.05463013561832692 79+2019,software_engineering,0.05360904582012484,0.053033723582236666,0.054117067260037684
80−2019,speech_audio,0.023639586459804105,0.022995753085323874,0.024347359340022245 80+2019,speech_audio,0.02368358156967879,0.02329441343162739,0.024124719078224555
81−2019,sustainable_computing,0.0012334822943510396,0.0010818397933963389,0.0014065304288246503 81+2019,sustainable_computing,0.0012357779008820898,0.001149579149999722,0.0013312403541698703
82−2020,ai_accelerators,0.002864752020728307,0.0026244824964091586,0.003081390641282 82+2020,ai_accelerators,0.001975957064365208,0.0018529833309354747,0.0020898250048715992
83−2020,ai_agents,0.013542224068493536,0.012992904568113329,0.014120719153969638 83+2020,ai_agents,0.009340722380764194,0.009097552180746347,0.00961685555116454
84−2020,ai_safety,0.005687258850367532,0.005354577970541019,0.006006870124200861 84+2020,ai_safety,0.003922775591375722,0.0037433459418212676,0.0040912782894221645
85−2020,algorithms,0.051708113893035174,0.050675350987129296,0.05261163291745289 85+2020,algorithms,0.05349835458321875,0.05282388850138681,0.05411090127317197
86−2020,applied_cryptography,0.019853479595572703,0.019189614019770748,0.02050753196552792 86+2020,applied_cryptography,0.020540847676474818,0.020194645429529357,0.02089328739157277
87−2020,autonomous_scientific_discovery,0.0010191652350240797,0.0008765877150984312,0.0011567261385078994 87+2020,autonomous_scientific_discovery,0.0003514836068410001,0.00030595464222429023,0.00040247604721171514
88−2020,blockchain,0.02558738417776776,0.024866575604810315,0.02635400237981243 88+2020,blockchain,0.01764884784402141,0.017297273179251175,0.017973105130021618
89−2020,cloud_computing,0.014405610161376628,0.013856158644774269,0.014960342326865123 89+2020,cloud_computing,0.014904361856926139,0.01457181829936569,0.015266362654593598
90−2020,compilers,0.0019459191145407945,0.0017531754301968623,0.0021597853944292675 90+2020,compilers,0.0013421938769006066,0.001241074046486894,0.0014451348658989875
91−2020,computational_biology,0.03182119018785028,0.030902357281662768,0.03264523544653167 91+2020,computational_biology,0.03292290489363523,0.03243474333501487,0.033376646555005364
92−2020,computational_complexity,0.006815557495705072,0.006442391641084856,0.007158051581378319 92+2020,computational_complexity,0.007051526039836023,0.006846509112166979,0.007273707387165567
93−2020,computer_architecture,0.03078794322246317,0.030081084363646603,0.031549632818880784 93+2020,computer_architecture,0.031853884804434886,0.031391128408070644,0.03230102476593559
94−2020,computer_graphics,0.01662524290984876,0.01601343573098262,0.01723656602923367 94+2020,computer_graphics,0.017200842832192986,0.016871031012509537,0.017515994325269853
95−2020,computer_networks,0.0957364040893345,0.09446024727518516,0.09708274952826203 95+2020,computer_networks,0.0990509942615293,0.09815539399855401,0.09981319465817765
96−2020,computer_vision,0.14793913876137096,0.14634024628946404,0.1494509445320641 96+2020,computer_vision,0.15306109440703988,0.15208864125179156,0.1539799144419697
97−2020,cybersecurity,0.05756787394034978,0.05661330864336609,0.058604905370771954 97+2020,cybersecurity,0.059560991511579835,0.0589581880200546,0.06014230533180599
98−2020,database_systems,0.026510617624693717,0.025774319148336944,0.027203394753147262 98+2020,database_systems,0.02742846944369069,0.02702503728822202,0.027842828550667366
99−2020,deep_learning_architectures,0.03152371363393133,0.0307570074210719,0.03233869377869152 99+2020,deep_learning_architectures,0.032615129092826274,0.03216156954731461,0.033086991282113856
100−2020,distributed_computing,0.019188117835919675,0.018590392387416565,0.019765468780803784 100+2020,distributed_computing,0.019852449731470167,0.019483437473251732,0.02025674693726655
101−2020,edge_computing,0.012523938941617144,0.012010703875292194,0.01300623820655082 101+2020,edge_computing,0.012957543329915626,0.012675946683761275,0.01326896144789392
102−2020,federated_learning,0.001605317261385079,0.0014441254541358041,0.0017929123130650292 102+2020,federated_learning,0.0005536322097391919,0.0004971762936144716,0.0006183288684595365
103−2020,formal_verification,0.007982580899540934,0.0076094150449207195,0.008356758878530993 103+2020,formal_verification,0.008258954181471169,0.008024434484730496,0.008492153538237956
104−2020,foundation_models,0.0002411496324668375,0.00017148907257723828,0.00030627763539583743 104+2020,foundation_models,0.00024949872609957,0.00021029828756458283,0.00028774305637760633
105−2020,generative_ai,0.002636804010476807,0.002397414594305348,0.0028673923451713744 105+2020,generative_ai,0.0027280955598332547,0.0025950599252232286,0.002869371937038905
106−2020,hci,0.0638096009237615,0.06280803785169121,0.06500254351254682 106+2020,hci,0.06601881985281396,0.06535860433690706,0.06667471011708229
107−2020,information_retrieval,0.020730067310671133,0.020134586137944627,0.021334349564874528 107+2020,information_retrieval,0.02144778465163968,0.02108259682644905,0.021837876820475652
108−2020,llm,0.004331892302926184,0.0040396963978933725,0.004641954403357083 108+2020,llm,0.0044818712768689185,0.004304262785899133,0.004667811568363561
109−2020,ml_theory,0.03398801644746106,0.03321501746134565,0.034762775649871853 109+2020,ml_theory,0.035164751111362025,0.03466211134199355,0.035677999129486195
110−2020,multimodal_ai,0.0019934449545160105,0.0017979289295068576,0.002233714478835159 110+2020,multimodal_ai,0.0013749747314246378,0.0012756760595955935,0.0014779157204230188
111−2020,nlp,0.050600937843242176,0.04966423874165657,0.05163246859774128 111+2020,nlp,0.05235284583346233,0.051698093793309426,0.052982192711359115
112−2020,operating_systems,0.0038390317402202377,0.0035775076041343955,0.004148081716281296 112+2020,operating_systems,0.0026479645821078455,0.002510421580000765,0.002780044108460921
113−2020,privacy_preserving_computation,0.007317219139887909,0.006954614583039963,0.0076702745233334264 113+2020,privacy_preserving_computation,0.007570556236466516,0.007346553730552303,0.007836536447757335
114−2020,programming_languages,0.007262652434731179,0.0068884744557411205,0.0076094150449207195 114+2020,programming_languages,0.007514100320341795,0.007289050648241398,0.007739149992442192
115−2020,quantum_computing,0.014255991776269466,0.013671599966203846,0.014735782732982228 115+2020,quantum_computing,0.014749563377229326,0.014436233042737128,0.015025514431771204
116−2020,reinforcement_learning,0.009989227476272284,0.009571044089897765,0.010416211944123692 116+2020,reinforcement_learning,0.010335074967993138,0.01009190476797529,0.010581887485180323
117−2020,robotics,0.03582568225983608,0.03507406990170952,0.036706054439969577 117+2020,robotics,0.03706604067375583,0.036544233210283716,0.037618898891096535
118−2020,scientific_computing,0.03474666967076914,0.033850939603458474,0.03563936336497028 118+2020,scientific_computing,0.035949670461354104,0.035486140072591384,0.03641957490516316
119−2020,software_engineering,0.0488565634945222,0.04796413383276536,0.04981011666713605 119+2020,software_engineering,0.05054807767605595,0.049996084509042964,0.051122015804014195
120−2020,speech_audio,0.03298557326724307,0.0321855549609936,0.03387364639366885 120+2020,speech_audio,0.02275173419826225,0.022346389826279685,0.023155166353730917
121−2020,sustainable_computing,0.003347931393809671,0.0031103021939335902,0.003601402540344157 121+2020,sustainable_computing,0.001154614542679762,0.0010726624063696843,0.001243077320930029
122−2021,ai_accelerators,0.002754393651148362,0.002549302690734593,0.0029638951698505995 122+2021,ai_accelerators,0.0022344669660267994,0.0021092366316618066,0.002359697300391792
123−2021,ai_agents,0.007192150382395403,0.00683195478883717,0.007540180186007944 123+2021,ai_agents,0.00875181136733635,0.008529035547524913,0.00900781795087393
124−2021,ai_safety,0.003666644123813192,0.0034356043788022683,0.00392561740465108 124+2021,ai_safety,0.004461777912947028,0.004269415174338515,0.0046515018695099915
125−2021,algorithms,0.04266921106873708,0.041859726604193556,0.04348457627766522 125+2021,algorithms,0.05192228563250264,0.05128526575665957,0.0525297869295311
126−2021,applied_cryptography,0.016060312914408705,0.015518439074017988,0.016593181864960463 126+2021,applied_cryptography,0.01954308818004544,0.0191733456178328,0.019934969676369034
127−2021,autonomous_scientific_discovery,0.000487366690875731,0.000401360804250602,0.0005821936940777963 127+2021,autonomous_scientific_discovery,0.00039537005563804856,0.00034970570871424225,0.0004481904216684258
128−2021,blockchain,0.023479607048660218,0.022897082562710787,0.02407073212327216 128+2021,blockchain,0.019047533856915398,0.018727302001896345,0.019408912821797235
129−2021,cloud_computing,0.017057834180650586,0.016497362486143495,0.017505505846930104 129+2021,cloud_computing,0.013837951947331699,0.013552471510098932,0.014155634470544036
130−2021,compilers,0.0015635429132619606,0.0013981469774444048,0.0017180227173155575 130+2021,compilers,0.0012684043866397123,0.001184321162137503,0.00135785462547185
131−2021,computational_biology,0.028254036395926992,0.02762266497693278,0.028929072341977044 131+2021,computational_biology,0.03438109379752044,0.033903429522156824,0.03482570620963558
132−2021,computational_complexity,0.005535250652027533,0.005230701602208807,0.0058330736004563455 132+2021,computational_complexity,0.0067356029840599674,0.006530717212014956,0.006953861566810384
133−2021,computer_architecture,0.024111051976959244,0.02349713901785688,0.02477035693178042 133+2021,computer_architecture,0.02933967833694116,0.028900432939155948,0.029801420469792655
134−2021,computer_graphics,0.019779148644635435,0.019214266391839877,0.020317788075614608 134+2021,computer_graphics,0.016045583841708858,0.01569493890548688,0.01637757840313434
135−2021,computer_networks,0.07592629074988312,0.07482026974974491,0.07702665153355111 135+2021,computer_networks,0.09239136268493837,0.09160674991502227,0.09308527291267868
136−2021,computer_vision,0.13686844480774374,0.13533566553854387,0.13833065514432816 136+2021,computer_vision,0.16654918868633378,0.16563004275721416,0.1674919942036245
137−2021,cybersecurity,0.07008156592461473,0.0690958061471421,0.07105938669716812 137+2021,cybersecurity,0.05685278279693007,0.056236828452331966,0.05750049197631357
138−2021,database_systems,0.031606060695162794,0.03094403589606372,0.032227949413836804 138+2021,database_systems,0.025640016458843944,0.025257929763672467,0.026030198400629728
139−2021,deep_learning_architectures,0.031203229704816034,0.03043472667770286,0.03187576633450262 139+2021,deep_learning_architectures,0.037969837379465805,0.037445569529670646,0.03844401309551496
140−2021,distributed_computing,0.014842998826791494,0.014365188345540778,0.015302542245797471 140+2021,distributed_computing,0.018061792224985242,0.017730647440828667,0.01837343685707641
141−2021,edge_computing,0.01744375803089155,0.01690335437659699,0.017962108893743767 141+2021,edge_computing,0.014151027783244181,0.013819122672057534,0.014447242249136806
142−2021,federated_learning,0.0014356367228963841,0.0012855674771312551,0.0015768848520845767 142+2021,federated_learning,0.0011646421095944327,0.001075191870762295,0.0012434477700055458
143−2021,formal_verification,0.00994801421963992,0.009553269252822022,0.010365694089557854 143+2021,formal_verification,0.008070200547435461,0.007818711200958907,0.00830652807842997
144−2021,foundation_models,0.00036607633794285676,0.0002955074053273663,0.00044546638713528357 144+2021,foundation_models,0.00044546218938404565,0.00038821403653147756,0.0004964935506377801
145−2021,generative_ai,0.004322347122698791,0.004073040315443128,0.004600432822786708 145+2021,generative_ai,0.003506449362219797,0.003341503121813335,0.003671977029178668
146−2021,hci,0.07592555565683504,0.07497508034567014,0.07705939992884299 146+2021,hci,0.06159364545503337,0.06099276347567848,0.062190502173640805
