SPB Git
spboucher.ai

spb/wp44_uqo

Public

WP44: category-level actor and impact indicators

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Simon-Pierre Boucher committed (Nov 4, 2026) parent cc7bdc1 Browse files

64 changed files +2,163 −547

64 changed files
Load all 64 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", 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) 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 2026-11-03, 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]
added bibliometrics/05_category_impact.py +28 −0
@@ -0,0 +1,28 @@
1+#!/usr/bin/env python3
2+"""WP44 — category-level actor and impact indicators (after classification): per primary category and year —
3+works, share with a China affiliation (any author country CN), share with a company affiliation, international share,
4+share with an arXiv version, mean/median c36 and Gini of c36 (complete windows only), mean NCS36, share in top 10 % FWCI.
5+Outputs results/bibliometrics/category_actors_yearly.csv and category_impact_yearly.csv (read-only DuckDB)."""
6+import argparse, os, duckdb
7+
8+
9+def main():
10+ ap = argparse.ArgumentParser(); ap.add_argument("--root", required=True); a = ap.parse_args(); R = a.root
11+ out = f"{R}/results/bibliometrics"; con = duckdb.connect(f"{R}/data/processed/wp44.duckdb", read_only=True); con.execute(f"SET threads TO 16; SET memory_limit='{os.environ.get('WP44_DUCKDB_MEM', '30GB')}';")
12+ P = f"read_parquet('{R}/data/processed/classification/primary/*.parquet')"; W = f"read_parquet('{R}/data/processed/corpus/work_citation_windows.parquet')"
13+ con.execute(f"""COPY (SELECT c.publication_year AS year, c.period, p.primary_category, count(*) n,
14+ round(avg(list_contains(c.countries, 'CN')::int), 4) share_cn, round(avg(list_contains(c.countries, 'US')::int), 4) share_us,
15+ round(avg(list_contains(c.institution_types, 'company')::int), 4) share_company, round(avg((c.n_countries >= 2)::int), 4) share_intl,
16+ round(avg((c.arxiv_id_linked IS NOT NULL)::int), 4) share_arxiv, round(avg(c.is_oa::int), 4) share_oa, round(avg(c.n_authors), 3) mean_authors
17+ FROM corpus_works c JOIN {P} p ON p.work_id = c.id WHERE c.eligible GROUP BY 1,2,3 ORDER BY 1,3) TO '{out}/category_actors_yearly.csv' (HEADER)""")
18+ con.execute(f"""COPY (WITH J AS (SELECT w.year, w.c36, w.c36_complete, w.ncs36, w.work_id, p.primary_category, c.is_top10 FROM {W} w JOIN {P} p ON p.work_id = w.work_id JOIN corpus_works c ON c.id = w.work_id),
19+ Rk AS (SELECT *, row_number() OVER (PARTITION BY year, primary_category ORDER BY c36, work_id) rk FROM J)
20+ SELECT year, primary_category, count(*) n, any_value(c36_complete) c36_complete, round(avg(c36), 3) mean_c36, quantile_cont(c36, 0.5) med_c36,
21+ round(sum((c36 = 0)::int) * 1.0 / count(*), 4) share_uncited_36, round(avg(ncs36), 3) mean_ncs36,
22+ round((2 * sum(c36 * rk) / (count(*) * sum(c36)) - (count(*) + 1.0) / count(*)), 4) gini_c36, round(avg(is_top10::int), 4) share_top10_fwci
23+ FROM Rk GROUP BY 1,2 HAVING count(*) >= 200 ORDER BY 1,2) TO '{out}/category_impact_yearly.csv' (HEADER)""")
24+ print(open(f"{out}/category_actors_yearly.csv").read()[:600])
25+
26+
27+if __name__ == "__main__":
28+ main()
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", aggfunc="sum", fill_value=0); base = piv.loc[2018] # aggfunc="sum" over the (year, period) rows (fixed 2026-11-04) 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 2026-11-03)
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 manuscript/tables/TABLES.json +118 −55
@@ -103,7 +103,7 @@
103 "caption": "Fractional output of the fifteen largest countries by year.", 103 "caption": "Fractional output of the fifteen largest countries by year.",
104 "label": "tab:countries", 104 "label": "tab:countries",
105 "sources": [ 105 "sources": [
106− "./results/bibliometrics/countries_yearly.csv" 106+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/countries_yearly.csv"
107 ], 107 ],
108 "rows": 15 108 "rows": 15
109 }, 109 },
@@ -112,7 +112,7 @@
112 "caption": "Structural-change estimates by series: ITS level and slope changes at 2022-12 (HAC p-values), placebo p-values, Bai\u2013Perron (BIC) and PELT break dates.", 112 "caption": "Structural-change estimates by series: ITS level and slope changes at 2022-12 (HAC p-values), placebo p-values, Bai\u2013Perron (BIC) and PELT break dates.",
113 "label": "tab:breaks", 113 "label": "tab:breaks",
114 "sources": [ 114 "sources": [
115− "./results/econometrics/breaks.csv" 115+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/econometrics/breaks.csv"
116 ], 116 ],
117 "rows": 41 117 "rows": 41
118 }, 118 },
@@ -121,7 +121,7 @@
121 "caption": "Out-of-time performance of the emerging-topic detectors at horizon 12 months (test cutoffs 2023-01\u20132024-10).", 121 "caption": "Out-of-time performance of the emerging-topic detectors at horizon 12 months (test cutoffs 2023-01\u20132024-10).",
122 "label": "tab:emerging_h12", 122 "label": "tab:emerging_h12",
123 "sources": [ 123 "sources": [
124− "./results/emerging/metrics_h12.json" 124+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/emerging/metrics_h12.json"
125 ], 125 ],
126 "rows": 3 126 "rows": 3
127 }, 127 },
@@ -130,7 +130,7 @@
130 "caption": "Out-of-time performance of the emerging-topic detectors at horizon 24 months (test cutoffs 2023-01\u20132024-04).", 130 "caption": "Out-of-time performance of the emerging-topic detectors at horizon 24 months (test cutoffs 2023-01\u20132024-04).",
131 "label": "tab:emerging_h24", 131 "label": "tab:emerging_h24",
132 "sources": [ 132 "sources": [
133− "./results/emerging/metrics_h24.json" 133+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/emerging/metrics_h24.json"
134 ], 134 ],
135 "rows": 3 135 "rows": 3
136 }, 136 },
@@ -139,7 +139,7 @@
139 "caption": "Preregistered known-emergence set: detection month, peak-growth month and lead time.", 139 "caption": "Preregistered known-emergence set: detection month, peak-growth month and lead time.",
140 "label": "tab:known_set", 140 "label": "tab:known_set",
141 "sources": [ 141 "sources": [
142− "./results/emerging/known_set_lead_times.csv" 142+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/emerging/known_set_lead_times.csv"
143 ], 143 ],
144 "rows": 12 144 "rows": 12
145 }, 145 },
@@ -148,8 +148,8 @@
148 "caption": "Team size and collaboration indicators by publication year (all eligible works; periods of the source file summed).", 148 "caption": "Team size and collaboration indicators by publication year (all eligible works; periods of the source file summed).",
149 "label": "tab:collaboration", 149 "label": "tab:collaboration",
150 "sources": [ 150 "sources": [
151− "./results/bibliometrics/team_size_yearly.csv", 151+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/team_size_yearly.csv",
152− "./results/bibliometrics/international_collab_yearly.csv" 152+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/international_collab_yearly.csv"
153 ], 153 ],
154 "rows": 9 154 "rows": 9
155 }, 155 },
@@ -158,7 +158,7 @@
158 "caption": "Share of typed affiliations by OpenAlex institution type and year (per cent of affiliation-work pairs).", 158 "caption": "Share of typed affiliations by OpenAlex institution type and year (per cent of affiliation-work pairs).",
159 "label": "tab:institution_types", 159 "label": "tab:institution_types",
160 "sources": [ 160 "sources": [
161− "./results/bibliometrics/institution_types_yearly.csv" 161+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/institution_types_yearly.csv"
162 ], 162 ],
163 "rows": 9 163 "rows": 9
164 }, 164 },
@@ -167,7 +167,7 @@
167 "caption": "Share of fractional output of the fifteen largest countries by year (per cent).", 167 "caption": "Share of fractional output of the fifteen largest countries by year (per cent).",
168 "label": "tab:country_shares", 168 "label": "tab:country_shares",
169 "sources": [ 169 "sources": [
170− "./results/bibliometrics/countries_yearly.csv" 170+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/countries_yearly.csv"
171 ], 171 ],
172 "rows": 15 172 "rows": 15
173 }, 173 },
@@ -176,7 +176,7 @@
176 "caption": "Concentration of output across countries by year, and the two largest producers under full counting.", 176 "caption": "Concentration of output across countries by year, and the two largest producers under full counting.",
177 "label": "tab:country_concentration", 177 "label": "tab:country_concentration",
178 "sources": [ 178 "sources": [
179− "./results/bibliometrics/countries_yearly.csv" 179+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/countries_yearly.csv"
180 ], 180 ],
181 "rows": 9 181 "rows": 9
182 }, 182 },
@@ -185,7 +185,7 @@
185 "caption": "The country co-authorship network by period: edges with at least fifty joint works.", 185 "caption": "The country co-authorship network by period: edges with at least fifty joint works.",
186 "label": "tab:country_network", 186 "label": "tab:country_network",
187 "sources": [ 187 "sources": [
188− "./results/bibliometrics/country_pairs_by_period.csv" 188+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/country_pairs_by_period.csv"
189 ], 189 ],
190 "rows": 2 190 "rows": 2
191 }, 191 },
@@ -194,7 +194,7 @@
194 "caption": "The twenty strongest country pairs by period (number of joint works).", 194 "caption": "The twenty strongest country pairs by period (number of joint works).",
195 "label": "tab:country_pairs", 195 "label": "tab:country_pairs",
196 "sources": [ 196 "sources": [
197− "./results/bibliometrics/country_pairs_by_period.csv" 197+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/country_pairs_by_period.csv"
198 ], 198 ],
199 "rows": 20 199 "rows": 20
200 }, 200 },
@@ -203,7 +203,7 @@
203 "caption": "Institutions with at least two hundred works per period: coverage and concentration (fractional counting).", 203 "caption": "Institutions with at least two hundred works per period: coverage and concentration (fractional counting).",
204 "label": "tab:institution_concentration", 204 "label": "tab:institution_concentration",
205 "sources": [ 205 "sources": [
206− "./results/bibliometrics/top_institutions_by_period.csv" 206+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/top_institutions_by_period.csv"
207 ], 207 ],
208 "rows": 3 208 "rows": 3
209 }, 209 },
@@ -212,7 +212,7 @@
212 "caption": "The twenty-five institutions with the largest fractional output in the post period, with their baseline output.", 212 "caption": "The twenty-five institutions with the largest fractional output in the post period, with their baseline output.",
213 "label": "tab:top_institutions", 213 "label": "tab:top_institutions",
214 "sources": [ 214 "sources": [
215− "./results/bibliometrics/top_institutions_by_period.csv" 215+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/top_institutions_by_period.csv"
216 ], 216 ],
217 "rows": 25 217 "rows": 25
218 }, 218 },
@@ -221,7 +221,7 @@
221 "caption": "Distribution of works over source types among the listed sources, by period (per cent).", 221 "caption": "Distribution of works over source types among the listed sources, by period (per cent).",
222 "label": "tab:source_types", 222 "label": "tab:source_types",
223 "sources": [ 223 "sources": [
224− "./results/bibliometrics/top_sources_by_period.csv" 224+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/top_sources_by_period.csv"
225 ], 225 ],
226 "rows": 2 226 "rows": 2
227 }, 227 },
@@ -230,7 +230,7 @@
230 "caption": "The twenty largest sources (venues) by period, with OpenAlex source type.", 230 "caption": "The twenty largest sources (venues) by period, with OpenAlex source type.",
231 "label": "tab:top_sources", 231 "label": "tab:top_sources",
232 "sources": [ 232 "sources": [
233− "./results/bibliometrics/top_sources_by_period.csv" 233+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/top_sources_by_period.csv"
234 ], 234 ],
235 "rows": 20 235 "rows": 20
236 }, 236 },
@@ -239,7 +239,7 @@
239 "caption": "Age of cited works by citing year (resolved references of eligible works; corpus version of the result file at generation time).", 239 "caption": "Age of cited works by citing year (resolved references of eligible works; corpus version of the result file at generation time).",
240 "label": "tab:cited_age", 240 "label": "tab:cited_age",
241 "sources": [ 241 "sources": [
242− "./results/graph/cited_age_yearly.csv" 242+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/cited_age_yearly.csv"
243 ], 243 ],
244 "rows": 9 244 "rows": 9
245 }, 245 },
@@ -248,7 +248,7 @@
248 "caption": "The ten most cited works of each publication year by 36-month window citations (c36), with total citations to date.", 248 "caption": "The ten most cited works of each publication year by 36-month window citations (c36), with total citations to date.",
249 "label": "tab:top_cited", 249 "label": "tab:top_cited",
250 "sources": [ 250 "sources": [
251− "./results/bibliometrics/top_cited_by_year.csv" 251+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/top_cited_by_year.csv"
252 ], 252 ],
253 "rows": 80 253 "rows": 80
254 }, 254 },
@@ -257,7 +257,7 @@
257 "caption": "DBLP records of 2018--2026 without an OpenAlex counterpart, by DBLP record type and year.", 257 "caption": "DBLP records of 2018--2026 without an OpenAlex counterpart, by DBLP record type and year.",
258 "label": "tab:dblp_only", 258 "label": "tab:dblp_only",
259 "sources": [ 259 "sources": [
260− "./results/bibliometrics/dblp_only_yearly.csv" 260+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/dblp_only_yearly.csv"
261 ], 261 ],
262 "rows": 9 262 "rows": 9
263 }, 263 },
@@ -266,7 +266,7 @@
266 "caption": "The OpenAlex \\texttt{created\\_date} relative to the publication date, by publication year.", 266 "caption": "The OpenAlex \\texttt{created\\_date} relative to the publication date, by publication year.",
267 "label": "tab:indexing_lag", 267 "label": "tab:indexing_lag",
268 "sources": [ 268 "sources": [
269− "./results/bibliometrics/indexing_lag.csv" 269+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/indexing_lag.csv"
270 ], 270 ],
271 "rows": 9 271 "rows": 9
272 }, 272 },
@@ -275,7 +275,7 @@
275 "caption": "Reference coverage by citing year: declared, resolved, internal and external references of eligible works.", 275 "caption": "Reference coverage by citing year: declared, resolved, internal and external references of eligible works.",
276 "label": "tab:reference_coverage", 276 "label": "tab:reference_coverage",
277 "sources": [ 277 "sources": [
278− "./results/graph/reference_coverage_yearly.csv" 278+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/reference_coverage_yearly.csv"
279 ], 279 ],
280 "rows": 9 280 "rows": 9
281 }, 281 },
@@ -284,16 +284,79 @@