147−2021,information_retrieval,0.02545553716189395,0.02480641324578865,0.026071251098964104 147+2021,information_retrieval,0.020650482136787305,0.02030162620534197,0.02101722811599907
148−2021,llm,0.006282840281922886,0.005931980370075244,0.006567321291529081 148+2021,llm,0.005096874608655205,0.004914306671198812,0.005278503318603861
149−2021,ml_theory,0.02933903373489016,0.028752429482523895,0.030069311923503275 149+2021,ml_theory,0.03570137932268279,0.035164498989212305,0.036212274361772545
150−2021,multimodal_ai,0.002090604628733905,0.0019162773123822011,0.0023111325431573126 150+2021,multimodal_ai,0.0016959765282573305,0.0015848793316278155,0.0018051952698713706
151−2021,nlp,0.06526082571531903,0.06426811930854207,0.06620942653921133 151+2021,nlp,0.05294201835518901,0.052335992987101274,0.05346203731863964
152−2021,operating_systems,0.002917584307821684,0.0027013566877295324,0.003149138617966262 152+2021,operating_systems,0.002366853319498363,0.002243411989910013,0.0025029071327620444
153−2021,privacy_preserving_computation,0.00935258885069672,0.008951117782488907,0.009758691005107425 153+2021,privacy_preserving_computation,0.0075871692577419185,0.007375977243859241,0.0077814104513659055
154−2021,programming_languages,0.005848400290508772,0.005548261798978514,0.00616221151273328 154+2021,programming_languages,0.007116661001484874,0.006905468987602197,0.007313451526915577
155−2021,quantum_computing,0.013643326972328156,0.013143243171720675,0.01413374429935341 155+2021,quantum_computing,0.016601964327244752,0.01626630230602716,0.016954666618959873
156−2021,reinforcement_learning,0.014537200118791036,0.014053692666417714,0.015000529266994614 156+2021,reinforcement_learning,0.011793119487629032,0.011537291804569118,0.012061559654364278
157−2021,robotics,0.048661689596669146,0.047878705236508834,0.049510832331156475 157+2021,robotics,0.039476179401399004,0.03896443458504034,0.03996824516521459
158−2021,scientific_computing,0.029843307565871686,0.02914496917019756,0.03047155483941157 158+2021,scientific_computing,0.036315007961071254,0.03584092169526092,0.03677657119344509
159−2021,software_engineering,0.054406441767398915,0.0535198092874596,0.05529351530316707 159+2021,software_engineering,0.044136536844553376,0.04361235844499705,0.04464649265613539
160−2021,speech_audio,0.028198904417321143,0.02751284207554992,0.028875925114601003 160+2021,speech_audio,0.02287600407893089,0.02246359375279532,0.02326193713437215
161−2021,sustainable_computing,0.0015811851464158332,0.00143343144375215,0.0017421705239449208 161+2021,sustainable_computing,0.0012827164248528544,0.0011887489489596937,0.001383839919852586
162−2022,ai_accelerators,0.0027899579837432427,0.002599314127148261,0.0030002356051780778 162+2022,ai_accelerators,0.002535539484648477,0.0023954246091861114,0.002663164017821934
163−2022,ai_agents,0.00887642508409796,0.008546479665964214,0.00923778452597547 163+2022,ai_agents,0.008066976784022354,0.007835013284371192,0.008284665606771907
164−2022,ai_safety,0.005799814133692849,0.005513161167030982,0.006086467100354717 164+2022,ai_safety,0.005270924445919495,0.005080580551109589,0.005467379686777902
165−2022,algorithms,0.03691409573423735,0.036232149635466435,0.03764970745690389 165+2022,algorithms,0.050321804747400674,0.04975072845460564,0.050886100568667436
166−2022,applied_cryptography,0.020220814408565557,0.019649275514077406,0.020792156965405307 166+2022,applied_cryptography,0.01837686217621018,0.018004426934174114,0.01874207086306481
167−2022,autonomous_scientific_discovery,0.0004633568502205526,0.00037696828492519535,0.000530111650676056 167+2022,autonomous_scientific_discovery,0.00042110296859749515,0.000370204823770192,0.0004791384518756033
168−2022,blockchain,0.02310894121650807,0.02253357373787615,0.023682443487480203 168+2022,blockchain,0.02100161839149372,0.02059884946104173,0.021372001648725183
169−2022,cloud_computing,0.013945863165748243,0.013480346601395306,0.014405587769473422 169+2022,cloud_computing,0.012674128754016984,0.012406210911919891,0.012960648284451487
170−2022,compilers,0.001260487702718622,0.0011348316077435569,0.001399985601905784 170+2022,compilers,0.001145542821354203,0.001053604980434771,0.0012419861071707055
171−2022,computational_biology,0.03835652299113863,0.037619765965523105,0.03915178863597691 171+2022,computational_biology,0.034858760993731634,0.034417851910933155,0.03532184043409292
172−2022,computational_complexity,0.006952316129792274,0.00660479849212686,0.007294041806829932 172+2022,computational_complexity,0.006318328863575129,0.0061228996151190245,0.006528032788932843
173−2022,computer_architecture,0.029792274767994346,0.029141906307674183,0.03044863152659066 173+2022,computer_architecture,0.027075493413128777,0.026614956649590585,0.027507391606133115
174−2022,computer_graphics,0.016270500922786948,0.015806260553148602,0.01675545491433134 174+2022,computer_graphics,0.014786780935455265,0.014468232598722774,0.015105017013630533
175−2022,computer_networks,0.06432152253301744,0.06338633359075381,0.06524506210814278 175+2022,computer_networks,0.08768398720275215,0.0869405887947355,0.08839789948129392
176−2022,computer_vision,0.19379311247529418,0.19239499208104818,0.1954822052644668 176+2022,computer_vision,0.17612096361206425,0.1750984506622558,0.17715101537561137
177−2022,cybersecurity,0.06195238157567507,0.06096372334716423,0.06289872904095603 177+2022,cybersecurity,0.056302894369175266,0.05570777416747408,0.05694926958262629
178−2022,database_systems,0.02675100459430097,0.026071578162017824,0.02739116349690441 178+2022,database_systems,0.02431155909805454,0.023886619810039738,0.02470964415014819
179−2022,deep_learning_architectures,0.029149596198903123,0.028498311496223772,0.02981043600047121 179+2022,deep_learning_architectures,0.03973713182485685,0.03925085603453045,0.04019972057319965
180−2022,distributed_computing,0.018414508043298996,0.017852884200054972,0.018901523580151573 180+2022,distributed_computing,0.01673527433252503,0.016396250756111792,0.017053242760508392
181−2022,edge_computing,0.016794722444011048,0.016305743530674485,0.017344467859526955 181+2022,edge_computing,0.015263198277046497,0.014961556510769351,0.015591515845783529
182−2022,federated_learning,0.0019123286954017722,0.0017100027487270775,0.002067533606460817 182+2022,federated_learning,0.001737941912771018,0.001631729394949976,0.0018512917690428651
183−2022,formal_verification,0.007843689053521643,0.007511878427728111,0.008219086637259648 183+2022,formal_verification,0.0071284167777414965,0.006915813308638142,0.007356900824897891
184−2022,foundation_models,0.0006871817693948874,0.0005929396981635885,0.0007913388918703123 184+2022,foundation_models,0.0006245171144454377,0.0005602810683881927,0.0006932586053997534
185−2022,generative_ai,0.004240893205408448,0.00400136127436223,0.004502120446602704 185+2022,generative_ai,0.0038541627634347013,0.003694420207232622,0.004006767981185976
186−2022,hci,0.06832746501917564,0.06732732103823348,0.06919724080158117 186+2022,hci,0.06209662885661622,0.06150141943818441,0.06272065527904316
187−2022,information_retrieval,0.022074241809447767,0.021508691213235773,0.02262222018612809 187+2022,information_retrieval,0.02006127405060016,0.019709760131898013,0.020421798859096448
188−2022,llm,0.004274925064464195,0.004020995039202084,0.004525680964410529 188+2022,llm,0.005827636845082284,0.005632118379895545,0.006025697987092123
189−2022,ml_theory,0.039524731999109934,0.03872357622481969,0.04022398853388133 189+2022,ml_theory,0.035920440088288874,0.03539303538513971,0.03636661295819482
190−2022,multimodal_ai,0.0025091951465333316,0.0023324912629746463,0.002725166559771725 190+2022,multimodal_ai,0.002280379635032198,0.0021545394364714428,0.0024159890655974935
191−2022,nlp,0.038846058194478984,0.03814595086323119,0.039605230434953335 191+2022,nlp,0.052955482635747717,0.052373700335276475,0.053590794974599995
192−2022,operating_systems,0.00212633673215618,0.0019299009149334413,0.0022971504862629093 192+2022,operating_systems,0.0019324343855554546,0.0018075309626663669,0.0020493529110527037
193−2022,privacy_preserving_computation,0.008591735493920076,0.008224289584942212,0.008937289755101507 193+2022,privacy_preserving_computation,0.007808248265180673,0.007590291792239215,0.008041148540503503
194−2022,programming_languages,0.007193811437322478,0.00686150996740795,0.00753357373787615 194+2022,programming_languages,0.006537802020937383,0.006334298658358805,0.006740457824574927
195−2022,quantum_computing,0.018094477676409378,0.017562108142776738,0.01861693215879789 195+2022,quantum_computing,0.016444427790654727,0.016103619879628786,0.016776492462078428
196−2022,reinforcement_learning,0.013777012788125499,0.01326064477283734,0.014222895587638581 196+2022,reinforcement_learning,0.012520675977324676,0.012244104112355983,0.012800058169308374
197−2022,robotics,0.04535988690951452,0.044546950876320365,0.04622200552363251 197+2022,robotics,0.04122348255723699,0.04069250918486242,0.04169106744249535
198−2022,scientific_computing,0.026089346719197895,0.025464403984345345,0.02674324925718923 198+2022,scientific_computing,0.03556535750036133,0.035086888173965484,0.03607135018815808
199−2022,software_engineering,0.04658110708255343,0.04577916595766961,0.04736655584497179 199+2022,software_engineering,0.04233333868633717,0.04178612786898702,0.042844595161241394
200−2022,speech_audio,0.024930954593646512,0.024324173091270825,0.025494738151022916 200+2022,speech_audio,0.02265748091208048,0.022302041457230394,0.023059223850130168
201−2022,sustainable_computing,0.0010864016544719173,0.0009756999437165408,0.001199721200539274 201+2022,sustainable_computing,0.001480997728542038,0.0013775063210053654,0.0015765042386868724
202−2023,ai_accelerators,0.0027909590845514565,0.0025953824957934885,0.002963932221061181 202+2023,ai_accelerators,0.0025202296104125358,0.0023894109377019415,0.0026408428505231172
203−2023,ai_agents,0.00871149267729601,0.008433967665136108,0.009047228124396367 203+2023,ai_agents,0.007866457777091286,0.007650370547830774,0.00807773781416183
204−2023,ai_safety,0.007007942917119511,0.006693676386054164,0.007327358497152605 204+2023,ai_safety,0.006328156276301646,0.006121604624991725,0.006525329962519662
205−2023,algorithms,0.033199438666390496,0.03251336425477843,0.03380088831276428 205+2023,algorithms,0.044968524712120116,0.04443327144401749,0.04549408478941346
206−2023,applied_cryptography,0.013054730026926325,0.01263995522363745,0.013464501234608123 206+2023,applied_cryptography,0.01768258661614649,0.01736578477018469,0.01799107438381253
207−2023,autonomous_scientific_discovery,0.0006981761312198765,0.0006090714024729398,0.0007977535019351114 207+2023,autonomous_scientific_discovery,0.0006304514347498526,0.0005657513562586491,0.0006934965782248378
208−2023,blockchain,0.025045323267185022,0.024498825900472664,0.025574540774649688 208+2023,blockchain,0.022615869093064088,0.022195712615017984,0.022956431077473024
209−2023,cloud_computing,0.013340400427283792,0.01292306564484711,0.013749095280096626 209+2023,cloud_computing,0.01204635078948281,0.011778408929714122,0.012319099841441463
210−2023,compilers,0.0012078447070103865,0.0010803402910463564,0.0013371818353188683 210+2023,compilers,0.001090680982117245,0.001000014185157282,0.0011671626217953366
211−2023,computational_biology,0.03591418018995045,0.03518057162007116,0.03662850164420479 211+2023,computational_biology,0.03243042180353242,0.03197551169014573,0.032867836889604804
212−2023,computational_complexity,0.004071530471897246,0.0038609140056459173,0.004288634158034494 212+2023,computational_complexity,0.005514873925474336,0.005303593888403791,0.00568434715177803
213−2023,computer_architecture,0.0285711127298454,0.027938477882943787,0.029237958207176784 213+2023,computer_architecture,0.025799648838550845,0.025440212714314084,0.026166216944643213
214−2023,computer_graphics,0.010818239153251988,0.010441224042393255,0.011221523141047868 214+2023,computer_graphics,0.014653267472173449,0.01434028773803482,0.014952140855459551