284 "caption": "Destination of the resolved references of eligible works by citing year: inside the corpus, computer science outside it, and the seven largest other OpenAlex fields (per cent of resolved references).", 284 "caption": "Destination of the resolved references of eligible works by citing year: inside the corpus, computer science outside it, and the seven largest other OpenAlex fields (per cent of resolved references).",
285 "label": "tab:field_flows", 285 "label": "tab:field_flows",
286 "sources": [ 286 "sources": [
287− "./results/graph/field_flows_yearly.csv" 287+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/field_flows_yearly.csv"
288 ], 288 ],
289 "rows": 9 289 "rows": 9
290 }, 290 },
291+ {
292+ "file": "tab_ai_growth.tex",
293+ "caption": "Volume, compound annual growth and corpus share of the nine AI categories, baseline (2018--2022) and post (2023--2025) periods.",
294+ "label": "tab:ai_growth",
295+ "sources": [
296+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/cagr_by_category.csv"
297+ ],
298+ "rows": 9
299+ },
300+ {
301+ "file": "tab_ai_counts.tex",
302+ "caption": "Yearly multi-label counts of the nine AI categories (all period rows summed; 2026 incomplete).",
303+ "label": "tab:ai_counts",
304+ "sources": [
305+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/category_counts_yearly.csv"
306+ ],
307+ "rows": 9
308+ },
309+ {
310+ "file": "tab_ai_breaks.tex",
311+ "caption": "Structural-change estimates for the nine AI categories: interrupted-time-series level and slope changes at 2022-12 (ChatGPT), level changes at the alternative event dates 2023-04 (GPT-4) and 2024-09 (o1), placebo p-values, and Bai--Perron and PELT break dates.",
312+ "label": "tab:ai_breaks",
313+ "sources": [
314+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/econometrics/breaks.csv"
315+ ],
316+ "rows": 9
317+ },
318+ {
319+ "file": "tab_ai_flows.tex",
320+ "caption": "Destination of the in-corpus references of the nine AI categories in 2019, 2022 and 2025 (per cent of in-corpus references of the citing category).",
321+ "label": "tab:ai_flows",
322+ "sources": [
323+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/category_flows_yearly.csv"
324+ ],
325+ "rows": 27
326+ },
327+ {
328+ "file": "tab_ai_landmark_flags.tex",
329+ "caption": "Specialist screening of the 234 AI/ML landmark candidates: records by primary category and data-quality flag.",
330+ "label": "tab:ai_landmark_flags",
331+ "sources": [
332+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/literature_synthesis/agent_E/contributions.csv"
333+ ],
334+ "rows": 9
335+ },
336+ {
337+ "file": "tab_ai_contribution_types.tex",
338+ "caption": "Contribution type of the 234 AI/ML landmark candidates by primary category (specialist judgement from abstract or metadata).",
339+ "label": "tab:ai_contribution_types",
340+ "sources": [
341+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/literature_synthesis/agent_E/contributions.csv"
342+ ],
343+ "rows": 9
344+ },
345+ {
346+ "file": "tab_ai_landmarks.tex",
347+ "caption": "The 234 AI/ML landmark candidates with the specialist's contribution type and data-quality flag (titles in literature_synthesis/agent_E/contributions.csv).",
348+ "label": "tab:ai_landmarks",
349+ "sources": [
350+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/literature_synthesis/agent_E/contributions.csv"
351+ ],
352+ "rows": 234
353+ },
291 { 354 {
292 "file": "tab_security_counts.tex", 355 "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.).", 356 "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", 357 "label": "tab:security_counts",
295 "sources": [ 358 "sources": [
296− "./results/graph/category_counts_yearly.csv" 359+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/category_counts_yearly.csv"
297 ], 360 ],
298 "rows": 9 361 "rows": 9
299 }, 362 },
@@ -302,8 +365,8 @@
302 "caption": "Growth and structural-change summary for the five security-and-privacy categories (multi-label counts).", 365 "caption": "Growth and structural-change summary for the five security-and-privacy categories (multi-label counts).",
303 "label": "tab:security_summary", 366 "label": "tab:security_summary",
304 "sources": [ 367 "sources": [
305− "./results/graph/category_counts_yearly.csv", 368+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/category_counts_yearly.csv",
306− "./results/econometrics/breaks.csv" 369+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/econometrics/breaks.csv"
307 ], 370 ],
308 "rows": 5 371 "rows": 5
309 }, 372 },
@@ -312,8 +375,8 @@
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.", 375 "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", 376 "label": "tab:security_llm_cooccurrence",
314 "sources": [ 377 "sources": [
315− "./results/graph/category_cooccurrence_yearly.csv", 378+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/category_cooccurrence_yearly.csv",
316− "./results/graph/category_counts_yearly.csv" 379+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/category_counts_yearly.csv"
317 ], 380 ],
318 "rows": 9 381 "rows": 9
319 }, 382 },
@@ -322,7 +385,7 @@
322 "caption": "Composition of the automatically selected landmark set of the security-and-privacy area, by category.", 385 "caption": "Composition of the automatically selected landmark set of the security-and-privacy area, by category.",
323 "label": "tab:security_landmarks", 386 "label": "tab:security_landmarks",
324 "sources": [ 387 "sources": [
325− "./literature_synthesis/landmarks/security_privacy_landmarks.csv" 388+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/literature_synthesis/landmarks/security_privacy_landmarks.csv"
326 ], 389 ],
327 "rows": 5 390 "rows": 5
328 }, 391 },
@@ -331,8 +394,8 @@
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.", 394 "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", 395 "label": "tab:systems_categories",
333 "sources": [ 396 "sources": [
334− "./results/bibliometrics/cagr_by_category.csv", 397+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/cagr_by_category.csv",
335− "./results/econometrics/breaks.csv" 398+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/econometrics/breaks.csv"
336 ], 399 ],
337 "rows": 11 400 "rows": 11
338 }, 401 },
@@ -341,7 +404,7 @@
341 "caption": "The eligible corpus by protocol period: completeness, source membership and document type.", 404 "caption": "The eligible corpus by protocol period: completeness, source membership and document type.",
342 "label": "tab:p9-periods", 405 "label": "tab:p9-periods",
343 "sources": [ 406 "sources": [
344− "./results/bibliometrics/volume_monthly.csv" 407+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/volume_monthly.csv"
345 ], 408 ],
346 "rows": 13 409 "rows": 13
347 }, 410 },
@@ -350,8 +413,8 @@
350 "caption": "Classification coverage by publication year: eligible works, works with a primary category, and multi-label assignments.", 413 "caption": "Classification coverage by publication year: eligible works, works with a primary category, and multi-label assignments.",
351 "label": "tab:p9-classification", 414 "label": "tab:p9-classification",
352 "sources": [ 415 "sources": [
353− "./results/graph/category_counts_yearly.csv", 416+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/category_counts_yearly.csv",
354− "./data/metadata/prisma_counts.json" 417+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/data/metadata/prisma_counts.json"
355 ], 418 ],
356 "rows": 9 419 "rows": 9
357 }, 420 },
@@ -360,7 +423,7 @@
360 "caption": "Share of primary-category assignments by top-level area and year (per cent), with the two protocol periods.", 423 "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", 424 "label": "tab:p9-area-shares",
362 "sources": [ 425 "sources": [
363− "./results/graph/category_counts_yearly.csv" 426+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/category_counts_yearly.csv"
364 ], 427 ],
365 "rows": 11 428 "rows": 11
366 }, 429 },
@@ -369,8 +432,8 @@
369 "caption": "The forty categories by protocol period: primary-category output, composition shares, multi-label growth and compound annual growth rates.", 432 "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", 433 "label": "tab:p9-category-periods",
371 "sources": [ 434 "sources": [
372− "./results/graph/category_counts_yearly.csv", 435+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/category_counts_yearly.csv",
373− "./results/bibliometrics/cagr_by_category.csv" 436+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/cagr_by_category.csv"
374 ], 437 ],
375 "rows": 40 438 "rows": 40
376 }, 439 },
@@ -379,7 +442,7 @@
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).", 442 "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", 443 "label": "tab:p9-ai-shares",
381 "sources": [ 444 "sources": [
382− "./results/bibliometrics/category_share_ci_yearly.csv" 445+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/category_share_ci_yearly.csv"
383 ], 446 ],
384 "rows": 10 447 "rows": 10
385 }, 448 },
@@ -388,7 +451,7 @@
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).", 451 "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", 452 "label": "tab:p9-its-total-events",
390 "sources": [ 453 "sources": [
391− "./results/econometrics/its_results.json" 454+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/econometrics/its_results.json"
392 ], 455 ],
393 "rows": 5 456 "rows": 5
394 }, 457 },
@@ -397,7 +460,7 @@
397 "caption": "Robustness diagnostics of the structural-change analysis of the total monthly series.", 460 "caption": "Robustness diagnostics of the structural-change analysis of the total monthly series.",
398 "label": "tab:p9-total-robustness", 461 "label": "tab:p9-total-robustness",
399 "sources": [ 462 "sources": [
400− "./results/econometrics/its_results.json" 463+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/econometrics/its_results.json"
401 ], 464 ],
402 "rows": 12 465 "rows": 12
403 }, 466 },
@@ -406,7 +469,7 @@
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).", 469 "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", 470 "label": "tab:p9-its-slopes",
408 "sources": [ 471 "sources": [
409− "./results/econometrics/its_results.json" 472+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/econometrics/its_results.json"
410 ], 473 ],
411 "rows": 40 474 "rows": 40
412 }, 475 },
@@ -415,7 +478,7 @@
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.", 478 "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", 479 "label": "tab:p9-its-levels",
417 "sources": [ 480 "sources": [
418− "./results/econometrics/its_results.json" 481+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/econometrics/its_results.json"
419 ], 482 ],
420 "rows": 40 483 "rows": 40
421 }, 484 },
@@ -424,7 +487,7 @@
424 "caption": "Data-driven break and changepoint dates by category, and whether any falls in the H3 window (October 2022 to June 2023).", 487 "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", 488 "label": "tab:p9-unknown-breaks",
426 "sources": [ 489 "sources": [
427− "./results/econometrics/its_results.json" 490+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/econometrics/its_results.json"
428 ], 491 ],
429 "rows": 40 492 "rows": 40
430 }, 493 },
@@ -433,7 +496,7 @@
433 "caption": "ITS level change (HAC $p$) by category at the five pre-specified event dates.", 496 "caption": "ITS level change (HAC $p$) by category at the five pre-specified event dates.",
434 "label": "tab:p9-altdates", 497 "label": "tab:p9-altdates",
435 "sources": [ 498 "sources": [
436− "./results/econometrics/its_results.json" 499+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/econometrics/its_results.json"
437 ], 500 ],
438 "rows": 40 501 "rows": 40
439 }, 502 },
@@ -442,7 +505,7 @@
442 "caption": "Hypothesis H3: break dates and ITS estimates for the eight categories named in the protocol.", 505 "caption": "Hypothesis H3: break dates and ITS estimates for the eight categories named in the protocol.",
443 "label": "tab:p9-h3", 506 "label": "tab:p9-h3",
444 "sources": [ 507 "sources": [
445− "./results/econometrics/its_results.json" 508+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/econometrics/its_results.json"
446 ], 509 ],
447 "rows": 8 510 "rows": 8
448 }, 511 },
@@ -451,7 +514,7 @@
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.", 514 "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", 515 "label": "tab:p9-cross-area",
453 "sources": [ 516 "sources": [
454− "./results/graph/cross_area_share_yearly.csv" 517+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/cross_area_share_yearly.csv"
455 ], 518 ],
456 "rows": 11 519 "rows": 11
457 }, 520 },
@@ -460,7 +523,7 @@
460 "caption": "Modularity of the yearly category citation-flow network at the category and area partitions, and the within-category self-share of citations.", 523 "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", 524 "label": "tab:p9-modularity",
462 "sources": [ 525 "sources": [
463− "./results/graph/modularity_yearly.csv" 526+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/modularity_yearly.csv"
464 ], 527 ],
465 "rows": 9 528 "rows": 9
466 }, 529 },
@@ -469,7 +532,7 @@
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).", 532 "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", 533 "label": "tab:p9-area-flow-change",
471 "sources": [ 534 "sources": [
472− "./results/graph/area_flows_yearly.csv" 535+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/area_flows_yearly.csv"
473 ], 536 ],
474 "rows": 8 537 "rows": 8
475 }, 538 },
@@ -478,7 +541,7 @@
478 "caption": "The fifteen most frequent co-classification pairs (works carrying both categories in their label set), 2018--2022 and 2023--2026.", 541 "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", 542 "label": "tab:p9-cooccurrence",
480 "sources": [ 543 "sources": [
481− "./results/graph/category_cooccurrence_yearly.csv" 544+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/category_cooccurrence_yearly.csv"
482 ], 545 ],
483 "rows": 15 546 "rows": 15
484 }, 547 },
@@ -487,7 +550,7 @@
487 "caption": "Categories most often co-assigned with Large Language Models, by span.", 550 "caption": "Categories most often co-assigned with Large Language Models, by span.",
488 "label": "tab:p9-llm-partners", 551 "label": "tab:p9-llm-partners",
489 "sources": [ 552 "sources": [
490− "./results/graph/category_cooccurrence_yearly.csv" 553+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/category_cooccurrence_yearly.csv"
491 ], 554 ],
492 "rows": 15 555 "rows": 15
493 }, 556 },
@@ -496,7 +559,7 @@
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).", 559 "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", 560 "label": "tab:p9-openness",
498 "sources": [ 561 "sources": [
499− "./results/bibliometrics/volume_yearly.csv" 562+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/volume_yearly.csv"
500 ], 563 ],
501 "rows": 9 564 "rows": 9
502 }, 565 },
@@ -505,7 +568,7 @@
505 "caption": "Composition of the landmark set by area and by publication year.", 568 "caption": "Composition of the landmark set by area and by publication year.",
506 "label": "tab:p9-landmarks", 569 "label": "tab:p9-landmarks",
507 "sources": [ 570 "sources": [
508− "./literature_synthesis/landmarks/selection_report.json" 571+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/literature_synthesis/landmarks/selection_report.json"
509 ], 572 ],
510 "rows": 17 573 "rows": 17
511 }, 574 },
@@ -514,8 +577,8 @@
514 "caption": "Out-of-time evaluation design of the emerging-topic detector: cutoffs, panel sizes and positive rates.", 577 "caption": "Out-of-time evaluation design of the emerging-topic detector: cutoffs, panel sizes and positive rates.",
515 "label": "tab:p9-emerging-setup", 578 "label": "tab:p9-emerging-setup",
516 "sources": [ 579 "sources": [
517− "./results/emerging/metrics_h12.json", 580+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/emerging/metrics_h12.json",
518− "./results/emerging/metrics_h24.json" 581+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/emerging/metrics_h24.json"
519 ], 582 ],
520 "rows": 2 583 "rows": 2
521 }, 584 },
@@ -524,8 +587,8 @@
524 "caption": "AUC of the three detectors at each test cutoff, by horizon.", 587 "caption": "AUC of the three detectors at each test cutoff, by horizon.",
525 "label": "tab:p9-emerging-cutoffs", 588 "label": "tab:p9-emerging-cutoffs",
526 "sources": [ 589 "sources": [
527− "./results/emerging/metrics_h12.json", 590+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/emerging/metrics_h12.json",
528− "./results/emerging/metrics_h24.json" 591+ "/Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/emerging/metrics_h24.json"
529 ], 592 ],
530 "rows": 8 593 "rows": 8
531 } 594 }
added manuscript/tables/tab_ai_breaks.tex +23 −0
@@ -0,0 +1,23 @@
1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/econometrics/breaks.csv
2+\begin{longtable}{lrrrrrrrrrrrr}
3+\caption{Structural-change estimates for the nine AI categories: interrupted-time-series level and slope changes at 2022-12 (ChatGPT), level changes at the alternative event dates 2023-04 (GPT-4) and 2024-09 (o1), placebo p-values, and Bai--Perron and PELT break dates.}\label{tab:ai_breaks}\\
4+\toprule
5+\textbf{Category} & \textbf{Level 2022-12} & \textbf{p} & \textbf{Slope 2022-12} & \textbf{p } & \textbf{Level 2023-04} & \textbf{p } & \textbf{Level 2024-09} & \textbf{p } & \textbf{Placebo p (level)} & \textbf{Placebo p (slope)} & \textbf{Bai--Perron (BIC)} & \textbf{PELT} \\
6+\midrule
7+\endfirsthead
8+\toprule
9+\textbf{Category} & \textbf{Level 2022-12} & \textbf{p} & \textbf{Slope 2022-12} & \textbf{p } & \textbf{Level 2023-04} & \textbf{p } & \textbf{Level 2024-09} & \textbf{p } & \textbf{Placebo p (level)} & \textbf{Placebo p (slope)} & \textbf{Bai--Perron (BIC)} & \textbf{PELT} \\
10+\midrule
11+\endhead
12+Large language models & 0.106 & 0.075 & 0.034 & 0.000 & 0.236 & 0.002 & 0.457 & 0.001 & 0.938 & 0.200 & 2022-07-01 & 2023-10 \\
13+Foundation models & -0.075 & 0.340 & 0.006 & 0.021 & 0.023 & 0.756 & 0.084 & 0.173 & 0.692 & 0.246 & 2021-10-01 & 2020-02; 2023-10 \\
14+Multimodal AI & 0.025 & 0.600 & 0.024 & 0.000 & 0.150 & 0.002 & 0.330 & 0.001 & 0.969 & 0.200 & 2022-04-01 & 2020-12; 2023-12 \\
15+Generative AI & 0.036 & 0.556 & 0.019 & 0.000 & 0.150 & 0.017 & 0.324 & 0.000 & 0.969 & 0.308 & 2021-11-01 & 2020-12; 2023-12 \\
16+AI agents & -0.210 & 0.028 & 0.040 & 0.000 & -0.151 & 0.090 & 0.226 & 0.003 & 0.631 & 0.400 & 2022-07-01; 2024-07-01 & 2023-12; 2025-08 \\
17+Reinforcement learning & -0.164 & 0.013 & 0.012 & 0.000 & -0.158 & 0.006 & 0.054 & 0.071 & 0.615 & 0.385 & 2021-07-01; 2024-12-01 & 2021-01; 2024-12 \\
18+ML theory & -0.080 & 0.090 & 0.009 & 0.000 & -0.047 & 0.296 & 0.081 & 0.026 & 0.662 & 0.215 & 2020-05-01; 2022-07-01 & 2020-02; 2023-12 \\
19+DL architectures & -0.071 & 0.126 & -0.002 & 0.136 & -0.051 & 0.260 & 0.008 & 0.792 & 0.400 & 0.708 & 2021-07-01 & 2020-02; 2023-12 \\
20+AI safety and alignment & -0.206 & 0.002 & 0.016 & 0.000 & -0.080 & 0.244 & 0.150 & 0.010 & 0.446 & 0.308 & 2021-06-01; 2022-09-01 & 2020-02; 2024-02 \\
21+\bottomrule
22+\multicolumn{13}{p{0.95\linewidth}}{\footnotesize Monthly log counts, seasonally adjusted; HAC p-values (Part IV, Section~\ref{sec:m-its}). Level changes are in log points of the monthly count; slope changes in log points per month. Placebo p-values are the share of placebo event dates (Section~\ref{sec:m-placebo}) whose t-statistic exceeds the true one. Bai--Perron dates: breaks selected by BIC with trimming h=15 months; PELT: penalised exact linear-time change points. The categories are multi-label, so the series overlap.}\\
23+\end{longtable}
added manuscript/tables/tab_ai_contribution_types.tex +20 −0
@@ -0,0 +1,20 @@
1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/literature_synthesis/agent_E/contributions.csv
2+\begin{table}[htbp]\centering\caption{Contribution type of the 234 AI/ML landmark candidates by primary category (specialist judgement from abstract or metadata).}\label{tab:ai_contribution_types}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrrrrrrr}
5+\toprule
6+\textbf{Primary category} & \textbf{benchmark} & \textbf{empirical} & \textbf{engineering} & \textbf{method} & \textbf{survey} & \textbf{theory} & \textbf{total} \\
7+\midrule
8+Large language models & 1 & 8 & 6 & 2 & 12 & 4 & 33 \\
9+Foundation models & 1 & 0 & 1 & 16 & 6 & 1 & 25 \\
10+Multimodal AI & 1 & 2 & 0 & 16 & 6 & 0 & 25 \\
11+Generative AI & 0 & 9 & 1 & 5 & 8 & 2 & 25 \\
12+AI agents & 1 & 0 & 4 & 0 & 13 & 7 & 25 \\
13+Reinforcement learning & 0 & 5 & 2 & 11 & 3 & 4 & 25 \\
14+ML theory & 0 & 0 & 1 & 4 & 11 & 9 & 25 \\
15+DL architectures & 0 & 2 & 1 & 11 & 11 & 0 & 25 \\
16+AI safety and alignment & 0 & 7 & 1 & 2 & 14 & 2 & 26 \\
17+\bottomrule
18+\end{tabular}
19+\begin{tablenotes}\footnotesize\item Types follow CLAUDE.md~\S1 item 6: theory, method, empirical, benchmark, engineering, survey. Judged from the abstract for 190 records and from title/venue/type for 44 records without abstract (column judged\_from of the source file).\end{tablenotes}
20+\end{threeparttable}\end{table}
added manuscript/tables/tab_ai_counts.tex +20 −0
@@ -0,0 +1,20 @@
1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/category_counts_yearly.csv
2+\begin{table}[htbp]\centering\caption{Yearly multi-label counts of the nine AI categories (all period rows summed; 2026 incomplete).}\label{tab:ai_counts}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrrrrrrrrr}
5+\toprule
6+\textbf{Category} & \textbf{2018} & \textbf{2019} & \textbf{2020} & \textbf{2021} & \textbf{2022} & \textbf{2023} & \textbf{2024} & \textbf{2025} & \textbf{2026} \\
7+\midrule
8+Large language models & 15{,}397 & 18{,}253 & 21{,}706 & 24{,}153 & 25{,}206 & 40{,}491 & 69{,}900 & 116{,}242 & 132{,}202 \\
9+Foundation models & 4{,}486 & 7{,}802 & 11{,}029 & 14{,}509 & 16{,}948 & 25{,}052 & 36{,}652 & 52{,}535 & 56{,}874 \\
10+Multimodal AI & 3{,}989 & 5{,}553 & 6{,}666 & 8{,}561 & 10{,}369 & 15{,}746 & 24{,}979 & 42{,}707 & 47{,}737 \\
11+Generative AI & 5{,}118 & 8{,}354 & 10{,}228 & 12{,}580 & 13{,}703 & 21{,}583 & 36{,}186 & 56{,}768 & 55{,}673 \\
12+AI agents & 10{,}042 & 10{,}745 & 10{,}642 & 10{,}593 & 10{,}324 & 12{,}255 & 16{,}189 & 29{,}517 & 45{,}112 \\
13+Reinforcement learning & 8{,}157 & 11{,}558 & 14{,}307 & 16{,}674 & 18{,}491 & 22{,}678 & 28{,}572 & 43{,}949 & 44{,}996 \\
14+ML theory & 28{,}277 & 37{,}723 & 43{,}455 & 47{,}121 & 48{,}940 & 58{,}770 & 70{,}964 & 91{,}772 & 84{,}417 \\
15+DL architectures & 43{,}313 & 62{,}096 & 74{,}987 & 90{,}347 & 99{,}122 & 118{,}793 & 142{,}181 & 174{,}461 & 130{,}581 \\
16+AI safety and alignment & 3{,}309 & 5{,}385 & 7{,}749 & 9{,}909 & 11{,}593 & 16{,}606 & 26{,}098 & 44{,}888 & 51{,}629 \\
17+\bottomrule
18+\end{tabular}
19+\begin{tablenotes}\footnotesize\item The source file splits each year into period rows (baseline / pre / post relative to 30 November 2022); they are summed here. 2026 covers publications dated up to the snapshot of 14 October 2026.\end{tablenotes}
20+\end{threeparttable}\end{table}
added manuscript/tables/tab_ai_flows.tex +41 −0
@@ -0,0 +1,41 @@
1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/category_flows_yearly.csv
2+\begin{longtable}{lrrrrrrrrr}
3+\caption{Destination of the in-corpus references of the nine AI categories in 2019, 2022 and 2025 (per cent of in-corpus references of the citing category).}\label{tab:ai_flows}\\
4+\toprule
5+\textbf{Category} & \textbf{Year} & \textbf{Self} & \textbf{LLM} & \textbf{NLP} & \textbf{Vision} & \textbf{HCI} & \textbf{Robotics} & \textbf{Softw. eng.} & \textbf{AI safety} \\
6+\midrule
7+\endfirsthead
8+\toprule
9+\textbf{Category} & \textbf{Year} & \textbf{Self} & \textbf{LLM} & \textbf{NLP} & \textbf{Vision} & \textbf{HCI} & \textbf{Robotics} & \textbf{Softw. eng.} & \textbf{AI safety} \\
10+\midrule
11+\endhead
12+Large language models & 2019 & 3.8 & 3.8 & 69.6 & 4.7 & 0.7 & 0.2 & 0.6 & 0.2 \\
13+Large language models & 2022 & 7.5 & 7.5 & 68.6 & 4.5 & 0.9 & 0.2 & 1.2 & 0.4 \\
14+Large language models & 2025 & 28.4 & 28.4 & 34.2 & 4.5 & 4.3 & 1.0 & 3.4 & 2.4 \\
15+Foundation models & 2019 & 1.2 & 0.0 & 18.7 & 31.3 & 0.0 & 2.0 & 0.0 & 0.0 \\
16+Foundation models & 2022 & 2.9 & 1.5 & 10.5 & 37.4 & 2.9 & 2.1 & 0.3 & 0.4 \\
17+Foundation models & 2025 & 3.6 & 4.1 & 6.7 & 43.8 & 1.9 & 1.0 & 0.6 & 0.6 \\
18+Multimodal AI & 2019 & 12.9 & 0.2 & 15.7 & 51.1 & 1.0 & 1.5 & 0.2 & 0.3 \\
19+Multimodal AI & 2022 & 15.2 & 1.0 & 18.5 & 50.3 & 0.7 & 0.9 & 0.2 & 0.1 \\
20+Multimodal AI & 2025 & 14.3 & 3.7 & 16.1 & 46.2 & 1.5 & 1.7 & 0.3 & 0.3 \\
21+Generative AI & 2019 & 13.3 & 0.5 & 3.0 & 53.6 & 0.5 & 0.6 & 0.3 & 0.3 \\
22+Generative AI & 2022 & 10.7 & 0.2 & 2.1 & 57.9 & 1.3 & 0.6 & 0.4 & 0.3 \\
23+Generative AI & 2025 & 13.6 & 13.7 & 3.5 & 33.8 & 6.6 & 1.5 & 1.3 & 3.7 \\
24+AI agents & 2019 & 45.5 & 0.1 & 2.7 & 1.7 & 1.8 & 7.5 & 2.3 & 0.9 \\
25+AI agents & 2022 & 39.3 & 0.2 & 2.1 & 1.6 & 2.9 & 8.8 & 1.5 & 0.9 \\
26+AI agents & 2025 & 30.7 & 4.2 & 3.8 & 1.9 & 4.3 & 12.0 & 1.3 & 1.5 \\
27+Reinforcement learning & 2019 & 47.2 & 0.0 & 2.2 & 5.0 & 0.8 & 14.4 & 0.4 & 0.3 \\
28+Reinforcement learning & 2022 & 43.5 & 0.1 & 1.2 & 5.0 & 1.0 & 19.2 & 0.6 & 0.3 \\
29+Reinforcement learning & 2025 & 39.0 & 0.5 & 0.9 & 4.4 & 0.8 & 23.1 & 0.5 & 0.3 \\
30+ML theory & 2019 & 44.1 & 0.2 & 3.0 & 13.7 & 0.8 & 1.1 & 1.2 & 1.1 \\
31+ML theory & 2022 & 40.8 & 0.2 & 3.6 & 15.5 & 1.3 & 1.0 & 1.5 & 1.6 \\
32+ML theory & 2025 & 38.5 & 0.7 & 2.8 & 15.4 & 1.3 & 1.1 & 1.6 & 1.7 \\
33+DL architectures & 2019 & 36.9 & 0.2 & 5.1 & 21.8 & 0.7 & 1.5 & 1.2 & 0.2 \\
34+DL architectures & 2022 & 35.1 & 0.4 & 5.4 & 22.1 & 1.0 & 1.4 & 1.0 & 0.3 \\
35+DL architectures & 2025 & 35.3 & 1.0 & 4.3 & 21.8 & 1.0 & 1.4 & 0.7 & 0.3 \\
36+AI safety and alignment & 2019 & 27.1 & 0.1 & 7.0 & 6.6 & 9.1 & 3.2 & 4.1 & 27.1 \\
37+AI safety and alignment & 2022 & 34.4 & 0.8 & 8.2 & 7.3 & 10.0 & 2.3 & 3.6 & 34.4 \\
38+AI safety and alignment & 2025 & 28.9 & 10.0 & 8.1 & 5.2 & 11.5 & 2.5 & 2.7 & 28.9 \\
39+\bottomrule
40+\multicolumn{10}{p{0.95\linewidth}}{\footnotesize Row shares of category-to-category citation flows (Part IV, eq.~\ref{eq:flow}); references outside the eligible corpus are excluded. `Self' is the share going to the citing category itself (for LLM, the Self and LLM columns coincide). Categories are multi-label, so a reference can count under several destination categories.}\\
41+\end{longtable}
added manuscript/tables/tab_ai_growth.tex +20 −0
@@ -0,0 +1,20 @@
1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/cagr_by_category.csv
2+\begin{table}[htbp]\centering\caption{Volume, compound annual growth and corpus share of the nine AI categories, baseline (2018--2022) and post (2023--2025) periods.}\label{tab:ai_growth}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrrrrrrrrr}
5+\toprule
6+\textbf{Category} & \textbf{2018} & \textbf{2022} & \textbf{2023} & \textbf{2025} & \textbf{CAGR 2018--22} & \textbf{CAGR 2023--25} & \textbf{Share 2018 (\%)} & \textbf{Share 2022 (\%)} & \textbf{Share 2025 (\%)} \\
7+\midrule
8+Large language models & 1{,}635 & 3{,}266 & 10{,}389 & 45{,}158 & 0.189 & 1.085 & 0.33 & 0.58 & 5.06 \\
9+Foundation models & 46 & 350 & 571 & 1{,}854 & 0.661 & 0.802 & 0.01 & 0.06 & 0.21 \\
10+Multimodal AI & 470 & 1{,}278 & 1{,}932 & 6{,}749 & 0.284 & 0.869 & 0.09 & 0.23 & 0.76 \\
11+Generative AI & 766 & 2{,}160 & 3{,}646 & 12{,}743 & 0.296 & 0.870 & 0.15 & 0.39 & 1.43 \\
12+AI agents & 5{,}436 & 4{,}521 & 4{,}991 & 8{,}819 & -0.045 & 0.329 & 1.09 & 0.81 & 0.99 \\
13+Reinforcement learning & 3{,}296 & 7{,}017 & 8{,}422 & 13{,}210 & 0.208 & 0.252 & 0.66 & 1.25 & 1.48 \\
14+ML theory & 13{,}534 & 20{,}131 & 22{,}893 & 32{,}798 & 0.104 & 0.197 & 2.71 & 3.59 & 3.67 \\
15+DL architectures & 11{,}992 & 22{,}270 & 25{,}607 & 34{,}141 & 0.167 & 0.155 & 2.40 & 3.97 & 3.82 \\
16+AI safety and alignment & 1{,}052 & 2{,}954 & 4{,}015 & 10{,}885 & 0.294 & 0.647 & 0.21 & 0.53 & 1.22 \\
17+\bottomrule
18+\end{tabular}
19+\begin{tablenotes}\footnotesize\item Counts are multi-label (a work counts in every category it is assigned to), restricted to the eligible mapping corpus. CAGR as in Part IV, eq.~\ref{eq:cagr}; the post-period CAGR uses 2023 and 2025 end points. Shares are over all eligible works of the year.\end{tablenotes}
20+\end{threeparttable}\end{table}
added manuscript/tables/tab_ai_landmark_flags.tex +20 −0
@@ -0,0 +1,20 @@
1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/literature_synthesis/agent_E/contributions.csv
2+\begin{table}[htbp]\centering\caption{Specialist screening of the 234 AI/ML landmark candidates: records by primary category and data-quality flag.}\label{tab:ai_landmark_flags}
3+\begin{threeparttable}\small
4+\begin{tabular}{lrrrrrrrr}
5+\toprule
6+\textbf{Primary category} & \textbf{abstract\_mismatch} & \textbf{identifier\_mismatch} & \textbf{no\_abstract} & \textbf{normalisation\_artefact} & \textbf{off\_scope} & \textbf{ok} & \textbf{year\_implausible} & \textbf{total} \\
7+\midrule
8+Large language models & 1 & 7 & 4 & 2 & 0 & 19 & 0 & 33 \\
9+Foundation models & 0 & 0 & 6 & 0 & 0 & 19 & 0 & 25 \\
10+Multimodal AI & 0 & 0 & 5 & 0 & 0 & 20 & 0 & 25 \\
11+Generative AI & 0 & 0 & 5 & 1 & 1 & 18 & 0 & 25 \\
12+AI agents & 0 & 0 & 2 & 1 & 4 & 18 & 0 & 25 \\
13+Reinforcement learning & 0 & 1 & 4 & 1 & 4 & 12 & 3 & 25 \\
14+ML theory & 1 & 1 & 6 & 2 & 0 & 11 & 4 & 25 \\
15+DL architectures & 0 & 3 & 4 & 0 & 0 & 14 & 4 & 25 \\
16+AI safety and alignment & 1 & 1 & 4 & 0 & 1 & 19 & 0 & 26 \\
17+\bottomrule
18+\end{tabular}
19+\begin{tablenotes}\footnotesize\item Flags assigned by Agent E from title, venue, identifiers, citation counts and (where available) the abstract: identifier\_mismatch = title/abstract do not correspond to the arXiv identifier or citation count attached to the record; abstract\_mismatch = stored abstract belongs to another work; year\_implausible = OpenAlex year contradicts the first publication; normalisation\_artefact = very small c36 with very large NCS36; off\_scope = not an AI contribution; no\_abstract = metadata only. `ok' records were judged from the abstract.\end{tablenotes}