215−2023,computer_networks,0.060370126439697365,0.059414439678745885,0.06123074488411439 215+2023,computer_networks,0.08177112721564277,0.08109721403510985,0.08245847689237878
216−2023,computer_vision,0.19943575732328578,0.19807204479498058,0.20088141827499295 216+2023,computer_vision,0.18009002846488228,0.17915365047142007,0.18112672704289906
217−2023,cybersecurity,0.04245027240505386,0.04168277320213828,0.04324686228010361 217+2023,cybersecurity,0.057498746977773435,0.05688472668354175,0.05807135449338499
218−2023,database_systems,0.02528619403245588,0.024732889448464128,0.025859658452236606 218+2023,database_systems,0.022833374838052787,0.0224598323629635,0.02320849344172895
219−2023,deep_learning_architectures,0.04469549048036845,0.04391676227800908,0.045337899793107136 219+2023,deep_learning_architectures,0.04035992472409869,0.03982313473062386,0.04082338533506918
220−2023,distributed_computing,0.011401216222820584,0.010990775929679699,0.011794376656713776 220+2023,distributed_computing,0.01544290789419764,0.015133080417232759,0.01573449168276945
221−2023,edge_computing,0.010612277194542124,0.010263014584899382,0.010982485088121463 221+2023,edge_computing,0.01437429271229664,0.014125185589141104,0.014658823325442183
222−2023,federated_learning,0.002612924171090388,0.0024294783926123655,0.002799686286191705 222+2023,federated_learning,0.0023594644945513234,0.002249135493470099,0.0024753493489012806
223−2023,formal_verification,0.005102503892331932,0.004824222522696542,0.005374298042081402 223+2023,formal_verification,0.006911323853445259,0.006714228973656587,0.007113895780073321
224−2023,foundation_models,0.0009966464273163738,0.000884850974304791,0.0011189145222962546 224+2023,foundation_models,0.0008999694231054147,0.0008313684263616964,0.0009780271913703807
225−2023,generative_ai,0.006363875436069174,0.006060168818988528,0.006671072933805921 225+2023,generative_ai,0.0057465648277449065,0.005570787087093714,0.0059168655215567105
226−2023,hci,0.07084218659455283,0.06984283473472798,0.07180043333465211 226+2023,hci,0.06397033095548067,0.06340178197098031,0.06462765538263673
227−2023,information_retrieval,0.02193669404292852,0.021444479870418507,0.022396530297353213 227+2023,information_retrieval,0.01980878407984037,0.01948063410805307,0.020147651726018415
228−2023,llm,0.018133379568108243,0.017649892597238478,0.01864863355294851 228+2023,llm,0.016374399889040547,0.016070876926423165,0.016689704412844817
229−2023,ml_theory,0.03995836543004159,0.039268218324330736,0.04065680337731067 229+2023,ml_theory,0.036082311739320944,0.0356795714821598,0.0365213029539802
230−2023,multimodal_ai,0.003372190713792004,0.0031697196357382395,0.0035854835218796757 230+2023,multimodal_ai,0.0030450804298417884,0.0029047261791806025,0.003182282423329225
231−2023,nlp,0.05928388074220777,0.05844510939256349,0.06009402687047203 231+2023,nlp,0.05353320745319686,0.052956620212903455,0.05405581228938982
232−2023,operating_systems,0.002047401504802288,0.0018920573156058654,0.0022271818585914063 232+2023,operating_systems,0.0018487988324039429,0.0017376423638146095,0.0019496710619639193
233−2023,privacy_preserving_computation,0.008744656043528955,0.008376106318261262,0.009052202629331308 233+2023,privacy_preserving_computation,0.007896404220241904,0.0076685354297945044,0.008110836514486198
234−2023,programming_languages,0.006765326711520604,0.006489547139688753,0.0070692078726340545 234+2023,programming_languages,0.006109074402726072,0.005919702553013085,0.006301362090324777
235−2023,quantum_computing,0.018403922818955948,0.01791685769541368,0.01886140272893777 235+2023,quantum_computing,0.016618699820006116,0.016293189863601832,0.016949686823249786
236−2023,reinforcement_learning,0.014700098442834502,0.014228829554261085,0.01516970916309627 236+2023,reinforcement_learning,0.013274154958658147,0.012999002310604508,0.013544500414521817
237−2023,robotics,0.044960797410232,0.04425502861358511,0.045807510513368906 237+2023,robotics,0.040599496269303634,0.04014001538301501,0.04108135818152588
238−2023,scientific_computing,0.023764751879839233,0.023231636222308588,0.02435072528863765 238+2023,scientific_computing,0.032189274129740604,0.03178097801931073,0.032598397707678584
239−2023,software_engineering,0.04580908140966415,0.04505496391593028,0.046598107709959015 239+2023,software_engineering,0.041365494762524706,0.040874254885210556,0.04187564818288136
240−2023,speech_audio,0.025637027538393867,0.025080144801729612,0.026157168756152677 240+2023,speech_audio,0.02315017668401459,0.022767177437404054,0.02356816598525374
241−2023,sustainable_computing,0.0018815846736375673,0.001722662331768643,0.0020351834225059403 241+2023,sustainable_computing,0.001699066616650853,0.0015910624052352688,0.001802342442305813
242−2024,ai_accelerators,0.002366081397785513,0.002207815313590616,0.0025120356049396077 242+2024,ai_accelerators,0.002325521653716305,0.002229526283126853,0.002434395276486948
243−2024,ai_agents,0.008114833817172524,0.007832829929645763,0.00840247778244982 243+2024,ai_agents,0.007975728043763073,0.007740404181071606,0.00819283306675464
244−2024,ai_safety,0.008552283750018916,0.008264639784741618,0.008864551469377818 244+2024,ai_safety,0.008405679140205971,0.008204798686891494,0.00860307807127724
245−2024,algorithms,0.02747313970289171,0.02695957164755842,0.027972309510959407 245+2024,algorithms,0.04050328615138335,0.04007654616768904,0.04099407938411533
246−2024,applied_cryptography,0.01709356247028069,0.01670838642878072,0.01751044675376842 246+2024,applied_cryptography,0.016800541900627214,0.016508364441692278,0.017096775501981423
247−2024,autonomous_scientific_discovery,0.0006272866961570896,0.0005460558202621857,0.0007180781916535104 247+2024,autonomous_scientific_discovery,0.000616533647729439,0.0005665078912250767,0.0006746716890723467
248−2024,blockchain,0.021151667193226774,0.02068395342854824,0.021608513491020127 248+2024,blockchain,0.020789081946245295,0.020467598858331115,0.021094002452614117
249−2024,cloud_computing,0.011842787647404351,0.011470886423048936,0.012178509836433161 249+2024,cloud_computing,0.011639776695839344,0.01140715692809406,0.011890682905658184
250−2024,compilers,0.0010564828566856244,0.0009491838165534908,0.001170728821851954 250+2024,compilers,0.0010383724593337922,0.0009626578008407031,0.001112092847804072
251−2024,computational_biology,0.03235753874343653,0.03176024075336766,0.032918932823922734 251+2024,computational_biology,0.03180286066204357,0.03137050968133593,0.03217196962219737
252−2024,computational_complexity,0.005260404223913839,0.0050306054462975185,0.0054929542589694435 252+2024,computational_complexity,0.0051702295370995065,0.0050180903951899565,0.005342716993479075
253−2024,computer_architecture,0.02360441320034978,0.02309811305008527,0.024094480931722503 253+2024,computer_architecture,0.023199782590766327,0.022850886740335228,0.023551382536143608
254−2024,computer_graphics,0.014596796344129116,0.014219805293511023,0.014987818807687564 254+2024,computer_graphics,0.014346575736967276,0.014061935942694821,0.014593425804366844
255−2024,computer_networks,0.07475441588282579,0.07388329899611203,0.07557497841409268 255+2024,computer_networks,0.07347296378270433,0.07286582686491287,0.07408848339482896
256−2024,computer_vision,0.18339469401661118,0.18207175187693078,0.18470711259658634 256+2024,computer_vision,0.18025091297005622,0.1793576490057719,0.18106481174767008
257−2024,cybersecurity,0.06265713692228753,0.06193823335340223,0.06346593782800149 257+2024,cybersecurity,0.06158305830434318,0.06101100701847843,0.062084803121607204
258−2024,database_systems,0.02066744588391253,0.02018308701172632,0.021099668427624275 258+2024,database_systems,0.02031316123571731,0.01999742434956376,0.020655432053530265
259−2024,deep_learning_architectures,0.039872598438844824,0.0392177303868589,0.040469690084605756 259+2024,deep_learning_architectures,0.03918909600753902,0.038687790605703744,0.03962181880130176
260−2024,distributed_computing,0.014583040056932688,0.014178261306177813,0.014947581667638016 260+2024,distributed_computing,0.014333055262236367,0.01406663430766381,0.01457983772726228
261−2024,edge_computing,0.014045169227552378,0.013651464287990625,0.014440043451525825 261+2024,edge_computing,0.013804404700257835,0.01351827765376498,0.014039018738025929
262−2024,federated_learning,0.0028475514496604723,0.002675253952525219,0.003027827593369653 262+2024,federated_learning,0.0027987382692981114,0.0026769863943462784,0.002915149556731236
263−2024,formal_verification,0.0038820242468318128,0.003682214175303705,0.0040801147824603675 263+2024,formal_verification,0.005723216953593674,0.005543327037298934,0.005893642537576779
264−2024,foundation_models,0.0014389076407463062,0.0013095985410998886,0.0015753012282988838 264+2024,foundation_models,0.0014142416568530553,0.0013331188084676029,0.0014994206476577805
265−2024,generative_ai,0.009831618459286665,0.00944362237891143,0.010120569271847622 265+2024,generative_ai,0.009663083290180485,0.009420999190123564,0.00988218258319486
266−2024,hci,0.06613197506810509,0.06536513083834024,0.06690535353428824 266+2024,hci,0.06499833022137073,0.06439812254687886,0.06555648921944947
267−2024,information_retrieval,0.019459643868066204,0.01899021056748812,0.01991373890842027 267+2024,information_retrieval,0.01912606355434352,0.01885024586983298,0.01942834756813981
268−2024,llm,0.034216013143673875,0.03362284203976393,0.03480553883147677 268+2024,llm,0.03362947679818934,0.0332200430221506,0.0341027610161448
269−2024,ml_theory,0.03674304310165759,0.03609333365737032,0.03742274125203307 269+2024,ml_theory,0.03611318800625728,0.03570750616195636,0.036529686230342924
270−2024,multimodal_ai,0.0046881426765424585,0.004473407033406227,0.00489448698448887 270+2024,multimodal_ai,0.004607777788293702,0.004451582503964879,0.0047653251200956166
271−2024,nlp,0.05341703881244724,0.052638226612821505,0.05413663019569236 271+2024,nlp,0.052501355427591774,0.05198176358368295,0.052959767123343236
272−2024,operating_systems,0.0017332921867498526,0.0016012318296641495,0.0018627388492682344 272+2024,operating_systems,0.0017035798160945027,0.001608868890604487,0.0018076874715225002
273−2024,privacy_preserving_computation,0.009197453619531362,0.008862969496350228,0.009505869578475263 273+2024,privacy_preserving_computation,0.009039789405085591,0.008820047889521497,0.00924137968332344
274−2024,programming_languages,0.003527112037163986,0.0033314976332307882,0.0037528527100573594 274+2024,programming_languages,0.005199974581507506,0.0050375260776156375,0.005375808355382974
275−2024,quantum_computing,0.017500748571294942,0.017088059955402118,0.01788344848109955 275+2024,quantum_computing,0.017200747952662115,0.016901742653988066,0.01751141466079157
276−2024,reinforcement_learning,0.013607719294705986,0.013212432382116645,0.013968134019252385 276+2024,reinforcement_learning,0.013374453603814937,0.013108674871792098,0.013619174196444386
277−2024,robotics,0.041249602787207205,0.04064693984513172,0.04187991586654751 277+2024,robotics,0.04054249552810298,0.04004081831321261,0.04097863224173527
278−2024,scientific_computing,0.030108385786820655,0.02956081677496686,0.030610490269490254 278+2024,scientific_computing,0.029592263043539985,0.029195775122056088,0.029967557620883184
279−2024,software_engineering,0.04200207169685179,0.04136371118950157,0.04267475414075709 279+2024,software_engineering,0.04128206549588369,0.04084998492467068,0.04173250011154392
280−2024,speech_audio,0.022597452977571292,0.02208393077652866,0.023041781054015897 280+2024,speech_audio,0.022210083840463805,0.02187920402261164,0.022523116631671173
281−2024,sustainable_computing,0.0017484241026659228,0.0016313580986243256,0.0018833044986268934 281+2024,sustainable_computing,0.0017184523382985024,0.0016190430478394957,0.0018076874715225002
282−2025,ai_accelerators,0.002218169672523209,0.0020761881929598136,0.0023531684563703713 282+2025,ai_accelerators,0.0021347609874109584,0.002043451234263698,0.002224950719053806