20+\end{threeparttable}\end{table}
added manuscript/tables/tab_ai_landmarks.tex +248 −0
@@ -0,0 +1,248 @@
1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/literature_synthesis/agent_E/contributions.csv
2+\begin{longtable}{lrrrrrrr}
3+\caption{The 234 AI/ML landmark candidates with the specialist's contribution type and data-quality flag (titles in literature_synthesis/agent_E/contributions.csv).}\label{tab:ai_landmarks}\\
4+\toprule
5+\textbf{OpenAlex id} & \textbf{Year} & \textbf{Type} & \textbf{Category} & \textbf{c36} & \textbf{NCS36} & \textbf{Contribution} & \textbf{Flag} \\
6+\midrule
7+\endfirsthead
8+\toprule
9+\textbf{OpenAlex id} & \textbf{Year} & \textbf{Type} & \textbf{Category} & \textbf{c36} & \textbf{NCS36} & \textbf{Contribution} & \textbf{Flag} \\
10+\midrule
11+\endhead
12+7172491899 & 2026 & article & Large language models & 3{,}026 & 3367.3 & engineering & identifier\_mismatch \\
13+7154011589 & 2026 & article & Large language models & 2{,}847 & 3168.1 & empirical & identifier\_mismatch \\
14+4362515116 & 2026 & article & Large language models & 1{,}549 & 1723.7 & survey & ok \\
15+4288253152 & 2024 & dissertation & Large language models & 364 & 3102.1 & engineering & normalisation\_artefact \\
16+4322718191 & 2023 & preprint & Large language models & 3{,}629 & 1206.5 & engineering & ok \\
17+4323655724 & 2023 & article & Large language models & 4{,}227 & 990.7 & survey & ok \\
18+4367369313 & 2026 & article & Large language models & 261 & 290.4 & survey & ok \\
19+4226278401 & 2022 & preprint & Large language models & 3{,}077 & 894.2 & method & identifier\_mismatch \\
20+4396531244 & 2024 & dissertation & Large language models & 20 & 778.8 & engineering & normalisation\_artefact \\
21+4221143046 & 2022 & preprint & Large language models & 2{,}465 & 716.4 & theory & identifier\_mismatch \\
22+4327810158 & 2023 & preprint & Large language models & 1{,}968 & 654.3 & engineering & abstract\_mismatch \\
23+7135096569 & 2026 & article & Large language models & 183 & 203.6 & empirical & ok \\
24+4414281281 & 2025 & article & Large language models & 1{,}099 & 374.4 & method & ok \\
25+4389437528 & 2023 & article & Large language models & 2{,}361 & 553.4 & survey & ok \\
26+4292779060 & 2023 & article & Large language models & 13{,}333 & 534.7 & theory & identifier\_mismatch \\
27+4360836968 & 2023 & preprint & Large language models & 1{,}420 & 472.1 & empirical & ok \\
28+4406800520 & 2025 & article & Large language models & 914 & 311.4 & survey & no\_abstract \\
29+4385223296 & 2023 & book-chapter & Large language models & 612 & 421.3 & survey & no\_abstract \\
30+4287689466 & 2025 & conference-paper & Large language models & 722 & 548.6 & empirical & identifier\_mismatch \\
31+4366420437 & 2023 & article & Large language models & 1{,}541 & 361.2 & survey & ok \\
32+4317910584 & 2023 & article & Large language models & 1{,}530 & 358.6 & survey & ok \\
33+4313564799 & 2023 & preprint & Large language models & 811 & 335.9 & empirical & no\_abstract \\
34+4382618722 & 2026 & article & Large language models & 94 & 98.9 & engineering & ok \\
35+4384464487 & 2023 & article & Large language models & 1{,}398 & 327.7 & empirical & ok \\
36+4392773210 & 2024 & article & Large language models & 1{,}085 & 326.7 & survey & ok \\
37+4387355843 & 2026 & book-chapter & Foundation models & 197 & 188.6 & method & ok \\
38+4392947521 & 2024 & article & Foundation models & 1{,}505 & 186.9 & method & no\_abstract \\
39+7124890304 & 2026 & article & Foundation models & 42 & 46.5 & method & no\_abstract \\
40+4410551352 & 2025 & article & Foundation models & 249 & 69.7 & method & ok \\
41+4322832366 & 2026 & conference-paper & Foundation models & 35 & 126.3 & method & ok \\
42+7133233472 & 2023 & conference-paper & Foundation models & 532 & 122.7 & method & ok \\
43+2990704537 & 2019 & article & Foundation models & 668 & 101.6 & benchmark & ok \\
44+7125271461 & 2026 & article & Foundation models & 25 & 70.6 & engineering & ok \\
45+7119541039 & 2026 & article & Foundation models & 24 & 26.7 & theory & ok \\
46+4393138529 & 2024 & preprint & Foundation models & 95 & 82.9 & survey & ok \\
47+4390690043 & 2026 & article & Foundation models & 22 & 10.0 & survey & ok \\
48+7128029160 & 2026 & article & Foundation models & 20 & 22.3 & method & ok \\
49+4402220065 & 2024 & article & Foundation models & 562 & 69.8 & method & no\_abstract \\
50+7118170538 & 2026 & article & Foundation models & 17 & 17.9 & survey & ok \\
51+4411932262 & 2025 & article & Foundation models & 110 & 26.9 & method & ok \\
52+4411099958 & 2025 & article & Foundation models & 108 & 18.4 & method & ok \\
53+4361229539 & 2023 & preprint & Foundation models & 161 & 53.5 & method & ok \\
54+4407135816 & 2025 & article & Foundation models & 103 & 6.9 & survey & ok \\
55+4385948838 & 2023 & article & Foundation models & 471 & 49.1 & method & no\_abstract \\
56+4406643700 & 2025 & article & Foundation models & 91 & 19.1 & survey & ok \\
57+4406472599 & 2025 & article & Foundation models & 86 & 29.3 & survey & no\_abstract \\
58+4393153153 & 2024 & conference-paper & Foundation models & 113 & 41.8 & method & ok \\
59+4402727669 & 2024 & conference-paper & Foundation models & 184 & 38.6 & method & ok \\
60+4401397240 & 2024 & article & Foundation models & 325 & 38.5 & method & no\_abstract \\
61+4385285835 & 2023 & preprint & Foundation models & 107 & 35.6 & method & ok \\
62+4318718936 & 2023 & preprint & Multimodal AI & 875 & 273.1 & method & ok \\
63+4393148714 & 2024 & conference-paper & Multimodal AI & 923 & 193.7 & method & ok \\
64+4396599905 & 2026 & article & Multimodal AI & 48 & 128.7 & survey & ok \\
65+4396952261 & 2024 & book-chapter & Multimodal AI & 189 & 147.2 & survey & no\_abstract \\
66+7133250524 & 2022 & conference-paper & Multimodal AI & 828 & 111.9 & benchmark & no\_abstract \\
67+4392947532 & 2024 & article & Multimodal AI & 938 & 111.1 & method & no\_abstract \\
68+7138474371 & 2026 & conference-paper & Multimodal AI & 31 & 159.4 & method & ok \\
69+7127319731 & 2026 & article & Multimodal AI & 29 & 77.7 & method & ok \\
70+2619383789 & 2018 & article & Multimodal AI & 587 & 103.1 & survey & ok \\
71+7148475777 & 2026 & conference-paper & Multimodal AI & 26 & 73.2 & empirical & ok \\
72+7160627408 & 2026 & conference-paper & Multimodal AI & 23 & 193.4 & method & ok \\
73+2995460200 & 2020 & conference-paper & Multimodal AI & 670 & 81.1 & method & ok \\
74+4415913530 & 2025 & article & Multimodal AI & 157 & 53.5 & method & ok \\
75+4289785095 & 2022 & preprint & Multimodal AI & 366 & 79.4 & method & ok \\
76+7125233009 & 2026 & article & Multimodal AI & 22 & 24.5 & method & ok \\
77+7143512276 & 2026 & article & Multimodal AI & 21 & 23.4 & survey & ok \\
78+4413447271 & 2025 & article & Multimodal AI & 142 & 46.4 & method & ok \\
79+7131830459 & 2026 & article & Multimodal AI & 20 & 142.1 & survey & ok \\
80+7206206818 & 2026 & article & Multimodal AI & 18 & 48.2 & method & ok \\
81+7133220561 & 2023 & conference-paper & Multimodal AI & 454 & 63.4 & method & no\_abstract \\
82+7164501397 & 2026 & article & Multimodal AI & 17 & 45.6 & empirical & ok \\
83+7165561001 & 2026 & conference-paper & Multimodal AI & 17 & 143.0 & method & no\_abstract \\
84+7138470740 & 2026 & conference-paper & Multimodal AI & 17 & 130.5 & method & ok \\
85+7138363754 & 2026 & conference-paper & Multimodal AI & 17 & 130.5 & method & ok \\
86+4410882873 & 2025 & article & Multimodal AI & 119 & 7.9 & survey & ok \\
87+4400228804 & 2024 & dissertation & Generative AI & 2 & 540.0 & empirical & normalisation\_artefact \\
88+7124262017 & 2026 & book & Generative AI & 87 & 783.7 & survey & ok \\
89+4406472463 & 2025 & article & Generative AI & 498 & 77.1 & method & ok \\
90+7123342962 & 2026 & article & Generative AI & 66 & 176.9 & method & ok \\
91+4383312437 & 2023 & article & Generative AI & 988 & 231.6 & empirical & ok \\
92+7133185348 & 2022 & conference-paper & Generative AI & 1{,}706 & 230.5 & method & no\_abstract \\
93+3195296536 & 2020 & dissertation & Generative AI & 48 & 221.7 & method & no\_abstract \\
94+4322617812 & 2023 & article & Generative AI & 873 & 204.6 & theory & ok \\
95+3096831136 & 2020 & article & Generative AI & 1{,}473 & 198.6 & survey & ok \\
96+4400770292 & 2024 & preprint & Generative AI & 198 & 174.2 & empirical & no\_abstract \\
97+7127274868 & 2026 & article & Generative AI & 48 & 341.0 & empirical & ok \\
98+7124974472 & 2026 & article & Generative AI & 42 & 46.7 & empirical & ok \\
99+7118187240 & 2026 & article & Generative AI & 40 & 44.5 & engineering & off\_scope \\
100+4406199489 & 2025 & article & Generative AI & 278 & 241.4 & survey & ok \\
101+4398243313 & 2024 & article & Generative AI & 463 & 139.4 & survey & ok \\
102+4409269614 & 2025 & review & Generative AI & 271 & 14.2 & survey & ok \\
103+3193917007 & 2021 & book-chapter & Generative AI & 195 & 127.1 & survey & ok \\
104+4288099666 & 2022 & preprint & Generative AI & 575 & 124.8 & method & ok \\
105+4394934695 & 2024 & article & Generative AI & 394 & 118.6 & empirical & ok \\
106+4321648656 & 2026 & article & Generative AI & 33 & 88.5 & survey & ok \\
107+4395957952 & 2023 & book & Generative AI & 400 & 116.7 & survey & no\_abstract \\
108+4394863715 & 2024 & preprint & Generative AI & 131 & 114.3 & empirical & no\_abstract \\
109+4407778399 & 2025 & article & Generative AI & 213 & 184.9 & empirical & ok \\
110+7124432017 & 2026 & article & Generative AI & 30 & 213.1 & theory & ok \\
111+7140170686 & 2026 & article & Generative AI & 30 & 213.1 & empirical & ok \\
112+4406728221 & 2025 & article & AI agents & 694 & 236.4 & survey & ok \\
113+7161734976 & 2026 & article & AI agents & 62 & 175.1 & engineering & ok \\
114+7135409568 & 2026 & article & AI agents & 52 & 23.7 & survey & ok \\
115+7134187809 & 2026 & article & AI agents & 46 & 129.9 & theory & off\_scope \\
116+4390730779 & 2026 & article & AI agents & 44 & 49.0 & survey & ok \\
117+4225355126 & 2022 & book-chapter & AI agents & 268 & 143.6 & survey & no\_abstract \\
118+4407570917 & 2026 & article & AI agents & 40 & 44.5 & engineering & ok \\
119+4410197190 & 2025 & review & AI agents & 267 & 33.1 & survey & ok \\
120+4413427262 & 2025 & article & AI agents & 258 & 87.9 & survey & ok \\
121+4414991289 & 2026 & article & AI agents & 31 & 23.9 & engineering & ok \\
122+7122857583 & 2026 & article & AI agents & 30 & 33.4 & theory & ok \\
123+4401023556 & 2024 & conference-paper & AI agents & 371 & 103.1 & survey & ok \\
124+4414023998 & 2025 & article & AI agents & 189 & 64.4 & survey & ok \\
125+7118083939 & 2026 & article & AI agents & 26 & 32.7 & theory & ok \\
126+4412505619 & 2025 & article & AI agents & 181 & 61.7 & survey & ok \\
127+4416209823 & 2025 & article & AI agents & 165 & 56.2 & survey & ok \\
128+7155930820 & 2026 & article & AI agents & 23 & 80.7 & theory & off\_scope \\
129+7162257503 & 2026 & article & AI agents & 22 & 77.2 & theory & off\_scope \\
130+4412709685 & 2025 & article & AI agents & 150 & 33.5 & engineering & no\_abstract \\
131+7128165235 & 2026 & article & AI agents & 21 & 23.4 & theory & ok \\
132+7118190454 & 2026 & article & AI agents & 21 & 22.2 & survey & ok \\
133+7130481397 & 2026 & article & AI agents & 21 & 9.6 & benchmark & ok \\
134+4390638955 & 2024 & article & AI agents & 240 & 72.3 & survey & off\_scope \\
135+4412203333 & 2025 & article & AI agents & 139 & 47.4 & survey & ok \\
136+4399982941 & 2024 & dissertation & AI agents & 8 & 68.2 & theory & normalisation\_artefact \\
137+2911283634 & 2020 & book-chapter & Reinforcement learning & 532 & 276.0 & survey & no\_abstract \\
138+4389473504 & 2023 & dissertation & Reinforcement learning & 28 & 232.9 & method & ok \\
139+2964083594 & 2025 & conference-paper & Reinforcement learning & 414 & 367.9 & empirical & year\_implausible \\
140+2098774185 & 2018 & article & Reinforcement learning & 1{,}077 & 203.3 & method & year\_implausible \\
141+1591675293 & 2025 & book-chapter & Reinforcement learning & 395 & 665.0 & method & year\_implausible \\
142+7123365826 & 2026 & article & Reinforcement learning & 55 & 155.3 & theory & off\_scope \\
143+2991046523 & 2021 & book-chapter & Reinforcement learning & 530 & 178.4 & survey & no\_abstract \\
144+2973229164 & 2019 & dissertation & Reinforcement learning & 47 & 171.2 & method & ok \\
145+2982316857 & 2019 & article & Reinforcement learning & 1{,}021 & 155.2 & engineering & no\_abstract \\
146+2902907165 & 2018 & article & Reinforcement learning & 733 & 138.3 & method & ok \\
147+7133481537 & 2026 & article & Reinforcement learning & 35 & 122.9 & theory & off\_scope \\
148+7133226216 & 2022 & conference-paper & Reinforcement learning & 482 & 95.8 & empirical & ok \\
149+3041807988 & 2020 & article & Reinforcement learning & 322 & 93.8 & theory & off\_scope \\
150+2781726626 & 2018 & preprint & Reinforcement learning & 775 & 90.8 & method & ok \\
151+2963923407 & 2018 & article & Reinforcement learning & 472 & 89.1 & method & ok \\
152+7122543576 & 2026 & article & Reinforcement learning & 25 & 70.6 & theory & off\_scope \\
153+2899138465 & 2018 & dissertation & Reinforcement learning & 20 & 88.6 & method & ok \\
154+4391631327 & 2024 & preprint & Reinforcement learning & 99 & 86.4 & empirical & identifier\_mismatch \\
155+4409147637 & 2025 & article & Reinforcement learning & 170 & 57.9 & method & ok \\
156+2895974005 & 2018 & dissertation & Reinforcement learning & 18 & 79.8 & empirical & normalisation\_artefact \\
157+7134057781 & 2026 & article & Reinforcement learning & 21 & 23.4 & engineering & ok \\
158+2754517384 & 2018 & conference-paper & Reinforcement learning & 452 & 70.0 & empirical & ok \\
159+2761873684 & 2018 & conference-paper & Reinforcement learning & 442 & 68.4 & method & ok \\
160+7133208539 & 2023 & conference-paper & Reinforcement learning & 293 & 67.6 & method & no\_abstract \\
161+7128436456 & 2026 & article & Reinforcement learning & 19 & 21.0 & survey & ok \\
162+2962862931 & 2024 & conference-paper & ML theory & 12{,}586 & 3498.4 & method & identifier\_mismatch \\
163+1698155719 & 2020 & book-chapter & ML theory & 1{,}212 & 971.7 & survey & ok \\
164+4288283362 & 2026 & article & ML theory & 174 & 576.9 & method & normalisation\_artefact \\
165+46659105 & 2020 & preprint & ML theory & 3{,}906 & 608.2 & theory & abstract\_mismatch \\
166+3145506661 & 2019 & book-chapter & ML theory & 1{,}581 & 491.8 & survey & no\_abstract \\
167+2291445757 & 2024 & book-chapter & ML theory & 769 & 426.8 & survey & ok \\
168+2100415632 & 2026 & preprint & ML theory & 116 & 667.5 & theory & year\_implausible \\
169+1568555062 & 2022 & book-chapter & ML theory & 576 & 308.7 & theory & ok \\
170+2751318774 & 2019 & book-chapter & ML theory & 978 & 304.2 & survey & ok \\
171+3208688400 & 2021 & book-chapter & ML theory & 833 & 280.5 & survey & no\_abstract \\
172+3105488314 & 2020 & dissertation & ML theory & 59 & 272.4 & theory & no\_abstract \\
173+57706852 & 2026 & preprint & ML theory & 72 & 414.3 & theory & year\_implausible \\
174+2946444988 & 2026 & preprint & ML theory & 68 & 2646.0 & method & year\_implausible \\
175+2945976633 & 2019 & article & ML theory & 1{,}556 & 236.6 & theory & no\_abstract \\
176+4400823820 & 2024 & article & ML theory & 782 & 235.4 & survey & ok \\
177+2999309192 & 2020 & article & ML theory & 1{,}781 & 222.8 & theory & ok \\
178+2999615587 & 2020 & article & ML theory & 1{,}767 & 221.0 & method & no\_abstract \\
179+2439568532 & 2018 & article & ML theory & 1{,}085 & 204.8 & theory & ok \\
180+3194464139 & 2020 & dissertation & ML theory & 21 & 197.5 & engineering & normalisation\_artefact \\
181+3041133507 & 2020 & article & ML theory & 1{,}538 & 192.4 & survey & ok \\
182+2436219157 & 2018 & article & ML theory & 956 & 180.4 & survey & ok \\
183+3135028703 & 2021 & article & ML theory & 1{,}518 & 176.9 & survey & no\_abstract \\
184+7124426637 & 2026 & article & ML theory & 46 & 51.2 & survey & ok \\
185+2165856466 & 2026 & preprint & ML theory & 45 & 259.0 & theory & year\_implausible \\
186+4402843978 & 2020 & report & ML theory & 455 & 158.3 & survey & ok \\
187+2964121744 & 2026 & article & DL architectures & 45{,}775 & 50938.5 & engineering & identifier\_mismatch \\
188+1710476689 & 2022 & book-chapter & DL architectures & 2{,}385 & 1018.2 & survey & ok \\
189+4322614756 & 2025 & conference-paper & DL architectures & 1{,}542 & 1370.5 & method & identifier\_mismatch \\
190+2963399829 & 2024 & preprint & DL architectures & 773 & 674.3 & method & year\_implausible \\
191+4382203079 & 2023 & conference-paper & DL architectures & 2{,}278 & 621.0 & empirical & ok \\
192+4406170795 & 2025 & article & DL architectures & 974 & 331.8 & method & ok \\
193+146900863 & 2018 & book-chapter & DL architectures & 999 & 464.0 & survey & no\_abstract \\
194+4396821501 & 2024 & preprint & DL architectures & 531 & 463.2 & method & ok \\
195+2950067852 & 2026 & preprint & DL architectures & 119 & 4630.6 & method & year\_implausible \\
196+4221150249 & 2026 & article & DL architectures & 107 & 119.1 & survey & ok \\
197+7148890333 & 2026 & article & DL architectures & 102 & 113.0 & method & no\_abstract \\
198+3177318507 & 2021 & conference-paper & DL architectures & 1{,}344 & 315.3 & method & ok \\
199+4389326242 & 2023 & preprint & DL architectures & 912 & 303.2 & method & ok \\
200+7118302872 & 2026 & article & DL architectures & 85 & 87.6 & method & ok \\
201+3137147200 & 2022 & book-chapter & DL architectures & 698 & 298.0 & survey & ok \\
202+3140854437 & 2021 & article & DL architectures & 2{,}452 & 293.1 & survey & ok \\
203+4210257598 & 2020 & article & DL architectures & 2{,}239 & 280.0 & survey & ok \\
204+2907492528 & 2020 & article & DL architectures & 2{,}230 & 278.9 & survey & ok \\
205+3152893301 & 2020 & article & DL architectures & 2{,}023 & 253.0 & survey & ok \\
206+7110333790 & 2025 & article & DL architectures & 496 & 169.0 & method & identifier\_mismatch \\
207+4385763767 & 2023 & conference-paper & DL architectures & 921 & 251.1 & survey & ok \\
208+3124006607 & 2026 & conference-paper & DL architectures & 69 & 529.6 & method & year\_implausible \\
209+4285723986 & 2020 & book-chapter & DL architectures & 471 & 244.4 & survey & no\_abstract \\
210+2892035503 & 2019 & article & DL architectures & 908 & 192.7 & survey & no\_abstract \\
211+2962776659 & 2024 & preprint & DL architectures & 212 & 184.9 & empirical & year\_implausible \\
212+4406063062 & 2025 & article & AI safety and alignment & 534 & 74.5 & survey & ok \\
213+7130596601 & 2026 & article & AI safety and alignment & 65 & 29.6 & survey & ok \\
214+3084163594 & 2020 & article & AI safety and alignment & 775 & 225.7 & survey & off\_scope \\
215+2963808661 & 2025 & conference-paper & AI safety and alignment & 390 & 126.3 & engineering & identifier\_mismatch \\
216+7133236347 & 2026 & article & AI safety and alignment & 51 & 44.5 & survey & ok \\
217+7128782475 & 2026 & article & AI safety and alignment & 51 & 23.2 & empirical & ok \\
218+4406571986 & 2025 & article & AI safety and alignment & 347 & 301.3 & empirical & ok \\
219+2981731882 & 2019 & article & AI safety and alignment & 1{,}097 & 166.8 & survey & no\_abstract \\
220+4402698709 & 2026 & article & AI safety and alignment & 44 & 38.4 & survey & ok \\
221+4408314709 & 2025 & article & AI safety and alignment & 303 & 42.3 & survey & ok \\
222+4407243994 & 2025 & article & AI safety and alignment & 300 & 41.8 & theory & ok \\
223+4330336443 & 2023 & preprint & AI safety and alignment & 495 & 146.6 & empirical & abstract\_mismatch \\
224+7128600296 & 2026 & review & AI safety and alignment & 41 & 12.8 & survey & ok \\
225+4399450035 & 2024 & conference-paper & AI safety and alignment & 359 & 138.5 & empirical & ok \\
226+4318586239 & 2023 & book-chapter & AI safety and alignment & 247 & 136.6 & survey & ok \\
227+4360615722 & 2023 & article & AI safety and alignment & 557 & 130.5 & theory & ok \\
228+2670253439 & 2018 & article & AI safety and alignment & 676 & 127.6 & survey & no\_abstract \\
229+7118631510 & 2026 & article & AI safety and alignment & 35 & 38.9 & method & no\_abstract \\
230+7127143535 & 2026 & article & AI safety and alignment & 33 & 50.0 & empirical & no\_abstract \\
231+4409249263 & 2025 & article & AI safety and alignment & 218 & 14.5 & survey & ok \\
232+4366262984 & 2023 & article & AI safety and alignment & 1{,}197 & 109.9 & survey & ok \\
233+2969896603 & 2021 & article & AI safety and alignment & 1{,}794 & 102.3 & survey & ok \\
234+4406152291 & 2025 & article & AI safety and alignment & 199 & 67.8 & empirical & ok \\
235+4410609100 & 2025 & conference-paper & AI safety and alignment & 193 & 171.5 & method & ok \\
236+4411542009 & 2025 & conference-paper & AI safety and alignment & 191 & 61.8 & empirical & ok \\
237+4360620450 & 2023 & article & Large language models & 3{,}137 & 113.5 & survey & ok \\
238+3185341429 & 2022 & article & Large language models & 2{,}462 & 253.7 & survey & ok \\
239+4384071683 & 2023 & article & Large language models & 2{,}448 & 38.1 & benchmark & ok \\
240+4384918448 & 2023 & preprint & Large language models & 2{,}431 & 88.5 & empirical & identifier\_mismatch \\
241+4393065402 & 2024 & article & Large language models & 1{,}820 & 226.1 & survey & ok \\
242+4319083882 & 2023 & article & Large language models & 1{,}590 & 146.0 & theory & no\_abstract \\
243+4399790131 & 2024 & review & AI safety and alignment & 1{,}484 & 81.8 & survey & ok \\
244+3133702157 & 2021 & conference-paper & Large language models & 1{,}464 & 232.1 & theory & ok \\
245+4321499901 & 2023 & article & Large language models & 1{,}389 & 325.5 & empirical & ok \\
246+\bottomrule
247+\multicolumn{8}{p{0.95\linewidth}}{\footnotesize Selection score and windows as in Part IV (Section~\ref{sec:m-score}); c36 and NCS36 as stored in the landmark file.}\\
248+\end{longtable}
modified manuscript/tables/tab_breaks.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/econometrics/breaks.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/econometrics/breaks.csv
2 \begin{longtable}{lrrrrrrrr} 2 \begin{longtable}{lrrrrrrrr}
3 \caption{Structural-change estimates by series: ITS level and slope changes at 2022-12 (HAC p-values), placebo p-values, Bai–Perron (BIC) and PELT break dates.}\label{tab:breaks}\\ 3 \caption{Structural-change estimates by series: ITS level and slope changes at 2022-12 (HAC p-values), placebo p-values, Bai–Perron (BIC) and PELT break dates.}\label{tab:breaks}\\
4 \toprule 4 \toprule
modified manuscript/tables/tab_cited_age.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/graph/cited_age_yearly.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/cited_age_yearly.csv
2 \begin{table}[htbp]\centering\caption{Age of cited works by citing year (resolved references of eligible works; corpus version of the result file at generation time).}\label{tab:cited_age} 2 \begin{table}[htbp]\centering\caption{Age of cited works by citing year (resolved references of eligible works; corpus version of the result file at generation time).}\label{tab:cited_age}
3 \begin{threeparttable}\small 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrrr} 4 \begin{tabular}{lrrrrrrr}
modified manuscript/tables/tab_collaboration.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/bibliometrics/team_size_yearly.csv, ./results/bibliometrics/international_collab_yearly.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/team_size_yearly.csv, /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/international_collab_yearly.csv
2 \begin{table}[htbp]\centering\caption{Team size and collaboration indicators by publication year (all eligible works; periods of the source file summed).}\label{tab:collaboration} 2 \begin{table}[htbp]\centering\caption{Team size and collaboration indicators by publication year (all eligible works; periods of the source file summed).}\label{tab:collaboration}
3 \begin{threeparttable}\small 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrrrr} 4 \begin{tabular}{lrrrrrrrr}
modified manuscript/tables/tab_countries.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/bibliometrics/countries_yearly.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/countries_yearly.csv
2 \begin{table}[htbp]\centering\caption{Fractional output of the fifteen largest countries by year.}\label{tab:countries} 2 \begin{table}[htbp]\centering\caption{Fractional output of the fifteen largest countries by year.}\label{tab:countries}
3 \begin{threeparttable}\small 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrrrrr} 4 \begin{tabular}{lrrrrrrrrr}
modified manuscript/tables/tab_country_concentration.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/bibliometrics/countries_yearly.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/countries_yearly.csv
2 \begin{table}[htbp]\centering\caption{Concentration of output across countries by year, and the two largest producers under full counting.}\label{tab:country_concentration} 2 \begin{table}[htbp]\centering\caption{Concentration of output across countries by year, and the two largest producers under full counting.}\label{tab:country_concentration}
3 \begin{threeparttable}\small 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrrrrr} 4 \begin{tabular}{lrrrrrrrrr}
modified manuscript/tables/tab_country_network.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/bibliometrics/country_pairs_by_period.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/country_pairs_by_period.csv
2 \begin{table}[htbp]\centering\caption{The country co-authorship network by period: edges with at least fifty joint works.}\label{tab:country_network} 2 \begin{table}[htbp]\centering\caption{The country co-authorship network by period: edges with at least fifty joint works.}\label{tab:country_network}
3 \begin{threeparttable}\small 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrr} 4 \begin{tabular}{lrrrrrr}
modified manuscript/tables/tab_country_pairs.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/bibliometrics/country_pairs_by_period.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/country_pairs_by_period.csv
2 \begin{table}[htbp]\centering\caption{The twenty strongest country pairs by period (number of joint works).}\label{tab:country_pairs} 2 \begin{table}[htbp]\centering\caption{The twenty strongest country pairs by period (number of joint works).}\label{tab:country_pairs}
3 \begin{threeparttable}\small 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrr} 4 \begin{tabular}{lrrrr}
modified manuscript/tables/tab_country_shares.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/bibliometrics/countries_yearly.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/countries_yearly.csv
2 \begin{table}[htbp]\centering\caption{Share of fractional output of the fifteen largest countries by year (per cent).}\label{tab:country_shares} 2 \begin{table}[htbp]\centering\caption{Share of fractional output of the fifteen largest countries by year (per cent).}\label{tab:country_shares}
3 \begin{threeparttable}\small 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrrrrr} 4 \begin{tabular}{lrrrrrrrrr}
modified manuscript/tables/tab_dblp_only.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/bibliometrics/dblp_only_yearly.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/dblp_only_yearly.csv
2 \begin{table}[htbp]\centering\caption{DBLP records of 2018--2026 without an OpenAlex counterpart, by DBLP record type and year.}\label{tab:dblp_only} 2 \begin{table}[htbp]\centering\caption{DBLP records of 2018--2026 without an OpenAlex counterpart, by DBLP record type and year.}\label{tab:dblp_only}
3 \begin{threeparttable}\small 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrrrr} 4 \begin{tabular}{lrrrrrrrr}
modified manuscript/tables/tab_emerging_h12.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/emerging/metrics_h12.json 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/emerging/metrics_h12.json
2 \begin{table}[htbp]\centering\caption{Out-of-time performance of the emerging-topic detectors at horizon 12 months (test cutoffs 2023-01–2024-10).}\label{tab:emerging_h12} 2 \begin{table}[htbp]\centering\caption{Out-of-time performance of the emerging-topic detectors at horizon 12 months (test cutoffs 2023-01–2024-10).}\label{tab:emerging_h12}
3 \begin{threeparttable}\small 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrr} 4 \begin{tabular}{lrrr}
modified manuscript/tables/tab_emerging_h24.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/emerging/metrics_h24.json 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/emerging/metrics_h24.json
2 \begin{table}[htbp]\centering\caption{Out-of-time performance of the emerging-topic detectors at horizon 24 months (test cutoffs 2023-01–2024-04).}\label{tab:emerging_h24} 2 \begin{table}[htbp]\centering\caption{Out-of-time performance of the emerging-topic detectors at horizon 24 months (test cutoffs 2023-01–2024-04).}\label{tab:emerging_h24}
3 \begin{threeparttable}\small 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrr} 4 \begin{tabular}{lrrr}
modified manuscript/tables/tab_field_flows.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/graph/field_flows_yearly.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/field_flows_yearly.csv