283−2025,ai_agents,0.010263398920242486,0.00995831330657409,0.010557835922943626 283+2025,ai_agents,0.00987746964741723,0.009682249899198065,0.01009646185206756
284−2025,ai_safety,0.01266777381186523,0.012339296167547456,0.013030059341275761 284+2025,ai_safety,0.01219143407553425,0.011944973343488196,0.012440890865104609
285−2025,algorithms,0.02622002241445325,0.025748515886214646,0.026684429868713683 285+2025,algorithms,0.03785112674163344,0.03743932283499843,0.03824159423860938
286−2025,applied_cryptography,0.01700519163426502,0.016601300876211932,0.017405765611852895 286+2025,applied_cryptography,0.01636575422248107,0.016069620536714305,0.016610506921732895
287−2025,autonomous_scientific_discovery,0.0008891299212002788,0.0007948053399001707,0.0009845600959887905 287+2025,autonomous_scientific_discovery,0.000855696429371444,0.000796335289637561,0.0009111654943774921
288−2025,blockchain,0.020962052540130133,0.020516032851255778,0.021342667645795777 288+2025,blockchain,0.02017382733748488,0.01991328233502083,0.020477353165180772
289−2025,cloud_computing,0.00851423364332623,0.00823946456689248,0.008794763443725873 289+2025,cloud_computing,0.012291115989426997,0.012060335558442723,0.012523772456431163
290−2025,compilers,0.0008926212690583952,0.0007924777746614266,0.0009764718067841543 290+2025,compilers,0.0008590564938846826,0.0008002273643653959,0.0009167656018995565
291−2025,computational_biology,0.032765135865801896,0.032314461046450055,0.0333916000498099 291+2025,computational_biology,0.03153308543524036,0.0311824627032839,0.03189544039245553
292−2025,computational_complexity,0.004898361044937139,0.0047062205344788055,0.005120759903499145 292+2025,computational_complexity,0.004714170512073831,0.004580299941758882,0.004862013350656332
293−2025,computer_architecture,0.015679255303260413,0.015315262892674803,0.0160718379735286 293+2025,computer_architecture,0.02263451458267999,0.022314076430267463,0.022954840732942072
294−2025,computer_graphics,0.014477455784988833,0.014120116331710631,0.01484998260144984 294+2025,computer_graphics,0.013933067514896286,0.0136916748801577,0.014181348281887012
295−2025,computer_networks,0.06930907389670499,0.06849535708924001,0.07009253235606629 295+2025,computer_networks,0.06670288069530934,0.06617697459791229,0.06721394650777295
296−2025,computer_vision,0.17544837634867858,0.17428651397062844,0.17643322739032222 296+2025,computer_vision,0.16885108194077325,0.16800818175708973,0.16958203597509072
297−2025,cybersecurity,0.06871438097820584,0.06803339357848026,0.06940590061063674 297+2025,cybersecurity,0.06613054970655437,0.06558070314950047,0.06666502396846019
298−2025,database_systems,0.01941305787374588,0.0189342777041362,0.01982509511013457 298+2025,database_systems,0.01868307871511133,0.018413153532547823,0.018980752430446664
299−2025,deep_learning_architectures,0.039732702407982615,0.03915296409614241,0.04029259822616253 299+2025,deep_learning_architectures,0.0382386541821603,0.03786392298732136,0.038624221585054436
300−2025,distributed_computing,0.009736981248746508,0.009414768634196356,0.010062064527091113 300+2025,distributed_computing,0.014056269880381703,0.01381263720263429,0.014313146812418798
301−2025,edge_computing,0.01344634438422516,0.013122812816039718,0.013771097924160942 301+2025,edge_computing,0.01294072846198647,0.012697039783163836,0.013170388871466331
302−2025,federated_learning,0.002960662983682604,0.002803494140936403,0.003127142087383782 302+2025,federated_learning,0.002849334707226379,0.0027412246315129253,0.0029546727297164103
303−2025,formal_verification,0.005380167049357185,0.005183487786683301,0.005612981762362572 303+2025,formal_verification,0.0051778594149007665,0.005019880381703329,0.00532858630885713
304−2025,foundation_models,0.00215765297631586,0.0020214322207183564,0.002289218601435875 304+2025,foundation_models,0.0020765198691814883,0.0019823820617355854,0.0021633495363110973
305−2025,generative_ai,0.014830081918658578,0.014443647899896075,0.015152798839010459 305+2025,generative_ai,0.01427243403073339,0.014036641503516868,0.014530094977823574
306−2025,hci,0.06767163175124845,0.06691534761604949,0.06846114188023047 306+2025,hci,0.06512701043860042,0.06458724407508624,0.06561657183817929
307−2025,information_retrieval,0.0202253781420676,0.019800248351210973,0.02066080740910567 307+2025,information_retrieval,0.019464853725191522,0.019179780251780836,0.01973494691098069
308−2025,llm,0.052554095525604964,0.05185338201048102,0.053167467155145025 308+2025,llm,0.05057793109627705,0.05014526678912235,0.05100300725773935
309−2025,ml_theory,0.038169742350165896,0.03751558013981684,0.03875751076207977 309+2025,ml_theory,0.03673446530173379,0.03632224138703463,0.037132716948165403
310−2025,multimodal_ai,0.007854368898142254,0.007602933663226913,0.008127857813694695 310+2025,multimodal_ai,0.007559025133282559,0.007395978002777653,0.007723780296581694
311−2025,nlp,0.049942567327733986,0.049287357713027495,0.05058974684236681 311+2025,nlp,0.04806460284037453,0.04766080708749608,0.04852733972492272
312−2025,operating_systems,0.0015862357102041625,0.0014755599831018764,0.0017073272917498285 312+2025,operating_systems,0.001526589310514762,0.001447067783701447,0.0016083508803369025
313−2025,privacy_preserving_computation,0.010261071355003742,0.009930266145447224,0.01056260743168305 313+2025,privacy_preserving_computation,0.009875229604408405,0.009669509654585368,0.010055553066618879
314−2025,programming_languages,0.0048797405230271845,0.004658563636215519,0.005091607148883875 314+2025,programming_languages,0.004696250168003226,0.004548351328345504,0.0048334808028314144
315−2025,quantum_computing,0.020332446143049833,0.01989468931077302,0.02074215581419978 315+2025,quantum_computing,0.01956789570359751,0.01930452264683482,0.0198592132968953
316−2025,reinforcement_learning,0.015373568401905345,0.015026528424808587,0.015750633970581903 316+2025,reinforcement_learning,0.014795484073294207,0.014556863491779043,0.015040236772546033
317−2025,robotics,0.038944821574667714,0.038327784029876626,0.03951874097341106 317+2025,robotics,0.03748039962367278,0.037061287576721474,0.037860310917969625
318−2025,scientific_computing,0.017992855150572133,0.01758428986566457,0.01838433222735192 318+2025,scientific_computing,0.02597441870883921,0.025647204426324986,0.026275872496751937
319−2025,software_engineering,0.042339575475376104,0.041742205856852406,0.042954110887535535 319+2025,software_engineering,0.04074750235204516,0.04033581044756059,0.04116639039469558
320−2025,speech_audio,0.02138217806572346,0.02093255065072905,0.021811730230533698 320+2025,speech_audio,0.020578155100577932,0.020276869315890865,0.020856508444962143
321−2025,sustainable_computing,0.00190743971315086,0.001785126159854853,0.0020262037294577822 321+2025,sustainable_computing,0.0018357152457327182,0.0017461135253796873,0.0019159367859862908
modified results/bibliometrics/diversity_yearly.csv +8 −8
@@ -1,9 +1,9 @@
1 year,n,rao_stirling,rs_ci_low,rs_ci_high,leinster_cobbold_q2,similarity_source 1 year,n,rao_stirling,rs_ci_low,rs_ci_high,leinster_cobbold_q2,similarity_source
2−2018,169727,0.8992868732205007,0.8988175897191749,0.8997310848387939,9.929192270928045,co-classification Jaccard 2+2018,499561,0.8966979269920947,0.8964150650642733,0.8969729208682462,9.680347846683333,co-classification Jaccard
3−2019,180248,0.896100206436324,0.8956356536840846,0.8965579428776569,9.624658199028467,co-classification Jaccard 3+2019,539741,0.8957695076396416,0.8955171860453549,0.896020741688698,9.594121426028362,co-classification Jaccard
4−2020,189370,0.900723632618864,0.9002795125878724,0.9011485807064236,10.0728907229337,co-classification Jaccard 4+2020,549101,0.896412402094723,0.8961367144172547,0.8966751994754848,9.653665305709902,co-classification Jaccard
5−2021,226728,0.9022200511081365,0.9018514633225768,0.9026284736924316,10.22704564006183,co-classification Jaccard 5+2021,558970,0.8939562806724264,0.8936855340180797,0.8942329598613235,9.43007286373045,co-classification Jaccard
6−2022,254663,0.890309885996106,0.8897676301389984,0.8907889622855558,9.116591855894152,co-classification Jaccard 6+2022,560433,0.8914726303600234,0.891161796756561,0.8917985624313133,9.214265519539925,co-classification Jaccard
7−2023,286460,0.8855095802901976,0.8849779485027648,0.8860485908545297,8.73435526338963,co-classification Jaccard 7+2023,634466,0.8905938872892762,0.8902601885615422,0.8908748156052915,9.140257113823784,co-classification Jaccard
8−2024,363470,0.8883809706057654,0.8879340384933714,0.8887799732899944,8.959045831405975,co-classification Jaccard 8+2024,739619,0.8900517958692582,0.889784104970656,0.890332496245109,9.095191757846983,co-classification Jaccard
9−2025,429633,0.889452772558058,0.8891043925701015,0.8898464283041351,9.04590755589223,co-classification Jaccard 9+2025,892840,0.893113751756877,0.892858747038931,0.893398088972461,9.355740485206343,co-classification Jaccard
modified tables/make_numbers.py +109 −0
@@ -71,5 +71,114 @@ if os.path.exists(p):
71 leads = [int(float(r["lead_months"])) for r in csv.DictReader(open(p)) if r.get("lead_months") not in ("", None)] 71 leads = [int(float(r["lead_months"])) for r in csv.DictReader(open(p)) if r.get("lead_months") not in ("", None)]
72 if leads: 72 if leads:
73 import statistics; emit("known set median lead months", int(statistics.median(leads))); emit("known set n detected", len(leads)); emit("known set n positive lead", sum(1 for l in leads if l > 0)) 73 import statistics; emit("known set median lead months", int(statistics.median(leads))); emit("known set n detected", len(leads)); emit("known set n positive lead", sum(1 for l in leads if l > 0))
74+# ---- per-category growth and yearly primary counts (appended by Agent F, 2026-11-03; used by Parts V–VII) ----
75+p = os.path.join(R, "results/bibliometrics/cagr_by_category.csv")
76+if os.path.exists(p):
77+ for row in csv.DictReader(open(p)):
78+ for k, v in row.items():
79+ if k == "category" or v in ("", None): continue
80+ try: v = int(v) if re.fullmatch(r"-?\d+", v) else float(v)
81+ except ValueError: pass
82+ emit(f"cagr {row['category']} {k}", v)
83+p = os.path.join(R, "results/graph/category_counts_yearly.csv")
84+if os.path.exists(p):
85+ agg = {}
86+ for row in csv.DictReader(open(p)): # one row per (year, period, category): sum the periods
87+ k = (row["category_id"], row["year"]); agg[k] = agg.get(k, 0) + int(float(row["n_primary"]))
88+ for (c, y), v in sorted(agg.items()): emit(f"catcount {c} {y} primary", v)
89+ post = {}; base = {}
90+ for (c, y), v in agg.items():
91+ (post if int(y) >= 2023 else base).setdefault(c, 0); (post if int(y) >= 2023 else base)[c] += v
92+ for c in sorted(set(base) | set(post)): emit(f"catcount {c} baseline primary", base.get(c, 0)); emit(f"catcount {c} post primary", post.get(c, 0))
93+# de-duplicate macro names (last definition dropped) so that independent appends cannot break the compile
94+_seen = set(); _out = []
95+for line in out:
96+ n = line.split("}")[0]
97+ if n in _seen: continue
98+ _seen.add(n); _out.append(line)
99+out = _out
74 open(os.path.join(R, "manuscript/tables/numbers.tex"), "w").write("% machine-generated by tables/make_numbers.py — do not edit\n" + "\n".join(out) + "\n") 100 open(os.path.join(R, "manuscript/tables/numbers.tex"), "w").write("% machine-generated by tables/make_numbers.py — do not edit\n" + "\n".join(out) + "\n")
75 print(len(out), "macros") 101 print(len(out), "macros")
102+
103+# ---- Part VII-b (security half of Part VII; Agent H, 2026-11-03): category-level macros for the five security_privacy ----
104+# categories, appended to numbers.tex. Sources: results/graph/category_counts_yearly.csv (eligible works, multi-label and
105+# primary counts), results/bibliometrics/cagr_by_category.csv, results/econometrics/breaks.csv,
106+# results/graph/category_flows_yearly.csv, results/graph/category_cooccurrence_yearly.csv,
107+# literature_synthesis/landmarks/security_privacy_landmarks.csv. Percentages are pre-formatted strings (1 or 2 decimals).
108+SEC = ["cybersecurity", "applied_cryptography", "blockchain", "privacy_preserving_computation", "federated_learning"]
109+extra = []
110+def emit2(k, v): extra.append(f"\\newcommand{{\\{name(k)}}}{{{fmt(v)}}}")
111+agg = {}; tot_primary = {}
112+p = os.path.join(R, "results/graph/category_counts_yearly.csv")
113+if os.path.exists(p):
114+ for row in csv.DictReader(open(p)):
115+ tot_primary[row["year"]] = tot_primary.get(row["year"], 0) + int(row["n_primary"])
116+ if row["category_id"] in SEC:
117+ a = agg.setdefault((row["category_id"], row["year"]), [0, 0]); a[0] += int(row["n_multilabel"]); a[1] += int(row["n_primary"])
118+ for (c, y), (m, pr) in sorted(agg.items()):
119+ emit2(f"sec count {c} {y} multilabel", m); emit2(f"sec count {c} {y} primary", pr)
120+ if tot_primary.get(y): emit2(f"sec share primary {c} {y} pct", f"{100 * pr / tot_primary[y]:.2f}")
121+ for c in SEC:
122+ for y0, y1, tag in [("2018", "2022", "baseline"), ("2023", "2025", "post"), ("2022", "2025", "span")]:
123+ m0, m1 = agg.get((c, y0), [0, 0])[0], agg.get((c, y1), [0, 0])[0]
124+ if m0 > 0:
125+ emit2(f"sec cagr multilabel {c} {tag} pct", f"{100 * ((m1 / m0) ** (1 / (int(y1) - int(y0))) - 1):.1f}"); emit2(f"sec ratio multilabel {c} {tag}", f"{m1 / m0:.2f}")
126+ p0, p1 = agg.get((c, y0), [0, 0])[1], agg.get((c, y1), [0, 0])[1]
127+ if p0 > 0: emit2(f"sec cagr primary {c} {tag} pct", f"{100 * ((p1 / p0) ** (1 / (int(y1) - int(y0))) - 1):.1f}")
128+ s = sum(v[0] for (cc, y), v in agg.items() if cc == c); emit2(f"sec count {c} all multilabel", s)
129+ s = sum(v[1] for (cc, y), v in agg.items() if cc == c); emit2(f"sec count {c} all primary", s)
130+p = os.path.join(R, "results/bibliometrics/cagr_by_category.csv")
131+if os.path.exists(p):
132+ for row in csv.DictReader(open(p)):
133+ if row["category"] in SEC:
134+ c = row["category"]
135+ for k in ("n_2018", "n_2022", "n_2023", "n_2025"): emit2(f"sec cagr {c} {k}", int(row[k]))
136+ for k in ("cagr_baseline", "cagr_post"): emit2(f"sec cagr {c} {k} pct", f"{100 * float(row[k]):.1f}")
137+ for k in ("share_2018", "share_2022", "share_2025"): emit2(f"sec cagr {c} {k} pct", f"{100 * float(row[k]):.2f}")
138+p = os.path.join(R, "results/econometrics/breaks.csv")
139+if os.path.exists(p):
140+ for row in csv.DictReader(open(p)):
141+ c = row["series"].replace("cat_", "")
142+ if c in SEC:
143+ for k in ("bp_breaks_bic", "pelt", "bocpd_resets"): emit2(f"sec break {c} {k}", (row[k] or "none").replace(";", ", "))
144+ for k in ("chatgpt_level", "gpt3_level", "dalle2_level", "gpt4_level", "o1_level"): emit2(f"sec break {c} {k}", float(row[k])); emit2(f"sec break {c} {k} p", float(row[k + "_p"]))
145+p = os.path.join(R, "results/graph/category_flows_yearly.csv")
146+if os.path.exists(p):
147+ want = SEC + ["computer_networks", "computer_vision", "ml_theory", "quantum_computing", "llm", "ai_safety", "ai_agents", "generative_ai", "distributed_computing", "edge_computing", "software_engineering", "database_systems", "hci"]
148+ YRS = ("2019", "2022", "2025", "2026"); rows = list(csv.DictReader(open(p))); tot = {}
149+ for row in rows:
150+ if row["citing_cat"] in SEC: tot[(row["citing_cat"], row["citing_year"])] = tot.get((row["citing_cat"], row["citing_year"]), 0) + int(row["n"])
151+ for row in rows:
152+ if row["citing_year"] not in YRS: continue
153+ if (row["citing_cat"] in SEC and row["cited_cat"] in want) or (row["cited_cat"] in SEC and row["citing_cat"] in ("llm", "ai_agents", "ai_safety", "generative_ai", "computer_networks", "distributed_computing", "edge_computing")):
154+ emit2(f"sec flow {row['citing_cat']} to {row['cited_cat']} {row['citing_year']} pct", f"{100 * float(row['row_share']):.1f}")
155+ for (c, y), n in sorted(tot.items()):
156+ if y in YRS: emit2(f"sec flow {c} total {y}", n)
157+p = os.path.join(R, "results/graph/category_cooccurrence_yearly.csv")
158+if os.path.exists(p):
159+ partners = ["llm", "ai_agents", "ai_safety", "generative_ai", "foundation_models"]
160+ for row in csv.DictReader(open(p)):
161+ a, b, y = row["cat_a"], row["cat_b"], row["year"]
162+ if a in SEC and b in partners: c, q = a, b
163+ elif b in SEC and a in partners: c, q = b, a
164+ elif a in SEC and b in SEC: c, q = a, b
165+ else: continue
166+ n = int(row["n"]); emit2(f"sec cooc {c} {q} {y}", n)
167+ m = agg.get((c, y), [0, 0])[0]
168+ if m: emit2(f"sec cooc {c} {q} {y} pct", f"{100 * n / m:.1f}")
169+p = os.path.join(R, "literature_synthesis/landmarks/security_privacy_landmarks.csv")
170+if os.path.exists(p):
171+ from collections import Counter
172+ L = list(csv.DictReader(open(p))); n = len(L); emit2("sec land n", n)
173+ wa = sum(r["has_abstract"] == "True" for r in L); emit2("sec land with abstract", wa); emit2("sec land with abstract pct", f"{100 * wa / n:.0f}")
174+ emit2("sec land oa pct", f"{100 * sum(r['is_oa'] == 'True' for r in L) / n:.0f}")
175+ emit2("sec land post", sum(r["period"] == "post" for r in L)); emit2("sec land baseline", sum(r["period"] == "baseline" for r in L)); emit2("sec land year 2026", sum(r["year"] == "2026" for r in L))
176+ emit2("sec land no doi", sum(not r["doi"] for r in L))
177+ for t, k in Counter(r["type"] for r in L).items(): emit2(f"sec land type {t}", k)
178+ for c in SEC:
179+ g = [r for r in L if r["primary_category"] == c]; emit2(f"sec land {c} n", len(g)); emit2(f"sec land {c} post", sum(r["period"] == "post" for r in g))
180+ emit2(f"sec land {c} with abstract", sum(r["has_abstract"] == "True" for r in g)); emit2(f"sec land {c} year 2026", sum(r["year"] == "2026" for r in g))
181+ emit2(f"sec land {c} preprints", sum(r["type"] == "preprint" for r in g)); emit2(f"sec land {c} articles", sum(r["type"] in ("article", "review") for r in g))
182+ emit2(f"sec land {c} conference", sum(r["type"] == "conference-paper" for r in g)); emit2(f"sec land {c} grey", sum(r["type"] in ("dissertation", "report", "book-chapter", "book") for r in g))
183+open(os.path.join(R, "manuscript/tables/numbers.tex"), "a").write("% ---- Part VII-b security macros (Agent H) ----\n" + "\n".join(extra) + "\n")
184+print(len(extra), "Part VII-b macros appended")
modified tables/make_tables.py +287 −0
@@ -92,6 +92,9 @@ def main():
92 if os.path.exists(p): 92 if os.path.exists(p):
93 d = pd.read_csv(p)[["concept", "total_works", "detection_month", "peak_growth_month", "lead_months"]]; write("tab_known_set", d, "Preregistered known-emergence set: detection month, peak-growth month and lead time.", "tab:known_set", [p]) 93 d = pd.read_csv(p)[["concept", "total_works", "detection_month", "peak_growth_month", "lead_months"]]; write("tab_known_set", d, "Preregistered known-emergence set: detection month, peak-growth month and lead time.", "tab:known_set", [p])
94 tables_part08() # Part VIII tables (appended 2026-11-03) 94 tables_part08() # Part VIII tables (appended 2026-11-03)
95+ tables_part07b() # Part VII-b security tables (Agent H, appended 2026-11-03)
96+ tables_part06() # Part VI systems table (appended by Agent F, 2026-11-03)
97+ tables_part09() # Part IX results tables (appended by the Part IX writer, 2026-11-03)
95 json.dump(INV, open(os.path.join(OUT, "TABLES.json"), "w"), indent=1); print(len(INV), "tables") 98 json.dump(INV, open(os.path.join(OUT, "TABLES.json"), "w"), indent=1); print(len(INV), "tables")
96 99
97 100
@@ -223,5 +226,289 @@ def tables_part08():
223 note="Shares of all resolved references of the citing year. Inside: cited work is an eligible work (computer science, 2018--2026); CS out: cited work outside the corpus whose primary OpenAlex field is Computer Science (field 17; mostly pre-2018 work); Other: all remaining external references, including those without a field. The seven field columns are the largest components of `Other': Eng. = Engineering (OpenAlex field 22), Soc. = Social Sciences (33), Med. = Medicine (27), Math. = Mathematics (26), Dec. = Decision Sciences (18), Psy. = Psychology (32), Phys. = Physics and Astronomy (31). The rise of the inside-corpus share is mechanical: the corpus starts in 2018, so later citing years have more corpus to cite. The 2026 row is incomplete.") 226 note="Shares of all resolved references of the citing year. Inside: cited work is an eligible work (computer science, 2018--2026); CS out: cited work outside the corpus whose primary OpenAlex field is Computer Science (field 17; mostly pre-2018 work); Other: all remaining external references, including those without a field. The seven field columns are the largest components of `Other': Eng. = Engineering (OpenAlex field 22), Soc. = Social Sciences (33), Med. = Medicine (27), Math. = Mathematics (26), Dec. = Decision Sciences (18), Psy. = Psychology (32), Phys. = Physics and Astronomy (31). The rise of the inside-corpus share is mechanical: the corpus starts in 2018, so later citing years have more corpus to cite. The 2026 row is incomplete.")
224 227
225 228
229+def tables_part06():
230+ """Part VI — the eleven systems categories: primary-category counts and CAGR (cagr_by_category.csv) joined with the
231+ structural-change estimates of breaks.csv (ITS slope change at 2022-12, HAC p, placebo p, Bai–Perron BIC break dates)."""
232+ B = f"{R}/results/bibliometrics"; E = f"{R}/results/econometrics"
233+ cats = ["computer_architecture", "ai_accelerators", "operating_systems", "distributed_computing", "cloud_computing", "edge_computing", "computer_networks", "database_systems", "information_retrieval", "quantum_computing", "sustainable_computing"]
234+ names = {"computer_architecture": "Computer architecture", "ai_accelerators": "Accelerators and AI hardware", "operating_systems": "Operating systems", "distributed_computing": "Distributed computing", "cloud_computing": "Cloud computing", "edge_computing": "Edge computing", "computer_networks": "Computer networks", "database_systems": "Database systems", "information_retrieval": "Information retrieval", "quantum_computing": "Quantum computing", "sustainable_computing": "Sustainable computing"}
235+ p, q = f"{B}/cagr_by_category.csv", f"{E}/breaks.csv"
236+ if os.path.exists(p) and os.path.exists(q):
237+ c = pd.read_csv(p).set_index("category"); b = pd.read_csv(q).set_index("series"); rows = []
238+ for k in cats:
239+ if k not in c.index: continue
240+ r = c.loc[k]; s = b.loc["cat_" + k] if ("cat_" + k) in b.index else None
241+ rows.append({"Category": names[k], "n 2018": int(r.n_2018), "n 2022": int(r.n_2022), "n 2025": int(r.n_2025), "CAGR base": float(r.cagr_baseline), "CAGR post": float(r.cagr_post),
242+ "Slope ch.": float(s.chatgpt_slope) if s is not None else float("nan"), "p": float(s.chatgpt_slope_p) if s is not None else float("nan"),
243+ "Placebo p": float(s.placebo_slope_p) if s is not None else float("nan"), "BP breaks": str(s.bp_breaks_bic) if s is not None else ""})
244+ write("tab_systems_categories", pd.DataFrame(rows), "The systems and infrastructure categories of Part VI: primary-category output, compound annual growth, and structural-change estimates of the monthly log-volume series.", "tab:systems_categories", [p, q],
245+ fmt={"CAGR base": "{:.3f}", "CAGR post": "{:.3f}", "Slope ch.": "{:.3f}", "p": "{:.3f}", "Placebo p": "{:.3f}"},
246+ note="Counts are eligible works whose primary category is the row category (multi-label memberships are larger). CAGR base: 2018 to 2022; CAGR post: 2023 to 2025 (complete years). Slope ch.: change in the monthly trend of the log-volume series at 2022-12 in the interrupted time-series model of Part IV, with its HAC p-value; Placebo p: share of placebo event dates whose slope statistic is at least as extreme; BP breaks: Bai--Perron break dates selected by BIC. 2026 is incomplete and excluded from the growth columns.")