2 \begin{table}[htbp]\centering\caption{Destination of the resolved references of eligible works by citing year: inside the corpus, computer science outside it, and the seven largest other OpenAlex fields (per cent of resolved references).}\label{tab:field_flows} 2 \begin{table}[htbp]\centering\caption{Destination of the resolved references of eligible works by citing year: inside the corpus, computer science outside it, and the seven largest other OpenAlex fields (per cent of resolved references).}\label{tab:field_flows}
3 \begin{threeparttable}\small 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrrrrrr} 4 \begin{tabular}{lrrrrrrrrrr}
modified manuscript/tables/tab_indexing_lag.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/bibliometrics/indexing_lag.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/indexing_lag.csv
2 \begin{table}[htbp]\centering\caption{The OpenAlex \texttt{created\_date} relative to the publication date, by publication year.}\label{tab:indexing_lag} 2 \begin{table}[htbp]\centering\caption{The OpenAlex \texttt{created\_date} relative to the publication date, by publication year.}\label{tab:indexing_lag}
3 \begin{threeparttable}\small 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrr} 4 \begin{tabular}{lrrrrr}
modified manuscript/tables/tab_institution_concentration.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/bibliometrics/top_institutions_by_period.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/top_institutions_by_period.csv
2 \begin{table}[htbp]\centering\caption{Institutions with at least two hundred works per period: coverage and concentration (fractional counting).}\label{tab:institution_concentration} 2 \begin{table}[htbp]\centering\caption{Institutions with at least two hundred works per period: coverage and concentration (fractional counting).}\label{tab:institution_concentration}
3 \begin{threeparttable}\small 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrr} 4 \begin{tabular}{lrrrrrr}
modified manuscript/tables/tab_institution_types.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/bibliometrics/institution_types_yearly.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/institution_types_yearly.csv
2 \begin{table}[htbp]\centering\caption{Share of typed affiliations by OpenAlex institution type and year (per cent of affiliation-work pairs).}\label{tab:institution_types} 2 \begin{table}[htbp]\centering\caption{Share of typed affiliations by OpenAlex institution type and year (per cent of affiliation-work pairs).}\label{tab:institution_types}
3 \begin{threeparttable}\small 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrrrrr} 4 \begin{tabular}{lrrrrrrrrr}
modified manuscript/tables/tab_known_set.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/emerging/known_set_lead_times.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/emerging/known_set_lead_times.csv
2 \begin{table}[htbp]\centering\caption{Preregistered known-emergence set: detection month, peak-growth month and lead time.}\label{tab:known_set} 2 \begin{table}[htbp]\centering\caption{Preregistered known-emergence set: detection month, peak-growth month and lead time.}\label{tab:known_set}
3 \begin{threeparttable}\small 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrr} 4 \begin{tabular}{lrrrr}
modified manuscript/tables/tab_p9_ai_shares.tex +2 −2
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/bibliometrics/category_share_ci_yearly.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/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} 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 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrr} 4 \begin{tabular}{lrrr}
@@ -14,7 +14,7 @@ Reinforcement Learning & 0.66 [0.64, 0.68] & 1.25 [1.22, 1.28] & 1.48 [1.46, 1.5
14 Machine Learning Theory & 2.71 [2.67, 2.76] & 3.59 [3.54, 3.64] & 3.67 [3.63, 3.71] \\ 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] \\ 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] \\ 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 \\ 17+AI \& ML area (sum) & 7.65 & 11.41 & 18.63 \\
18 \bottomrule 18 \bottomrule
19 \end{tabular} 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} 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}
modified manuscript/tables/tab_p9_altdates.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/econometrics/its_results.json 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/econometrics/its_results.json
2 \begin{longtable}{lrrrrr} 2 \begin{longtable}{lrrrrr}
3 \caption{ITS level change (HAC $p$) by category at the five pre-specified event dates.}\label{tab:p9-altdates}\\ 3 \caption{ITS level change (HAC $p$) by category at the five pre-specified event dates.}\label{tab:p9-altdates}\\
4 \toprule 4 \toprule
modified manuscript/tables/tab_p9_area_flow_change.tex +8 −8
@@ -1,17 +1,17 @@
1−% generated by tables/make_tables.py from ./results/graph/area_flows_yearly.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/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} 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 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrrrr} 4 \begin{tabular}{lrrrrrrrr}
5 \toprule 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} \\ 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 7 \midrule
8−AI \\& ML & -3.1 & +3.1 & +0.2 & -0.9 & +0.1 & +0.5 & -0.2 & +0.2 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 15 Computational science & +2.5 & +0.9 & -0.1 & -0.7 & -0.7 & +0.0 & -0.1 & -1.8 \\
16 \bottomrule 16 \bottomrule
17 \end{tabular} 17 \end{tabular}
modified manuscript/tables/tab_p9_area_shares.tex +2 −2
@@ -1,9 +1,9 @@
1−% generated by tables/make_tables.py from ./results/graph/category_counts_yearly.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/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} 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 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrrrr} 4 \begin{tabular}{lrrrrrrrr}
5 \toprule 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} \\ 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 7 \midrule
8 2018 & 7.65 & 33.45 & 7.89 & 6.60 & 22.66 & 9.99 & 5.75 & 5.99 \\ 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 \\ 9 2019 & 8.89 & 34.69 & 7.21 & 6.15 & 21.14 & 10.27 & 5.48 & 6.17 \\
modified manuscript/tables/tab_p9_category_periods.tex +36 −36
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/graph/category_counts_yearly.csv, ./results/bibliometrics/cagr_by_category.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/category_counts_yearly.csv, /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/cagr_by_category.csv
2 \begin{longtable}{lrrrrrrrr} 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}\\ 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 4 \toprule
@@ -9,45 +9,45 @@
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} \\ 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 10 \midrule
11 \endhead 11 \endhead
12−Computer Vision & Perception \\& interaction & 421{,}385 & 521{,}935 & 15.79 & 16.97 & 1.30 & 0.091 & 0.149 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 51 Autonomous Scientific Discovery & Computational science & 841 & 2{,}539 & 0.03 & 0.08 & 2.50 & 0.287 & 0.382 \\
52 \bottomrule 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.}\\ 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.}\\
modified manuscript/tables/tab_p9_classification.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/graph/category_counts_yearly.csv, ./data/metadata/prisma_counts.json 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/category_counts_yearly.csv, /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/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} 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 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrr} 4 \begin{tabular}{lrrrrr}
modified manuscript/tables/tab_p9_cooccurrence.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/graph/category_cooccurrence_yearly.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/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} 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 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrr} 4 \begin{tabular}{lrrrr}
modified manuscript/tables/tab_p9_cross_area.tex +7 −7
@@ -1,17 +1,17 @@
1−% generated by tables/make_tables.py from ./results/graph/cross_area_share_yearly.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/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} 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 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrrrrr} 4 \begin{tabular}{lrrrrrrrrr}
5 \toprule 5 \toprule
6 \textbf{Citing area} & \textbf{2018} & \textbf{2019} & \textbf{2020} & \textbf{2021} & \textbf{2022} & \textbf{2023} & \textbf{2024} & \textbf{2025} & \textbf{2026} \\ 6 \textbf{Citing area} & \textbf{2018} & \textbf{2019} & \textbf{2020} & \textbf{2021} & \textbf{2022} & \textbf{2023} & \textbf{2024} & \textbf{2025} & \textbf{2026} \\
7 \midrule 7 \midrule
8−AI \\& ML & 46.6 & 48.1 & 49.3 & 51.0 & 55.1 & 55.3 & 54.9 & 55.4 & 53.9 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 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 \\ 17 All areas, baseline area weights & 27.4 & 25.9 & 26.1 & 26.8 & 27.1 & 27.6 & 27.8 & 28.3 & 28.6 \\
modified manuscript/tables/tab_p9_emerging_cutoffs.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/emerging/metrics_h12.json, ./results/emerging/metrics_h24.json 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/emerging/metrics_h12.json, /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/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} 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 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrr} 4 \begin{tabular}{lrrrrrr}
modified manuscript/tables/tab_p9_emerging_setup.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/emerging/metrics_h12.json, ./results/emerging/metrics_h24.json 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/emerging/metrics_h12.json, /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/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} 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 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrr} 4 \begin{tabular}{lrrrrrr}
modified manuscript/tables/tab_p9_h3.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/econometrics/its_results.json 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/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} 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 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrrrr} 4 \begin{tabular}{lrrrrrrrr}
modified manuscript/tables/tab_p9_its_levels.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/econometrics/its_results.json 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/econometrics/its_results.json
2 \begin{longtable}{lrrrrrrrr} 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}\\ 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 4 \toprule
modified manuscript/tables/tab_p9_its_slopes.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/econometrics/its_results.json 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/econometrics/its_results.json
2 \begin{longtable}{lrrrrrrrr} 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}\\ 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 4 \toprule
modified manuscript/tables/tab_p9_its_total_events.tex +2 −2
@@ -1,9 +1,9 @@
1−% generated by tables/make_tables.py from ./results/econometrics/its_results.json 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/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} 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 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrr} 4 \begin{tabular}{lrrrrrr}
5 \toprule 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} \\ 6+\textbf{Event (month)} & \textbf{Pre-slope (SE)} & \textbf{Level change (SE)} & \textbf{HAC p (level)} & \textbf{Slope change (SE)} & \textbf{HAC p (slope)} & \textbf{$R^2$} \\
7 \midrule 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 \\ 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 \\ 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 \\
modified manuscript/tables/tab_p9_landmarks.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./literature_synthesis/landmarks/selection_report.json 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/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} 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 3 \begin{threeparttable}\small
4 \begin{tabular}{lrr} 4 \begin{tabular}{lrr}
modified manuscript/tables/tab_p9_llm_partners.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/graph/category_cooccurrence_yearly.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/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} 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 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrr} 4 \begin{tabular}{lrrr}
modified manuscript/tables/tab_p9_modularity.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/graph/modularity_yearly.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/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} 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 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrr} 4 \begin{tabular}{lrrrr}
modified manuscript/tables/tab_p9_openness.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/bibliometrics/volume_yearly.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/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} 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 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrrrrr} 4 \begin{tabular}{lrrrrrrrrr}
modified manuscript/tables/tab_p9_periods.tex +15 −15
@@ -1,24 +1,24 @@
1−% generated by tables/make_tables.py from ./results/bibliometrics/volume_monthly.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/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} 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 3 \begin{threeparttable}\small
4 \begin{tabular}{lrr} 4 \begin{tabular}{lrr}
5 \toprule 5 \toprule
6 \textbf{Indicator} & \textbf{Baseline} & \textbf{Post} \\ 6 \textbf{Indicator} & \textbf{Baseline} & \textbf{Post} \\
7 \midrule 7 \midrule
8−Eligible works (count) & 3{,}405{,}453.00 & 3{,}705{,}041.00 \\ 8+Eligible works (count) & 3,405,453 & 3,705,041 \\
9−Month-precision first date (\\%) & 77.43 & 82.78 \\ 9+Month-precision first date (\%) & 77.43 & 82.78 \\
10−DBLP-indexed (\\%) & 54.16 & 47.56 \\ 10+DBLP-indexed (\%) & 54.16 & 47.56 \\
11−OpenAlex primary field CS (\\%) & 72.33 & 77.15 \\ 11+OpenAlex primary field CS (\%) & 72.33 & 77.15 \\
12−arXiv-linked (\\%) & 8.70 & 13.66 \\ 12+arXiv-linked (\%) & 8.70 & 13.66 \\
13−With abstract (\\%) & 73.42 & 77.88 \\ 13+With abstract (\%) & 73.42 & 77.88 \\
14−With at least one reference (\\%) & 71.63 & 58.71 \\ 14+With at least one reference (\%) & 71.63 & 58.71 \\
15−Open access (\\%) & 47.04 & 60.23 \\ 15+Open access (\%) & 47.04 & 60.23 \\