247+
248+
249+
250+# ---------------- Part VII-b (security half of Part VII): tables appended by Agent H, 2026-11-03 ----------------
251+def tables_part07b():
252+ G = f"{R}/results/graph"; B = f"{R}/results/bibliometrics"; E = f"{R}/results/econometrics"
253+ SEC = ["cybersecurity", "applied_cryptography", "blockchain", "privacy_preserving_computation", "federated_learning"]
254+ NAMES = {"cybersecurity": "Cybersecurity", "applied_cryptography": "Applied cryptography", "blockchain": "Blockchain systems", "privacy_preserving_computation": "Privacy-preserving computation", "federated_learning": "Federated learning"}
255+ SHORT = {"cybersecurity": "Cyber.", "applied_cryptography": "Crypto.", "blockchain": "Blockch.", "privacy_preserving_computation": "Privacy", "federated_learning": "Fed. learn."}
256+ p = f"{G}/category_counts_yearly.csv"; agg = None
257+ if os.path.exists(p):
258+ d = pd.read_csv(p); d = d[d.category_id.isin(SEC)].groupby(["year", "category_id"])[["n_multilabel", "n_primary"]].sum().reset_index()
259+ agg = d
260+ t = pd.DataFrame({"Year": sorted(d.year.unique().astype(str))})
261+ for v, suffix in (("n_multilabel", " (all)"), ("n_primary", " (prim.)")):
262+ piv = d.pivot_table(index="year", columns="category_id", values=v, aggfunc="sum", fill_value=0).astype(int)
263+ for c in SEC: t[SHORT[c] + suffix] = piv[c].values
264+ write("tab_security_counts", t, "Eligible works in the five security-and-privacy categories by publication year: works carrying the category (all) and works whose primary category it is (prim.).", "tab:security_counts", [p], longtable=True,
265+ note="Counts of eligible works (Part III) per publication year, summed over the pre/baseline/post period labels of the result file. `all' counts every work to which the multi-label classifier assigns the category (a work can carry several categories); `prim.' counts works for which the category is the argmax of the assignment score (Part IV). Cyber. = cybersecurity; Crypto. = applied cryptography; Blockch. = blockchain systems; Privacy = privacy-preserving computation; Fed. learn. = federated learning. The 2026 row is incomplete (the corpus ends on 2026-11-02 and recent months are under-indexed).")
266+ p = f"{E}/breaks.csv"; q = f"{B}/cagr_by_category.csv"
267+ if os.path.exists(p) and agg is not None:
268+ b = pd.read_csv(p).set_index("series"); rows = []
269+ for c in SEC:
270+ g = agg[agg.category_id == c].set_index("year")
271+ m18, m22, m23, m25 = (int(g.loc[y, "n_multilabel"]) if y in g.index else 0 for y in (2018, 2022, 2023, 2025))
272+ r = b.loc[f"cat_{c}"]
273+ rows.append({"Category": NAMES[c], "Works 2018": m18, "Works 2025": m25, "CAGR 18--22 (%)": 100 * ((m22 / m18) ** 0.25 - 1) if m18 else float("nan"), "CAGR 23--25 (%)": 100 * ((m25 / m23) ** 0.5 - 1) if m23 else float("nan"),
274+ "Level 2022-12": float(r.chatgpt_level), "p": float(r.chatgpt_level_p), "Slope change": float(r.chatgpt_slope), "p ": float(r.chatgpt_slope_p), "Placebo p (slope)": float(r.placebo_slope_p), "Bai--Perron": str(r.bp_breaks_bic).replace(";", ", "), "PELT": str(r.pelt).replace(";", ", ")})
275+ write("tab_security_summary", pd.DataFrame(rows), "Growth and structural-change summary for the five security-and-privacy categories (multi-label counts).", "tab:security_summary", [f"{G}/category_counts_yearly.csv", p], fmt={"CAGR 18--22 (%)": "{:.1f}", "CAGR 23--25 (%)": "{:.1f}", "Level 2022-12": "{:.3f}", "p": "{:.3f}", "Slope change": "{:.3f}", "p ": "{:.3f}", "Placebo p (slope)": "{:.3f}"},
276+ note="Works: eligible works carrying the category (multi-label). CAGR: compound annual growth of the multi-label count between the years shown (2026 excluded as incomplete). Level and slope change: interrupted-time-series estimates on the log monthly series at the 2022-12 reference month with HAC p-values (Part IV, Section~\\ref{sec:m-its}); Placebo p: share of placebo event months whose slope t-statistic exceeds the one at 2022-12 (Section~\\ref{sec:m-placebo}); Bai--Perron: BIC-selected break months (Section~\\ref{sec:m-bp}); PELT: penalised change-point months. The ITS, placebo and break columns are those of Table~\\ref{tab:breaks}.")
277+ p = f"{G}/category_cooccurrence_yearly.csv"
278+ if os.path.exists(p) and agg is not None:
279+ d = pd.read_csv(p); pairs = [("cybersecurity", "llm"), ("cybersecurity", "ai_safety"), ("cybersecurity", "ai_agents"), ("applied_cryptography", "llm"), ("blockchain", "llm"), ("privacy_preserving_computation", "llm"), ("federated_learning", "llm")]
280+ LAB = {"llm": "LLM", "ai_safety": "AI safety", "ai_agents": "AI agents"}
281+ rows = []
282+ for y in sorted(d.year.unique()):
283+ if y < 2018 or y > 2026: continue
284+ row = {"Year": str(y)}
285+ for a, bb in pairs:
286+ n = d[(d.year == y) & (((d.cat_a == a) & (d.cat_b == bb)) | ((d.cat_a == bb) & (d.cat_b == a)))].n.sum()
287+ m = agg[(agg.year == y) & (agg.category_id == a)].n_multilabel.sum()
288+ row[f"{SHORT[a]} x {LAB[bb]}"] = int(n); row[f"{SHORT[a]} x {LAB[bb]} (%)"] = 100 * n / m if m else float("nan")
289+ rows.append(row)
290+ t = pd.DataFrame(rows); cols = ["Year"] + [c for c in t.columns if c != "Year"]
291+ write("tab_security_llm_cooccurrence", t[cols], "Co-classification of security-and-privacy works with the LLM, AI-safety and AI-agents categories, by publication year: number of works carrying both categories and share of the security category's works.", "tab:security_llm_cooccurrence", [p, f"{G}/category_counts_yearly.csv"], longtable=True, fmt={c: "{:.1f}" for c in cols if c.endswith("(%)")},
292+ note="A work is co-classified when the multi-label classifier assigns both categories (Part IV). Shares are relative to all eligible works carrying the security category in the year (multi-label count). Non-zero counts before 2022 reflect the breadth of the keyword patterns of the AI categories (for example `language model' matches pre-transformer work), not the use of LLMs; the table should be read for its trend, not its level. The 2026 row is incomplete.")
293+ p = f"{R}/literature_synthesis/landmarks/security_privacy_landmarks.csv"
294+ if os.path.exists(p):
295+ d = pd.read_csv(p); rows = []
296+ for c in SEC:
297+ g = d[d.primary_category == c]
298+ rows.append({"Category": NAMES[c], "Works": int(len(g)), "Post-period": int((g.period == "post").sum()), "Dated 2026": int((g.year == 2026).sum()), "Articles": int(g.type.isin(["article", "review"]).sum()), "Conf.": int((g.type == "conference-paper").sum()), "Preprints": int((g.type == "preprint").sum()), "Grey": int(g.type.isin(["dissertation", "report", "book-chapter", "book"]).sum()), "Abstract": int((g.has_abstract == True).sum()), "No DOI": int(g.doi.isna().sum())})
299+ write("tab_security_landmarks", pd.DataFrame(rows), "Composition of the automatically selected landmark set of the security-and-privacy area, by category.", "tab:security_landmarks", [p],
300+ note="Landmarks are the works with the highest field-normalised 36-month citation score (NCS36; 2025--2026 works ranked by 12-month citations) within each primary category, with a floor of fifteen per category, plus the 200 most cited works of the corpus overall (literature_synthesis/select_landmarks.py). Grey = dissertations, reports, book chapters and books as typed by OpenAlex. Abstract = works whose abstract is available in the metadata; the synthesis of Part VII reads these abstracts only, never full texts. The set is flagged for principal-investigator review; Part VII documents the records that are duplicates, misdated or misclassified.")
301+
302+
303+# ---------------- Part IX (Empirical Results): tables appended by Agent L/I/J (Part IX writer), 2026-11-03 ----------------
304+def _p9_names():
305+ import yaml
306+ o = yaml.safe_load(open(f"{R}/taxonomy/ontology.yaml"))
307+ return {c["id"]: c["name"] for c in o["categories"]}, {c["id"]: c["area"] for c in o["categories"]}, {a["id"]: a["name"] for a in o["areas"]}
308+
309+
310+_P9_AREA_SHORT = {"ai_ml": "AI \\& ML", "perception_interaction": "Perception \\& interaction", "software_pl": "Software \\& PL", "theory": "Theory", "data_distributed": "Data \\& distributed",
311+ "security_privacy": "Security \\& privacy", "hardware_emerging": "Hardware \\& emerging", "computational_science": "Computational science"}
312+_P9_EVENTS = [("gpt3", "GPT-3 (2020-06)"), ("dalle2", "DALL$\\cdot$E 2 (2022-05)"), ("chatgpt", "ChatGPT (2022-12)"), ("gpt4", "GPT-4 (2023-04)"), ("o1", "o1 (2024-10)")]
313+
314+
315+def _p9_estp(e, p):
316+ return f"{e:+.3f} ({p:.3f})" if pd.notna(e) and pd.notna(p) else ""
317+
318+
319+def tables_part09():
320+ import numpy as np
321+ names, area, area_names = _p9_names(); B = f"{R}/results/bibliometrics"; G = f"{R}/results/graph"; E = f"{R}/results/econometrics"; X = f"{R}/results/emerging"
322+ # --- corpus by period (volume_monthly, baseline vs post) ---
323+ p = f"{B}/volume_monthly.csv"
324+ if os.path.exists(p):
325+ m = pd.read_csv(p); m = m[m.period.isin(["baseline", "post"])]; rows = []
326+ tot = m.groupby("period").n.sum()
327+ def share(mask, label):
328+ s = m[mask].groupby("period").n.sum().reindex(["baseline", "post"]).fillna(0) / tot * 100
329+ rows.append({"Indicator": label, "Baseline": float(s["baseline"]), "Post": float(s["post"])})
330+ rows.append({"Indicator": "Eligible works (count)", "Baseline": float(tot["baseline"]), "Post": float(tot["post"])})
331+ share(m.date_precision.eq("month"), "Month-precision first date (\\%)"); share(m.in_dblp.astype(bool), "DBLP-indexed (\\%)"); share(m.cs_primary.astype(bool), "OpenAlex primary field CS (\\%)")
332+ share(m.has_arxiv.astype(bool), "arXiv-linked (\\%)")
333+ for c, lab in [("n_abstract", "With abstract (\\%)"), ("n_with_refs", "With at least one reference (\\%)"), ("n_oa", "Open access (\\%)")]:
334+ s = m.groupby("period")[c].sum() / tot * 100; rows.append({"Indicator": lab, "Baseline": float(s["baseline"]), "Post": float(s["post"])})
335+ for ty, lab in [("article", "Type: article (\\%)"), ("conference-paper", "Type: conference paper (\\%)"), ("preprint", "Type: preprint (\\%)"), ("book-chapter", "Type: book chapter (\\%)"), ("dissertation", "Type: dissertation (\\%)")]:
336+ share(m.type.eq(ty), lab)
337+ t = pd.DataFrame(rows)
338+ write("tab_p9_periods", t, "The eligible corpus by protocol period: completeness, source membership and document type.", "tab:p9-periods", [p], fmt={"Baseline": "{:,.2f}", "Post": "{:,.2f}"},
339+ note="Baseline: first date 2018-01-01 to 2022-11-29; post: 2022-11-30 to 2026-11-02 (right-truncated, see Part III). Percentages are over all eligible works of the period. The count row is an integer written with two decimals by the table generator.")
340+ # --- classification coverage by year ---
341+ p = f"{G}/category_counts_yearly.csv"; q = f"{R}/data/metadata/prisma_counts.json"
342+ if os.path.exists(p) and os.path.exists(q):
343+ cc = pd.read_csv(p); cc = cc[(cc.year >= 2018) & (cc.year <= 2026)]; P = json.load(open(q))["eligible_by_year"]
344+ y = cc.groupby("year")[["n_primary", "n_multilabel"]].sum(); y["eligible"] = [P[str(i)] for i in y.index]
345+ t = pd.DataFrame({"Year": y.index.astype(str), "Eligible works": y.eligible, "With primary category": y.n_primary, "Coverage (%)": 100 * y.n_primary / y.eligible, "Category assignments": y.n_multilabel, "Labels per classified work": y.n_multilabel / y.n_primary})
346+ write("tab_p9_classification", t, "Classification coverage by publication year: eligible works, works with a primary category, and multi-label assignments.", "tab:p9-classification", [p, q], fmt={"Coverage (%)": "{:.1f}", "Labels per classified work": "{:.2f}"},
347+ note="A work receives a primary category when at least one category score passes the threshold of Part IV; works without any assignment (no abstract, unmapped venue and no arXiv category) have no primary category and enter no category series. Assignments count every (work, category) pair. 2026 is incomplete.")
348+ cc["area"] = cc.category_id.map(area)
349+ ya = cc.groupby(["year", "area"]).n_primary.sum().unstack().fillna(0); sh = ya.div(ya.sum(1), axis=0) * 100
350+ per = cc[cc.period.isin(["baseline", "post"])].groupby(["period", "area"]).n_primary.sum().unstack().fillna(0); shp = per.div(per.sum(1), axis=0) * 100
351+ t = pd.concat([sh, shp.loc[["baseline", "post"]]]); t.index = [str(i) for i in sh.index] + ["Baseline (2018-01 to 2022-11)", "Post (2022-12 to 2026-11)"]
352+ t = t[[a for a in _P9_AREA_SHORT if a in t.columns]].rename(columns=_P9_AREA_SHORT).reset_index().rename(columns={"index": "Year / period"})
353+ write("tab_p9_area_shares", t, "Share of primary-category assignments by top-level area and year (per cent), with the two protocol periods.", "tab:p9-area-shares", [p], fmt={c: "{:.2f}" for c in _P9_AREA_SHORT.values()},
354+ note="Shares of works with a primary category (Table~\\ref{tab:p9-classification}); rows sum to 100. The period rows split at the reference date; the year rows split at the calendar year. 2026 is incomplete.")
355+ # category-level periods joined with CAGR
356+ cp = cc[cc.period.isin(["baseline", "post"])].groupby(["period", "category_id"])[["n_primary", "n_multilabel"]].sum().unstack(0)
357+ cp.columns = ["_".join(c) for c in cp.columns]
358+ r = f"{B}/cagr_by_category.csv"; cg = pd.read_csv(r).set_index("category") if os.path.exists(r) else None
359+ rows = []
360+ for k in cp.index:
361+ rr = cp.loc[k]; rows.append({"Category": names.get(k, k), "Area": _P9_AREA_SHORT.get(area.get(k), area.get(k)), "Primary, base": int(rr.n_primary_baseline), "Primary, post": int(rr.n_primary_post),
362+ "Share base (%)": 100 * rr.n_primary_baseline / cp.n_primary_baseline.sum(), "Share post (%)": 100 * rr.n_primary_post / cp.n_primary_post.sum(),
363+ "Multi-label post/base": rr.n_multilabel_post / rr.n_multilabel_baseline if rr.n_multilabel_baseline > 0 else float("nan"),
364+ "CAGR base": float(cg.loc[k, "cagr_baseline"]) if cg is not None and k in cg.index else float("nan"), "CAGR post": float(cg.loc[k, "cagr_post"]) if cg is not None and k in cg.index else float("nan")})
365+ t = pd.DataFrame(rows).sort_values("Share post (%)", ascending=False)
366+ write("tab_p9_category_periods", t, "The forty categories by protocol period: primary-category output, composition shares, multi-label growth and compound annual growth rates.", "tab:p9-category-periods", [p, r], longtable=True,
367+ fmt={"Share base (%)": "{:.2f}", "Share post (%)": "{:.2f}", "Multi-label post/base": "{:.2f}", "CAGR base": "{:.3f}", "CAGR post": "{:.3f}"},
368+ note="Primary counts and shares use the primary category; the multi-label ratio is the post-period count of all assignments to the category divided by the baseline count (periods split at the reference date). CAGR base: 2018 to 2022 (four years); CAGR post: 2023 to 2025 (two years), both on primary counts of complete calendar years (equation~\\eqref{eq:cagr}). Sorted by post-period share.")
369+ # --- AI categories: share with bootstrap CI (H2) ---
370+ p = f"{B}/category_share_ci_yearly.csv"
371+ if os.path.exists(p):
372+ s = pd.read_csv(p); ai = [k for k, a in area.items() if a == "ai_ml"]; rows = []
373+ for k in ai:
374+ r = {"Category": names[k]}
375+ for yv in (2018, 2022, 2025):
376+ z = s[(s.category == k) & (s.year == yv)]
377+ r[str(yv)] = f"{100 * float(z.share.iloc[0]):.2f} [{100 * float(z.ci_low.iloc[0]):.2f}, {100 * float(z.ci_high.iloc[0]):.2f}]" if len(z) else ""
378+ rows.append(r)
379+ r = {"Category": "AI \\& ML area (sum)"}
380+ for yv in (2018, 2022, 2025):
381+ z = s[(s.category.isin(ai)) & (s.year == yv)]; r[str(yv)] = f"{100 * float(z.share.sum()):.2f}" if len(z) else ""
382+ rows.append(r)
383+ write("tab_p9_ai_shares", pd.DataFrame(rows), "Share of primary-category assignments held by the AI and machine-learning categories in 2018, 2022 and 2025 (per cent, bootstrap 95\\% intervals).", "tab:p9-ai-shares", [p],
384+ note="Shares are of the year's works with a primary category; intervals from 500 multinomial resamples of the works (seed 44). The area row is the sum of the category shares (no interval). Categories 1--8 and 38 of the mandate.")
385+ # --- ITS total: events, robustness ---
386+ p = f"{E}/its_results.json"
387+ if os.path.exists(p):
388+ J = json.load(open(p)); S = J["series"]; T = S["total"]; rows = []
389+ for key, lab in _P9_EVENTS:
390+ e = T.get("its_" + key)
391+ if e: rows.append({"Event (month)": lab, "Pre-slope (SE)": f"{e['pre_slope']:.4f} ({e['pre_slope_se']:.4f})", "Level change (SE)": f"{e['level_change']:+.3f} ({e['level_se']:.3f})", "HAC p": e["level_p"], "Slope change (SE)": f"{e['slope_change']:+.4f} ({e['slope_se']:.4f})", "HAC p ": e["slope_p"], "R2": e["r2"]})
392+ write("tab_p9_its_total_events", pd.DataFrame(rows), "Interrupted time series of the total monthly series at the reference date and at four alternative event dates (log counts, month-of-year dummies, Newey--West standard errors).", "tab:p9-its-total-events", [p],
393+ fmt={"HAC p": "{:.3f}", "HAC p ": "{:.3g}", "R2": "{:.3f}"}, note="Equation~\\eqref{eq:its}; each event is coded as the first full month after the public release. Slopes are monthly log growth rates. A HAC $p$ printed as 0.000 is below 0.0005.")
394+ rows = [("Observations (months), first to last complete month", f"{T['n_months']} ({J['first']} to {J['last_complete']})"),
395+ ("Placebo rank of the level $t$-statistic (all candidate months)", f"rank {T['placebo_chatgpt']['level_t_rank']} of {T['placebo_chatgpt']['n_placebo']}, placebo $p$ = {T['placebo_chatgpt']['level_t_placebo_p']:.3f}"),
396+ ("Placebo rank of the slope $t$-statistic (all candidate months)", f"rank {T['placebo_chatgpt']['slope_t_rank']} of {T['placebo_chatgpt']['n_placebo']}, placebo $p$ = {T['placebo_chatgpt']['slope_t_placebo_p']:.3f}"),
397+ ("Placebo $p$, baseline candidate months only (level; slope)", f"{T['placebo_chatgpt_baseline_only']['level_t_placebo_p']:.3f}; {T['placebo_chatgpt_baseline_only']['slope_t_placebo_p']:.3f} (n = {T['placebo_chatgpt_baseline_only']['n_placebo']}; zero is attainable)"),
398+ ("Donut ITS, $\\pm$3 months removed: level change (HAC $p$)", f"{T['its_chatgpt_donut']['level_change']:+.3f} ({T['its_chatgpt_donut']['level_p']:.3f}), n = {T['its_chatgpt_donut']['n']}"),
399+ ("Donut ITS: slope change (HAC $p$)", f"{T['its_chatgpt_donut']['slope_change']:+.4f} ({T['its_chatgpt_donut']['slope_p']:.2g})"),
400+ ("Bai--Perron break dates, BIC (LWZ)", f"{', '.join(T['bai_perron']['break_dates_bic'])} ({', '.join(T['bai_perron']['break_dates_lwz'])}); m = {T['bai_perron']['m_bic']}, h = {T['bai_perron']['h']}"),
401+ ("Bai--Perron BIC by number of breaks m = 0, 1, 2, 3", ", ".join(f"{T['bai_perron']['bic_by_m'][k]:.1f}" for k in ("0", "1", "2", "3"))),
402+ ("PELT changepoints (piecewise-constant mean)", ", ".join(T["pelt"]) or "none"),
403+ ("BOCPD: max P(run length $\\le$ 3) over 2022-12 to 2023-03; MAP resets", f"{T['bocpd_cp_prob_2022_12_to_2023_03']:.3f}; {', '.join(T['bocpd_resets']) or 'none'}")]
404+ for lm, v in J.get("sensitivity_last_complete_total", {}).items():
405+ e = v["its_chatgpt"]; rows.append((f"Last complete month {lm}: level (p); slope (p); Bai--Perron", f"{e['level_change']:+.3f} ({e['level_p']:.3f}); {e['slope_change']:+.4f} ({e['slope_p']:.2g}); {', '.join(v['bai_perron'])}"))
406+ write("tab_p9_total_robustness", pd.DataFrame(rows, columns=["Diagnostic", "Total series"]), "Robustness diagnostics of the structural-change analysis of the total monthly series.", "tab:p9-total-robustness", [p],
407+ note="Placebo ranks follow equation~\\eqref{eq:placebo}; the baseline-only rank excludes the true event from the candidate set, so a value of zero means that no baseline month yields a larger $t$-statistic. Bai--Perron on the seasonally adjusted log series with trimming $h$; PELT with the $\\ell_2$ cost; BOCPD with hazard 1/60 (Part IV).")
408+ # categories: slopes
409+ rows = []; rows2 = []; rows3 = []; rows4 = []
410+ for sid, v in S.items():
411+ if sid == "total": continue
412+ k = sid[4:]; e = v["its_chatgpt"]; pl = v["placebo_chatgpt"]; plb = v["placebo_chatgpt_baseline_only"]; dn = v.get("its_chatgpt_donut", {})
413+ rows.append({"Category": names.get(k, k), "Works": int(v["total"]), "Pre-slope": e["pre_slope"], "Slope change": e["slope_change"], "SE": e["slope_se"], "HAC p": e["slope_p"], "Placebo p (all)": pl["slope_t_placebo_p"], "Placebo p (base)": plb["slope_t_placebo_p"], "Post slope": e["pre_slope"] + e["slope_change"]})
414+ rows2.append({"Category": names.get(k, k), "Level change": e["level_change"], "SE": e["level_se"], "HAC p": e["level_p"], "Placebo p (all)": pl["level_t_placebo_p"], "Placebo p (base)": plb["level_t_placebo_p"], "Donut level": dn.get("level_change", float("nan")), "Donut p": dn.get("level_p", float("nan")), "BOCPD P": v["bocpd_cp_prob_2022_12_to_2023_03"]})
415+ bp = v["bai_perron"]; inw = any("2022-10-01" <= d <= "2023-06-01" for d in bp["break_dates_bic"]); pin = any("2022-10" <= d <= "2023-06" for d in v["pelt"])
416+ rows3.append({"Category": names.get(k, k), "Bai--Perron (BIC)": ", ".join(d[:7] for d in bp["break_dates_bic"]) or "none", "Bai--Perron (LWZ)": ", ".join(d[:7] for d in bp["break_dates_lwz"]) or "none", "PELT": ", ".join(v["pelt"]) or "none", "BOCPD resets": ", ".join(v["bocpd_resets"]) or "none", "BIC break in window": "yes" if inw else "no", "PELT in window": "yes" if pin else "no"})
417+ r4 = {"Category": names.get(k, k)}
418+ for key, lab in _P9_EVENTS:
419+ ev = v.get("its_" + key); r4[lab.split(" (")[0]] = _p9_estp(ev["level_change"], ev["level_p"]) if ev else ""
420+ rows4.append(r4)
421+ t = pd.DataFrame(rows).sort_values("Slope change", ascending=False)
422+ write("tab_p9_its_slopes", t, "ITS slope change at December 2022 by category: pre-period slope, change, HAC and placebo $p$-values, and implied post-period slope (monthly log growth).", "tab:p9-its-slopes", [p], longtable=True,
423+ fmt={"Pre-slope": "{:.4f}", "Slope change": "{:+.4f}", "SE": "{:.4f}", "HAC p": "{:.3g}", "Placebo p (all)": "{:.3f}", "Placebo p (base)": "{:.3f}", "Post slope": "{:.4f}"},
424+ note="Multi-label monthly series of month-precision works (Part IV); Works: series total over the window. Placebo $p$ (all): rank of the slope $t$-statistic among all candidate event months; (base): among baseline months only, where zero is attainable. Sorted by slope change.")
425+ t = pd.DataFrame(rows2).sort_values("Category")
426+ write("tab_p9_its_levels", t, "ITS level change at December 2022 by category: full-sample and donut estimates, HAC and placebo $p$-values, and the BOCPD change probability over December 2022 to March 2023.", "tab:p9-its-levels", [p], longtable=True,
427+ fmt={"Level change": "{:+.3f}", "SE": "{:.3f}", "HAC p": "{:.3f}", "Placebo p (all)": "{:.3f}", "Placebo p (base)": "{:.3f}", "Donut level": "{:+.3f}", "Donut p": "{:.3f}", "BOCPD P": "{:.3f}"},
428+ note="Donut: the event month and three months on each side removed. BOCPD P: maximum posterior probability that a changepoint occurred within the last four months, taken over December 2022 to March 2023.")
429+ write("tab_p9_unknown_breaks", pd.DataFrame(rows3).sort_values("Category"), "Data-driven break and changepoint dates by category, and whether any falls in the H3 window (October 2022 to June 2023).", "tab:p9-unknown-breaks", [p], longtable=True,
430+ note="Bai--Perron on the seasonally adjusted log series with trimming $h = 15$ months (no break possible after April 2025); PELT with the $\\ell_2$ cost and a six-month minimum segment; BOCPD resets as defined in Part IV. The window test is the one pre-specified in H3; Bai--Perron dates carry no $p$-value.")