16−Type: article (\\%) & 52.63 & 50.31 \\ 16+Type: article (\%) & 52.63 & 50.31 \\
17−Type: conference paper (\\%) & 31.24 & 29.01 \\ 17+Type: conference paper (\%) & 31.24 & 29.01 \\
18−Type: preprint (\\%) & 5.29 & 11.85 \\ 18+Type: preprint (\%) & 5.29 & 11.85 \\
19−Type: book chapter (\\%) & 5.41 & 4.29 \\ 19+Type: book chapter (\%) & 5.41 & 4.29 \\
20−Type: dissertation (\\%) & 3.99 & 2.74 \\ 20+Type: dissertation (\%) & 3.99 & 2.74 \\
21 \bottomrule 21 \bottomrule
22 \end{tabular} 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} 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 first row is a count.\end{tablenotes}
24 \end{threeparttable}\end{table} 24 \end{threeparttable}\end{table}
modified manuscript/tables/tab_p9_total_robustness.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/econometrics/its_results.json 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/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} 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 3 \begin{threeparttable}\small
4 \begin{tabular}{lr} 4 \begin{tabular}{lr}
modified manuscript/tables/tab_p9_unknown_breaks.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/econometrics/its_results.json 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/econometrics/its_results.json
2 \begin{longtable}{lrrrrrr} 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}\\ 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 4 \toprule
modified manuscript/tables/tab_reference_coverage.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/graph/reference_coverage_yearly.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/reference_coverage_yearly.csv
2 \begin{table}[htbp]\centering\caption{Reference coverage by citing year: declared, resolved, internal and external references of eligible works.}\label{tab:reference_coverage} 2 \begin{table}[htbp]\centering\caption{Reference coverage by citing year: declared, resolved, internal and external references of eligible works.}\label{tab:reference_coverage}
3 \begin{threeparttable}\small 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrrr} 4 \begin{tabular}{lrrrrrrr}
modified manuscript/tables/tab_security_counts.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/graph/category_counts_yearly.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/category_counts_yearly.csv
2 \begin{longtable}{lrrrrrrrrrr} 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}\\ 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 4 \toprule
modified manuscript/tables/tab_security_landmarks.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./literature_synthesis/landmarks/security_privacy_landmarks.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/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} 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 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrrrrr} 4 \begin{tabular}{lrrrrrrrrr}
modified manuscript/tables/tab_security_llm_cooccurrence.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/graph/category_cooccurrence_yearly.csv, ./results/graph/category_counts_yearly.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/category_cooccurrence_yearly.csv, /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/category_counts_yearly.csv
2 \begin{longtable}{lrrrrrrrrrrrrrr} 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}\\ 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 4 \toprule
modified manuscript/tables/tab_security_summary.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/graph/category_counts_yearly.csv, ./results/econometrics/breaks.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/graph/category_counts_yearly.csv, /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/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} 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 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrrrrrrr} 4 \begin{tabular}{lrrrrrrrrrrr}
modified manuscript/tables/tab_source_types.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/bibliometrics/top_sources_by_period.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/top_sources_by_period.csv
2 \begin{table}[htbp]\centering\caption{Distribution of works over source types among the listed sources, by period (per cent).}\label{tab:source_types} 2 \begin{table}[htbp]\centering\caption{Distribution of works over source types among the listed sources, by period (per cent).}\label{tab:source_types}
3 \begin{threeparttable}\small 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrrr} 4 \begin{tabular}{lrrrrrrr}
modified manuscript/tables/tab_systems_categories.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/bibliometrics/cagr_by_category.csv, ./results/econometrics/breaks.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/cagr_by_category.csv, /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/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} 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 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrrrrrr} 4 \begin{tabular}{lrrrrrrrrr}
modified manuscript/tables/tab_top_cited.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/bibliometrics/top_cited_by_year.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/top_cited_by_year.csv
2 \begin{longtable}{lrrrr} 2 \begin{longtable}{lrrrr}
3 \caption{The ten most cited works of each publication year by 36-month window citations (c36), with total citations to date.}\label{tab:top_cited}\\ 3 \caption{The ten most cited works of each publication year by 36-month window citations (c36), with total citations to date.}\label{tab:top_cited}\\
4 \toprule 4 \toprule
modified manuscript/tables/tab_top_institutions.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/bibliometrics/top_institutions_by_period.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/top_institutions_by_period.csv
2 \begin{table}[htbp]\centering\caption{The twenty-five institutions with the largest fractional output in the post period, with their baseline output.}\label{tab:top_institutions} 2 \begin{table}[htbp]\centering\caption{The twenty-five institutions with the largest fractional output in the post period, with their baseline output.}\label{tab:top_institutions}
3 \begin{threeparttable}\small 3 \begin{threeparttable}\small
4 \begin{tabular}{lrrrrr} 4 \begin{tabular}{lrrrrr}
modified manuscript/tables/tab_top_sources.tex +1 −1
@@ -1,4 +1,4 @@
1−% generated by tables/make_tables.py from ./results/bibliometrics/top_sources_by_period.csv 1+% generated by tables/make_tables.py from /Users/simon-pierreboucher/Desktop/UQO/Recherche/Recherche-UQO/informatique/wp44/results/bibliometrics/top_sources_by_period.csv
2 \begin{longtable}{lrrrrrr} 2 \begin{longtable}{lrrrrrr}
3 \caption{The twenty largest sources (venues) by period, with OpenAlex source type.}\label{tab:top_sources}\\ 3 \caption{The twenty largest sources (venues) by period, with OpenAlex source type.}\label{tab:top_sources}\\
4 \toprule 4 \toprule
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,818,1421,1599,1906,0.14804838919248775,0.09178523385922643,0.001637437670274501,0.002535539484648477,0.0021347609874109584 2+ai_accelerators,409,710,799,953,0.14804838919248775,0.09178523385922643,0.002409752131363896,0.0027899579837432427,0.002218169672523209
3−ai_agents,5436,4521,4991,8819,-0.045032119445381436,0.3292782120448039,0.010881554004415877,0.008066976784022354,0.00987746964741723 3+ai_agents,2718,2260,2495,4409,-0.045032119445381436,0.3292782120448039,0.016013951816741003,0.00887642508409796,0.010263398920242486
4−ai_safety,1052,2954,4015,10885,0.2944899640897396,0.6465368010192893,0.0021058489353652505,0.005270924445919495,0.01219143407553425 4+ai_safety,526,1477,2007,5442,0.2944899640897396,0.6465368010192893,0.0030990944281110255,0.005799814133692849,0.01266777381186523
5−algorithms,29148,28202,28531,33795,-0.008214426291722088,0.08834786213446155,0.058347228866945176,0.050321804747400674,0.03785112674163344 5+algorithms,9716,9400,9510,11265,-0.008214426291722088,0.08834786213446155,0.05724486970252229,0.03691409573423735,0.02622002241445325
6−applied_cryptography,10782,10299,11219,14612,-0.01139242765304549,0.14124203040296623,0.021582949829950696,0.01837686217621018,0.01636575422248107 6+applied_cryptography,3594,5149,3739,7306,0.09407412599520226,0.3977303237525569,0.0211751813205913,0.020220814408565557,0.01700519163426502
7−autonomous_scientific_discovery,86,236,400,764,0.28707361667658393,0.38202749610852527,0.00017215114870856613,0.00042110296859749515,0.000855696429371444 7+autonomous_scientific_discovery,43,118,200,382,0.28707361667658393,0.38202749610852527,0.0002533480235908253,0.0004633568502205526,0.0008891299212002788
8−blockchain,6441,11770,14349,18012,0.1626681909797465,0.12039239265851642,0.012893320335254354,0.02100161839149372,0.02017382733748488 8+blockchain,2147,5885,7174,9006,0.2867038655679659,0.12039239265851642,0.012649725736034926,0.02310894121650807,0.020962052540130133
9−cloud_computing,9396,7103,7643,10974,-0.06755182852386099,0.19825858205350233,0.01880851387518241,0.012674128754016984,0.012291115989426997 9+cloud_computing,3132,3551,3821,3658,0.03192353243026824,-0.0216259646860103,0.01845316302061546,0.013945863165748243,0.00851423364332623
10−compilers,725,642,692,767,-0.029938529449611395,0.05279699035007379,0.0014512742187640748,0.001145542821354203,0.0008590564938846826 10+compilers,241,321,346,383,0.073549490456021,0.05279699035007379,0.0014238551713437854,0.001260487702718622,0.0008926212690583952
11−computational_biology,12968,19536,20576,28154,0.10787474203944436,0.1697406366696017,0.025958791819217272,0.034858760993731634,0.03153308543524036 11+computational_biology,4322,9768,10288,14077,0.22606494630771068,0.1697406366696017,0.025468350154463738,0.03835652299113863,0.032765135865801896
12−computational_complexity,3837,3541,3499,4209,-0.019870339595214892,0.0967748714322898,0.007680743692962421,0.006318328863575129,0.004714170512073831 12+computational_complexity,1279,1770,1166,2104,0.08469177413199191,0.34326939885786634,0.007535630748201524,0.006952316129792274,0.004898361044937139
13−computer_architecture,19554,15174,16369,20209,-0.06143119143120024,0.1111209535203932,0.03914236699822444,0.027075493413128777,0.02263451458267999 13+computer_architecture,6518,7587,8184,6736,0.03869713083776305,-0.0927735404534441,0.03840284692476742,0.029792274767994346,0.015679255303260413
14−computer_graphics,10198,8287,9297,12440,-0.050553314316049036,0.15674804638240403,0.020413923424766944,0.014786780935455265,0.013933067514896286 14+computer_graphics,3399,4143,3099,6220,0.05073548076582268,0.4167212372990894,0.02002824143084679,0.016270500922786948,0.014477455784988833
15−computer_networks,58744,49141,51881,59555,-0.04364282894058291,0.07140814904065751,0.1175912451132094,0.08768398720275215,0.06670288069530934 15+computer_networks,19581,16380,17293,29777,-0.04364282894058291,0.3122016357047006,0.11536958370402665,0.06432152253301744,0.06930907389670499
16−computer_vision,69738,98704,114261,150757,0.09072763371648507,0.1486553322088684,0.13959856754230215,0.17612096361206425,0.16885108194077325 16+computer_vision,23246,49352,57130,75378,0.20708855155154882,0.1486553322088684,0.13696111991610055,0.19379311247529418,0.17544837634867858
17−cybersecurity,29440,31554,36481,59044,0.017487667119059358,0.2721974435411547,0.05893174206953705,0.056302894369175266,0.06613054970655437 17+cybersecurity,9813,15777,12160,29522,0.12603520471872232,0.5581172943745198,0.057818339647394544,0.06195238157567507,0.06871438097820584
18−database_systems,16265,13625,14487,16681,-0.043311373704336575,0.07305457648307523,0.032558586438893346,0.02431155909805454,0.01868307871511133 18+database_systems,5421,6812,7243,8340,0.05875000550434839,0.07305457648307523,0.03194345429228506,0.02675100459430097,0.01941305787374588
19−deep_learning_architectures,11992,22270,25607,34141,0.16736629336390108,0.15467235481054153,0.02400507645712936,0.03973713182485685,0.0382386541821603 19+deep_learning_architectures,3997,7423,12803,17070,0.16736629336390108,0.15467235481054153,0.023551546503109897,0.029149596198903123,0.039732702407982615
20−distributed_computing,11519,9379,9798,12550,-0.050083482537713886,0.13175688541453123,0.023058245139232245,0.01673527433252503,0.014056269880381703 20+distributed_computing,3839,4689,3266,4183,0.0512554351002068,0.13175688541453123,0.022622603749943536,0.018414508043298996,0.009736981248746508
21−edge_computing,5185,8554,9120,11554,0.1333267023064395,0.1255602893280665,0.010379112861092039,0.015263198277046497,0.01294072846198647 21+edge_computing,1728,4277,3040,5777,0.25423217055612524,0.3785241917965825,0.010183019397817279,0.016794722444011048,0.01344634438422516
22−federated_learning,143,974,1497,2544,0.6154954153196714,0.303609910055608,0.00028625132866656923,0.001737941912771018,0.002849334707226379 22+federated_learning,71,487,748,1272,0.6154954153196714,0.303609910055608,0.0004212647369010234,0.0019123286954017722,0.002960662983682604
23−formal_verification,5307,3995,4385,4623,-0.06853417817252339,0.026779402163364052,0.010623327281353028,0.0071284167777414965,0.0051778594149007665 23+formal_verification,1769,1997,1461,2311,0.030836383835269654,0.2575428068501022,0.010422619854236509,0.007843689053521643,0.005380167049357185
24−foundation_models,46,350,571,1854,0.6608390716251988,0.8019254150494273,9.208084698365164e-05,0.0006245171144454377,0.0020765198691814883 24+foundation_models,23,175,285,927,0.6608390716251988,0.8019254150494273,0.00013551173354858095,0.0006871817693948874,0.00215765297631586
25−generative_ai,766,2160,3646,12743,0.29585451247314687,0.8695087811589706,0.0015333462780321122,0.0038541627634347013,0.01427243403073339 25+generative_ai,255,1080,1823,6371,0.43409875951610655,0.8695087811589706,0.0015043766362059857,0.004240893205408448,0.014830081918658578
26−hci,34225,34801,40587,58148,0.004181154443647861,0.19694422895345198,0.06851015191337995,0.06209662885661622,0.06512701043860042 26+hci,11408,17400,20293,29074,0.11130912772658119,0.19694422895345198,0.06721578377826352,0.06832746501917564,0.06767163175124845
27−information_retrieval,12095,11243,12568,17379,-0.018095875045832766,0.1759241392022266,0.024211257484071016,0.02006127405060016,0.019464853725191522 27+information_retrieval,4031,5621,6284,8689,0.08665554196594227,0.1759241392022266,0.023753832134349083,0.022074241809447767,0.0202253781420676
28−llm,1635,3266,10389,45158,0.1888432761609986,1.0848771832846835,0.003272873583005879,0.005827636845082284,0.05057793109627705 28+llm,545,1088,5194,22579,0.1888432761609986,1.0848771832846835,0.0032110389036511576,0.004274925064464195,0.052554095525604964
29−ml_theory,13534,20131,22893,32798,0.10435777777595234,0.19693985094098543,0.02709178658862481,0.035920440088288874,0.03673446530173379 29+ml_theory,4511,10065,11446,16399,0.22217278554507214,0.19693985094098543,0.026579939157195573,0.039524731999109934,0.038169742350165896