431+ write("tab_p9_altdates", pd.DataFrame(rows4).sort_values("Category"), "ITS level change (HAC $p$) by category at the five pre-specified event dates.", "tab:p9-altdates", [p], longtable=True,
432+ note="Each column re-estimates equation~\\eqref{eq:its} with the event at the stated month; entries are the log level change with the HAC $p$-value in parentheses. The five estimates of a row are not independent: they are the same series with the break moved.")
433+ # H3 focus table
434+ h3 = [("llm", "break expected"), ("generative_ai", "break expected"), ("ai_agents", "break expected"), ("ai_safety", "break expected"), ("formal_verification", "no break expected"), ("computational_complexity", "no break expected"), ("operating_systems", "no break expected"), ("computer_architecture", "no break expected")]
435+ rows = []
436+ for k, pred in h3:
437+ v = S.get("cat_" + k)
438+ if not v: continue
439+ bp = v["bai_perron"]; e = v["its_chatgpt"]
440+ rows.append({"Category": names[k], "H3 prediction": pred, "Bai--Perron (BIC)": ", ".join(d[:7] for d in bp["break_dates_bic"]) or "none", "PELT": ", ".join(v["pelt"]) or "none", "In window": "yes" if any("2022-10-01" <= d <= "2023-06-01" for d in bp["break_dates_bic"]) else "no",
441+ "Level change (p)": _p9_estp(e["level_change"], e["level_p"]), "Slope change (p)": f"{e['slope_change']:+.4f} ({e['slope_p']:.2g})", "Placebo p (slope)": v["placebo_chatgpt"]["slope_t_placebo_p"], "BOCPD P": v["bocpd_cp_prob_2022_12_to_2023_03"]})
442+ write("tab_p9_h3", pd.DataFrame(rows), "Hypothesis H3: break dates and ITS estimates for the eight categories named in the protocol.", "tab:p9-h3", [p], fmt={"Placebo p (slope)": "{:.3f}", "BOCPD P": "{:.3f}"},
443+ note="In window: a Bai--Perron (BIC) break date between October 2022 and June 2023. The protocol predicted a break in that window for the first four categories and none for the last four.")
444+ # --- cross-area shares, modularity, flows ---
445+ p = f"{G}/cross_area_share_yearly.csv"
446+ if os.path.exists(p):
447+ cs = pd.read_csv(p); piv = cs.pivot(index="citing_year", columns="citing_area", values="share_cross") * 100
448+ cs["period"] = np.where(cs.citing_year <= 2022, "baseline", "post"); g = cs.groupby(["period", "citing_area"])[["n", "n_cross"]].sum(); w = g.loc["baseline"].n / g.loc["baseline"].n.sum()
449+ yr = cs.groupby("citing_year")[["n", "n_cross"]].sum(); raw = 100 * yr.n_cross / yr.n
450+ adj = pd.Series({y: 100 * float((w.reindex(gg.set_index("citing_area").index) * gg.set_index("citing_area").share_cross).sum()) for y, gg in cs.groupby("citing_year")})
451+ piv["All areas, raw"] = raw; piv["All areas, baseline area weights"] = adj; piv["Internal references (count)"] = yr.n
452+ t = piv.T.reset_index().rename(columns={"citing_area": "Citing area", "index": "Citing area"}); t.columns = ["Citing area"] + [str(c) for c in t.columns[1:]]
453+ t["Citing area"] = t["Citing area"].map(lambda a: _P9_AREA_SHORT.get(a, a))
454+ write("tab_p9_cross_area", t, "Cross-area share of internal references by citing area and citing year (per cent), with the aggregate share and its composition-adjusted counterpart.", "tab:p9-cross-area", [p], fmt={c: "{:,.1f}" for c in t.columns[1:]},
455+ note="Equation~\\eqref{eq:cross-area}: the share of an area's resolved internal references whose cited work has a primary category in another area. The composition-adjusted aggregate fixes each area's weight at its 2018--2022 share of internal references, so that it moves only when the areas' own cross-area rates move. The last row is the number of internal references with both ends classified. 2026 is incomplete.")
456+ p = f"{G}/modularity_yearly.csv"
457+ if os.path.exists(p):
458+ d = pd.read_csv(p); t = pd.DataFrame({"Year": d.year.astype(str), "Internal citations": d.edges.astype(int), "Q (categories)": d.Q_categories, "Q (areas)": d.Q_areas, "Self-share (%)": 100 * d.self_share})
459+ write("tab_p9_modularity", t, "Modularity of the yearly category citation-flow network at the category and area partitions, and the within-category self-share of citations.", "tab:p9-modularity", [p], fmt={"Q (categories)": "{:.4f}", "Q (areas)": "{:.4f}", "Self-share (%)": "{:.2f}"},
460+ note="Equation~\\eqref{eq:modularity} at the fixed partitions; the citation counts are internal references of works of the citing year whose two ends have a primary category. The 2018 network is small because few cited works of 2018 or later existed; 2026 is incomplete.")
461+ p = f"{G}/area_flows_yearly.csv"
462+ if os.path.exists(p):
463+ af = pd.read_csv(p); A = [a for a in _P9_AREA_SHORT if a in set(af.citing_area)]
464+ def mat(mask):
465+ mm = af[mask].pivot_table(index="citing_area", columns="cited_area", values="n", aggfunc="sum", fill_value=0).reindex(index=A, columns=A).fillna(0); return mm.div(mm.sum(1), axis=0) * 100
466+ base = mat(af.citing_year <= 2022); post = mat(af.citing_year >= 2023); diff = (post - base)
467+ t = diff.copy(); t.index = [_P9_AREA_SHORT[a] for a in t.index]; t.columns = [_P9_AREA_SHORT[a] for a in t.columns]; t = t.reset_index().rename(columns={"index": "Citing area"})
468+ write("tab_p9_area_flow_change", t, "Change in the area citation-flow matrix between 2018--2022 and 2023--2026: post-period row share minus baseline row share (percentage points).", "tab:p9-area-flow-change", [p], fmt={c: "{:+.1f}" for c in t.columns[1:]},
469+ note="Rows: citing area; columns: cited area; row shares of internal references with both ends classified, split by citing calendar year (2022 entirely in the first span). Figure~\\ref{fig:p9-area-flow-matrix} shows the two underlying matrices.")
470+ p = f"{G}/category_cooccurrence_yearly.csv"
471+ if os.path.exists(p):
472+ co = pd.read_csv(p); co["period"] = np.where(co.year <= 2022, "baseline", "post"); rows = []
473+ tb = co[co.period == "baseline"].groupby(["cat_a", "cat_b"]).n.sum().sort_values(ascending=False).head(15); tp = co[co.period == "post"].groupby(["cat_a", "cat_b"]).n.sum().sort_values(ascending=False).head(15)
474+ for i in range(15):
475+ rows.append({"Rank": i + 1, "Pair, 2018--2022": f"{names.get(tb.index[i][0], tb.index[i][0])} / {names.get(tb.index[i][1], tb.index[i][1])}", "Works": int(tb.iloc[i]), "Pair, 2023--2026": f"{names.get(tp.index[i][0], tp.index[i][0])} / {names.get(tp.index[i][1], tp.index[i][1])}", "Works ": int(tp.iloc[i])})
476+ write("tab_p9_cooccurrence", pd.DataFrame(rows), "The fifteen most frequent co-classification pairs (works carrying both categories in their label set), 2018--2022 and 2023--2026.", "tab:p9-cooccurrence", [p],
477+ note="Counts of works with both labels, summed over the calendar years of each span (2022 in the first); the post span is incomplete. Co-classification depends on abstract availability and vocabulary (Part IV), so the pairs describe the classifier's view of the literature.")
478+ ll = co[(co.cat_a == "llm") | (co.cat_b == "llm")].copy(); ll["partner"] = np.where(ll.cat_a == "llm", ll.cat_b, ll.cat_a)
479+ g = ll.groupby(["partner", "period"]).n.sum().unstack().fillna(0); g = g.sort_values("post", ascending=False).head(15)
480+ t = pd.DataFrame({"Partner category": [names.get(k, k) for k in g.index], "Works, 2018--2022": g.baseline.astype(int).values, "Works, 2023--2026": g.post.astype(int).values, "Ratio": (g.post / g.baseline.replace(0, np.nan)).values})
481+ write("tab_p9_llm_partners", t, "Categories most often co-assigned with Large Language Models, by span.", "tab:p9-llm-partners", [p], fmt={"Ratio": "{:.2f}"}, note="Works carrying both the LLM label and the partner label; spans split by calendar year; the second span is incomplete.")
482+ # --- openness by year ---
483+ p = f"{B}/volume_yearly.csv"
484+ if os.path.exists(p):
485+ v = pd.read_csv(p); v = v[v.year <= 2026]
486+ t = pd.DataFrame({"Year": v.year.astype(str), "Works": v.n, "DBLP (%)": 100 * v.n_dblp / v.n, "arXiv-linked (%)": 100 * v.n_arxiv / v.n, "Preprint type (%)": 100 * v.n_preprint / v.n, "Open access (%)": 100 * v.n_oa / v.n, "Abstract (%)": 100 * v.n_abstract / v.n, "Conference (%)": 100 * v.n_conf / v.n, "Article (%)": 100 * v.n_article / v.n, "Mean refs.": v.mean_refs})
487+ write("tab_p9_openness", t, "Openness and metadata indicators by publication year: arXiv linkage, preprint share, open access, abstract availability and document type (per cent of eligible works).", "tab:p9-openness", [p], fmt={c: "{:.1f}" for c in t.columns[2:]},
488+ note="arXiv-linked: the work is linked to an arXiv e-print (any version); preprint type: the surviving record's OpenAlex type is preprint; open access: any OA location in OpenAlex. Mean refs.: declared references per work, which falls in recent years because references are indexed with a lag. 2026 is incomplete.")
489+ # --- landmarks ---
490+ p = f"{R}/literature_synthesis/landmarks/selection_report.json"
491+ if os.path.exists(p):
492+ Sj = json.load(open(p)); rows = [{"Group": "Area", "Level": area_names.get(k, k), "Landmarks": v} for k, v in Sj["by_area"].items()] + [{"Group": "Year", "Level": k, "Landmarks": v} for k, v in sorted(Sj["by_year_landmarks"].items())]
493+ write("tab_p9_landmarks", pd.DataFrame(rows), "Composition of the landmark set by area and by publication year.", "tab:p9-landmarks", [p], note="Selection by field-normalised impact within category and year with per-area quotas (Part IV); the specialist review of the set (checkpoint H-3) is pending, so the composition is reported without any assessment of the works.")
494+ # --- emerging: setup and AUC by cutoff ---
495+ fs = {h: f"{X}/metrics_h{h}.json" for h in (12, 24)}
496+ if all(os.path.exists(f) for f in fs.values()):
497+ M = {h: json.load(open(f)) for h, f in fs.items()}; rows = []
498+ for h in (12, 24):
499+ m = M[h]; rows.append({"Horizon (months)": str(h), "Training cutoffs": f"{m['train_cutoffs'][0]} to {m['train_cutoffs'][-1]} ({len(m['train_cutoffs'])})", "Test cutoffs": f"{m['test_cutoffs'][0]} to {m['test_cutoffs'][-1]} ({len(m['test_cutoffs'])})", "Train rows": m["n_train"], "Test rows": m["n_test"], "Positive rate, train": m["pos_rate_train"], "Positive rate, test": m["pos_rate_test"]})
500+ write("tab_p9_emerging_setup", pd.DataFrame(rows), "Out-of-time evaluation design of the emerging-topic detector: cutoffs, panel sizes and positive rates.", "tab:p9-emerging-setup", list(fs.values()), fmt={"Positive rate, train": "{:.4f}", "Positive rate, test": "{:.4f}"},
501+ note="Rows are eligible (topic, cutoff) pairs; the label of equation~\\eqref{eq:label} requires top-decile future growth and a doubling of volume, so the positive rate is below ten per cent.")
502+ cut = sorted(set(M[12]["auc_by_cutoff"]) | set(M[24]["auc_by_cutoff"])); rows = []
503+ for c in cut:
504+ r = {"Test cutoff": c}
505+ for h in (12, 24):
506+ a = M[h]["auc_by_cutoff"].get(c, {})
507+ for mdl, lab in (("zscore_baseline", "z-score"), ("logistic", "logistic"), ("gbm", "GBM")): r[f"h{h}: {lab}"] = a.get(mdl, float("nan"))
508+ rows.append(r)
509+ t = pd.DataFrame(rows); write("tab_p9_emerging_cutoffs", t, "AUC of the three detectors at each test cutoff, by horizon.", "tab:p9-emerging-cutoffs", list(fs.values()), fmt={c: "{:.3f}" for c in t.columns[1:]},
510+ note="Within-cutoff AUC on the eligible topics of the cutoff; blank where the label window of the cutoff extends beyond the panel. Pooled values are in Tables~\\ref{tab:emerging_h12} and~\\ref{tab:emerging_h24}.")
511+
512+
226 if __name__ == "__main__": 513 if __name__ == "__main__":
227 main() 514 main()
228 515
Menu

Theme
↑↓ navigate ↵ open esc close / or ⌘K anywhere