30−multimodal_ai,470,1278,1932,6749,0.2841278026110756,0.8690294865336103,0.000940826045267745,0.002280379635032198,0.007559025133282559 30+multimodal_ai,156,639,966,3374,0.421121021734181,0.8690294865336103,0.000923050938664247,0.0025091951465333316,0.007854368898142254
31−nlp,23372,29678,33965,42914,0.06153590511416507,0.12404497631921241,0.04678507729786753,0.052955482635747717,0.04806460284037453 31+nlp,7790,9892,16982,21457,0.06153590511416507,0.12404497631921241,0.045901162847788904,0.038846058194478984,0.049942567327733986
32−operating_systems,1556,1083,1173,1363,-0.0866132316020718,0.07795075704417309,0.003114734737099173,0.0019324343855554546,0.001526589310514762 32+operating_systems,778,541,586,681,-0.0866132316020718,0.07795075704417309,0.004583831682643304,0.00212633673215618,0.0015862357102041625
33−privacy_preserving_computation,3105,4376,5010,8817,0.08956696494067806,0.3266047789454696,0.006215457171396486,0.007808248265180673,0.009875229604408405 33+privacy_preserving_computation,1035,2188,2505,4408,0.20580406040260257,0.3266047789454696,0.006098028009686143,0.008591735493920076,0.010261071355003742
34−programming_languages,4390,3664,3876,4193,-0.044187433279510024,0.04008910470076277,0.008787715614309364,0.006537802020937383,0.004696250168003226 34+programming_languages,1463,1832,1938,2096,0.057780486211923554,0.04008910470076277,0.008621688554757541,0.007193811437322478,0.0048797405230271845
35−quantum_computing,6224,9216,10544,17471,0.10310847324827721,0.2872300901577727,0.012458938948396692,0.016444427790654727,0.01956789570359751 35+quantum_computing,3112,4608,5272,8735,0.10310847324827721,0.2872300901577727,0.018335326730573216,0.018094477676409378,0.020332446143049833
36−reinforcement_learning,3296,7017,8422,13210,0.20792805552654703,0.25240210895212,0.006597792862132953,0.012520675977324676,0.014795484073294207 36+reinforcement_learning,1648,3508,4211,6605,0.20792805552654703,0.25240210895212,0.009709710299480931,0.013777012788125499,0.015373568401905345
37−robotics,17496,23103,25759,33464,0.07196943376447251,0.1397889079167418,0.035022749974477595,0.04122348255723699,0.03748039962367278 37+robotics,5832,11551,12879,16732,0.18632919081853316,0.1397889079167418,0.034361062176318444,0.04535988690951452,0.038944821574667714
38−scientific_computing,16882,19932,20423,23191,0.04239362249738088,0.06561412676886369,0.033793670843000154,0.03556535750036133,0.02597441870883921 38+scientific_computing,5627,6644,6807,7730,0.04239362249738088,0.06561412676886369,0.033155204141552805,0.026089346719197895,0.017992855150572133
39−software_engineering,29016,23725,26245,36381,-0.04908396762848111,0.17737287914735167,0.05808299687125296,0.04233333868633717,0.04074750235204516 39+software_engineering,9672,11862,13122,18190,0.05236158017872561,0.17737287914735167,0.05698562986442935,0.04658110708255343,0.042339575475376104
40−speech_audio,12096,12698,14688,18373,0.012216438797881812,0.11842973684211522,0.024213259241614136,0.02265748091208048,0.020578155100577932 40+speech_audio,4032,6349,7344,9186,0.12020163164106301,0.11842973684211522,0.023755796072516454,0.024930954593646512,0.02138217806572346
41−sustainable_computing,583,830,1078,1639,0.09232628514238472,0.23304832154514776,0.0011670246476406284,0.001480997728542038,0.0018357152457327182 41+sustainable_computing,291,276,539,819,-0.012971780149642509,0.23304832154514776,0.0017174639273657107,0.0010864016544719173,0.00190743971315086
added results/bibliometrics/category_actors_yearly.csv +731 −0
@@ -0,0 +1,731 @@
1+year,period,primary_category,n,share_cn,share_us,share_company,share_intl,share_arxiv,share_oa,mean_authors
2+2018,baseline,ai_accelerators,816,0.2169,0.2598,0.0686,0.2059,0.0846,0.3407,4.152
3+2018,pre,ai_accelerators,2,0.0,0.5,0.5,0.0,1.0,1.0,3.5
4+2018,pre,ai_agents,15,0.2,0.2667,0.1333,0.2,1.0,0.9333,2.8
5+2018,baseline,ai_agents,5421,0.1443,0.1136,0.0249,0.162,0.0644,0.3935,2.779
6+2018,baseline,ai_safety,1049,0.0305,0.2574,0.0362,0.1163,0.1153,0.4471,2.359
7+2018,pre,ai_safety,3,0.0,0.6667,0.3333,0.6667,1.0,1.0,5.667
8+2018,baseline,algorithms,29011,0.1579,0.1516,0.0271,0.2131,0.09,0.4858,2.763
9+2018,post,algorithms,8,0.0,0.375,0.125,0.5,1.0,1.0,3.375
10+2018,pre,algorithms,129,0.031,0.3488,0.0465,0.4031,0.9612,0.9457,2.829
11+2018,post,applied_cryptography,1,0.0,1.0,0.0,0.0,1.0,1.0,4.0
12+2018,baseline,applied_cryptography,10764,0.1455,0.085,0.0334,0.1335,0.0295,0.3748,2.785
13+2018,pre,applied_cryptography,17,0.2353,0.2353,0.0,0.3529,0.9412,0.8235,2.647
14+2018,baseline,autonomous_scientific_discovery,86,0.0465,0.2558,0.093,0.1744,0.0814,0.4535,3.279
15+2018,baseline,blockchain,6435,0.06,0.0783,0.0242,0.0779,0.0292,0.3711,2.276
16+2018,pre,blockchain,6,0.1667,0.1667,0.1667,0.6667,1.0,1.0,4.0
17+2018,baseline,cloud_computing,9395,0.1214,0.0919,0.0467,0.1236,0.0141,0.3222,2.815
18+2018,pre,cloud_computing,1,0.0,0.0,0.0,1.0,1.0,1.0,3.0
19+2018,baseline,compilers,725,0.08,0.1862,0.0538,0.1269,0.029,0.4193,3.036
20+2018,baseline,computational_biology,12957,0.152,0.2667,0.0485,0.2222,0.0203,0.5632,4.615
21+2018,pre,computational_biology,11,0.0,0.3636,0.0909,0.3636,1.0,1.0,5.727
22+2018,baseline,computational_complexity,3822,0.0905,0.1434,0.016,0.1617,0.0858,0.505,2.388
23+2018,pre,computational_complexity,15,0.0,0.4667,0.0667,0.4,1.0,1.0,2.8
24+2018,pre,computer_architecture,12,0.0,0.25,0.1667,0.1667,1.0,1.0,4.333
25+2018,baseline,computer_architecture,19542,0.1087,0.1964,0.0902,0.1498,0.0224,0.2796,3.477
26+2018,baseline,computer_graphics,10187,0.1091,0.1343,0.0433,0.1222,0.0295,0.3746,2.787
27+2018,pre,computer_graphics,11,0.1818,0.0909,0.0909,0.1818,1.0,1.0,3.0
28+2018,post,computer_networks,1,0.0,0.0,0.0,0.0,1.0,1.0,1.0
29+2018,baseline,computer_networks,58594,0.2048,0.1244,0.0651,0.1942,0.0558,0.3386,3.309
30+2018,pre,computer_networks,149,0.2282,0.3826,0.1007,0.3624,0.9933,0.9732,3.383
31+2018,pre,computer_vision,148,0.2635,0.3311,0.0811,0.3919,1.0,0.9662,4.284
32+2018,baseline,computer_vision,69585,0.2526,0.1157,0.0496,0.1485,0.0931,0.3713,3.558
33+2018,post,computer_vision,5,0.4,0.2,0.0,0.0,1.0,1.0,4.8
34+2018,baseline,cybersecurity,29406,0.098,0.1238,0.0334,0.1028,0.0443,0.368,2.755
35+2018,post,cybersecurity,2,0.0,0.0,0.0,0.0,1.0,1.0,1.5
36+2018,pre,cybersecurity,32,0.1563,0.5313,0.0938,0.3438,1.0,1.0,3.594
37+2018,pre,database_systems,9,0.1111,0.3333,0.0,0.5556,1.0,1.0,4.556
38+2018,baseline,database_systems,16256,0.1085,0.1054,0.0369,0.122,0.028,0.4086,2.862
39+2018,pre,deep_learning_architectures,41,0.1951,0.439,0.1463,0.4878,1.0,0.9512,4.39
40+2018,baseline,deep_learning_architectures,11950,0.225,0.1468,0.0507,0.1544,0.1396,0.4147,3.261
41+2018,post,deep_learning_architectures,1,0.0,1.0,0.0,0.0,1.0,1.0,2.0
42+2018,baseline,distributed_computing,11489,0.1597,0.1834,0.0498,0.1591,0.0638,0.3925,3.251
43+2018,post,distributed_computing,1,0.0,0.0,0.0,0.0,1.0,1.0,2.0
44+2018,pre,distributed_computing,29,0.1034,0.4483,0.0345,0.4483,1.0,1.0,3.207
45+2018,pre,edge_computing,3,0.0,1.0,0.0,0.3333,1.0,1.0,3.333
46+2018,baseline,edge_computing,5182,0.115,0.1324,0.0556,0.191,0.0152,0.3203,3.361
47+2018,baseline,federated_learning,143,0.1399,0.2448,0.0629,0.1818,0.028,0.4336,3.413
48+2018,baseline,formal_verification,5278,0.0892,0.1408,0.0396,0.2363,0.1465,0.5731,2.672
49+2018,pre,formal_verification,29,0.1034,0.2069,0.0,0.3793,1.0,1.0,2.517
50+2018,baseline,foundation_models,46,0.2174,0.1957,0.0435,0.1087,0.2391,0.4348,3.891
51+2018,baseline,generative_ai,766,0.1841,0.2272,0.0927,0.1253,0.2911,0.4739,3.358
52+2018,pre,hci,12,0.0833,0.25,0.0,0.25,1.0,1.0,4.333
53+2018,post,hci,4,0.25,0.0,0.0,0.0,1.0,1.0,3.75
54+2018,baseline,hci,34209,0.0517,0.1338,0.035,0.1137,0.0169,0.376,2.968
55+2018,pre,information_retrieval,16,0.0625,0.3125,0.0625,0.375,1.0,1.0,3.563
56+2018,baseline,information_retrieval,12077,0.1673,0.124,0.054,0.1456,0.0745,0.3958,2.937
57+2018,post,information_retrieval,2,0.0,0.0,0.0,0.0,1.0,1.0,3.0
58+2018,baseline,llm,1633,0.1249,0.0802,0.0374,0.0986,0.0245,0.2817,2.994
59+2018,pre,llm,2,0.0,0.5,0.0,0.5,1.0,1.0,5.0
60+2018,post,ml_theory,1,0.0,1.0,1.0,1.0,1.0,1.0,5.0
61+2018,baseline,ml_theory,13468,0.1529,0.2073,0.0595,0.161,0.2493,0.5418,3.019
62+2018,pre,ml_theory,65,0.1385,0.4769,0.0923,0.3538,1.0,0.9538,3.385
63+2018,pre,multimodal_ai,1,0.0,0.0,0.0,1.0,1.0,1.0,3.0
64+2018,baseline,multimodal_ai,469,0.3369,0.2132,0.0959,0.1812,0.1279,0.4051,3.62
65+2018,baseline,nlp,23347,0.1141,0.1332,0.0572,0.1247,0.1158,0.5092,3.04
66+2018,pre,nlp,22,0.0909,0.4545,0.1364,0.5,1.0,1.0,4.182
67+2018,post,nlp,3,0.0,0.3333,0.0,0.6667,1.0,1.0,2.333
68+2018,baseline,operating_systems,1554,0.1338,0.1577,0.0727,0.1178,0.0161,0.3333,3.027
69+2018,pre,operating_systems,2,0.0,0.5,0.5,0.5,1.0,1.0,4.0
70+2018,baseline,privacy_preserving_computation,3098,0.1724,0.2198,0.0462,0.1756,0.0594,0.4032,3.255
71+2018,pre,privacy_preserving_computation,7,0.0,0.1429,0.2857,0.5714,1.0,0.8571,3.571
72+2018,baseline,programming_languages,4383,0.0418,0.1476,0.0331,0.141,0.102,0.5086,2.277
73+2018,post,programming_languages,1,0.0,0.0,0.0,1.0,1.0,1.0,5.0
74+2018,pre,programming_languages,6,0.0,0.3333,0.1667,0.6667,1.0,1.0,3.667
75+2018,baseline,quantum_computing,6187,0.2017,0.1474,0.0391,0.2022,0.0438,0.5495,3.594
76+2018,pre,quantum_computing,37,0.1081,0.1892,0.1081,0.4595,1.0,1.0,2.946
77+2018,baseline,reinforcement_learning,3289,0.1538,0.2195,0.09,0.145,0.2277,0.4865,3.212
78+2018,pre,reinforcement_learning,7,0.0,0.2857,0.0,0.2857,1.0,1.0,2.571
79+2018,pre,robotics,58,0.1034,0.569,0.0517,0.2586,1.0,0.931,3.793
80+2018,baseline,robotics,17437,0.1746,0.1812,0.0468,0.1681,0.0818,0.3658,3.657
81+2018,post,robotics,1,0.0,0.0,0.0,0.0,1.0,1.0,5.0
82+2018,pre,scientific_computing,467,0.1178,0.3405,0.0086,0.3426,1.0,0.9529,2.835
83+2018,baseline,scientific_computing,16414,0.226,0.1732,0.0344,0.2204,0.073,0.4828,3.055
84+2018,post,scientific_computing,1,0.0,1.0,0.0,1.0,1.0,1.0,3.0
85+2018,pre,software_engineering,16,0.125,0.3125,0.0,0.4375,1.0,1.0,5.5
86+2018,baseline,software_engineering,29000,0.0525,0.0828,0.0377,0.1001,0.0277,0.3541,2.471
87+2018,post,speech_audio,1,0.0,0.0,0.0,0.0,1.0,1.0,4.0
88+2018,baseline,speech_audio,12081,0.1153,0.0959,0.0561,0.1069,0.0699,0.351,3.024
89+2018,pre,speech_audio,14,0.2857,0.3571,0.1429,0.4286,1.0,0.9286,3.714
90+2018,pre,sustainable_computing,2,0.0,0.5,0.0,0.5,1.0,1.0,2.5
91+2018,baseline,sustainable_computing,581,0.1515,0.1549,0.062,0.2375,0.0224,0.4131,3.444
92+2019,baseline,ai_accelerators,992,0.2339,0.253,0.0665,0.2177,0.0938,0.3538,4.363
93+2019,baseline,ai_agents,5503,0.1703,0.1188,0.0305,0.1652,0.0836,0.4136,2.856
94+2019,post,ai_agents,2,0.0,0.0,0.0,0.0,1.0,1.0,2.0
95+2019,pre,ai_agents,1,0.0,0.0,0.0,0.0,1.0,0.0,1.0
96+2019,pre,ai_safety,1,0.0,0.0,0.0,0.0,1.0,1.0,1.0
97+2019,baseline,ai_safety,1478,0.0359,0.2659,0.0514,0.1238,0.1543,0.4926,2.565
98+2019,pre,algorithms,35,0.0571,0.2,0.0286,0.3714,1.0,0.9429,2.743
99+2019,baseline,algorithms,29204,0.1835,0.1542,0.031,0.2234,0.1227,0.5259,2.852
100+2019,post,algorithms,13,0.0,0.3846,0.0769,0.3846,1.0,1.0,3.385
101+2019,baseline,applied_cryptography,11352,0.1684,0.0847,0.0344,0.1373,0.0419,0.4066,2.885
102+2019,pre,applied_cryptography,3,0.0,0.0,0.0,0.0,1.0,1.0,3.0
103+2019,baseline,autonomous_scientific_discovery,114,0.0263,0.3333,0.0702,0.1842,0.114,0.5439,4.149
104+2019,baseline,blockchain,9045,0.0826,0.0871,0.0238,0.0905,0.0305,0.3787,2.545
105+2019,post,blockchain,2,0.0,0.0,0.0,0.0,1.0,1.0,2.5
106+2019,pre,cloud_computing,1,0.0,1.0,0.0,1.0,1.0,1.0,3.0
107+2019,baseline,cloud_computing,8755,0.136,0.0899,0.0474,0.1261,0.0162,0.3676,2.871
108+2019,baseline,compilers,689,0.0827,0.1872,0.0639,0.1176,0.0392,0.4673,2.955
109+2019,pre,computational_biology,1,0.0,1.0,0.0,0.0,1.0,1.0,5.0
110+2019,baseline,computational_biology,14875,0.1787,0.261,0.0529,0.2223,0.026,0.5882,4.68
111+2019,pre,computational_complexity,10,0.0,0.3,0.0,0.2,1.0,1.0,2.6
112+2019,baseline,computational_complexity,3941,0.1058,0.1373,0.018,0.1913,0.1056,0.5227,2.486
113+2019,post,computational_complexity,1,0.0,0.0,0.0,0.0,1.0,1.0,2.0
114+2019,post,computer_architecture,1,0.0,0.0,0.0,1.0,1.0,1.0,15.0
115+2019,baseline,computer_architecture,18844,0.1183,0.2029,0.0916,0.1516,0.0272,0.3022,3.567
116+2019,pre,computer_architecture,1,0.0,0.0,0.0,0.0,1.0,1.0,3.0
117+2019,pre,computer_graphics,2,0.5,1.0,0.0,0.5,1.0,1.0,4.0
118+2019,baseline,computer_graphics,9796,0.1134,0.1365,0.0475,0.1241,0.0481,0.4282,2.898
119+2019,baseline,computer_networks,58278,0.2277,0.1205,0.0616,0.2006,0.0655,0.3664,3.394
120+2019,pre,computer_networks,18,0.2778,0.3333,0.2222,0.3889,1.0,0.8333,3.111
121+2019,post,computer_networks,3,0.0,0.0,0.0,0.0,1.0,1.0,2.0
122+2019,baseline,computer_vision,81632,0.2773,0.1183,0.0604,0.1579,0.1216,0.415,3.712
123+2019,post,computer_vision,11,0.3636,0.0,0.2727,0.3636,1.0,1.0,7.182
124+2019,pre,computer_vision,27,0.2963,0.2963,0.0,0.2963,1.0,1.0,4.37
125+2019,baseline,cybersecurity,31551,0.103,0.128,0.0366,0.117,0.0553,0.4075,2.834
126+2019,pre,cybersecurity,3,0.0,0.3333,0.0,0.3333,1.0,1.0,5.0
127+2019,post,cybersecurity,7,0.0,0.4286,0.0,0.2857,1.0,1.0,3.714
128+2019,post,database_systems,1,1.0,0.0,0.0,0.0,1.0,1.0,4.0
129+2019,pre,database_systems,4,0.0,0.25,0.5,0.0,1.0,1.0,2.75
130+2019,baseline,database_systems,16574,0.1078,0.1104,0.0404,0.1234,0.0336,0.4532,2.934
131+2019,post,deep_learning_architectures,7,0.0,0.2857,0.1429,0.2857,1.0,1.0,3.0
132+2019,baseline,deep_learning_architectures,15350,0.2506,0.1516,0.0593,0.1745,0.1738,0.4552,3.556
133+2019,pre,deep_learning_architectures,6,0.1667,0.5,0.0,0.3333,1.0,1.0,3.167
134+2019,pre,distributed_computing,5,0.2,0.2,0.0,0.4,1.0,1.0,3.0
135+2019,baseline,distributed_computing,11487,0.1791,0.1977,0.0504,0.1726,0.0868,0.4341,3.368
136+2019,post,distributed_computing,3,0.0,0.0,0.0,0.3333,1.0,1.0,2.0
137+2019,baseline,edge_computing,6649,0.1396,0.1148,0.0486,0.1785,0.0203,0.3443,3.359
138+2019,post,edge_computing,1,0.0,0.0,0.0,0.0,1.0,1.0,3.0
139+2019,baseline,federated_learning,216,0.1806,0.25,0.0648,0.1713,0.0648,0.4907,3.472
140+2019,post,formal_verification,7,0.0,0.0,0.0,0.7143,1.0,0.8571,2.571
141+2019,baseline,formal_verification,4788,0.1005,0.1316,0.0357,0.2435,0.1976,0.5777,2.776
142+2019,pre,formal_verification,7,0.1429,0.2857,0.0,0.7143,1.0,1.0,2.571
143+2019,baseline,foundation_models,83,0.3012,0.2651,0.1205,0.2289,0.1928,0.4458,4.434
144+2019,baseline,generative_ai,1266,0.237,0.1943,0.0853,0.1469,0.2488,0.466,3.575
145+2019,post,hci,2,0.5,0.0,0.0,0.5,1.0,1.0,4.0
146+2019,baseline,hci,37208,0.0587,0.1417,0.0349,0.1171,0.0223,0.3994,3.066
147+2019,pre,hci,2,0.0,1.0,0.0,0.0,1.0,1.0,3.0
148+2019,post,information_retrieval,2,0.0,0.0,0.0,1.0,1.0,1.0,2.0
149+2019,baseline,information_retrieval,12298,0.1943,0.1383,0.0692,0.1511,0.0916,0.436,3.35
150+2019,pre,information_retrieval,4,0.5,0.25,0.25,0.25,1.0,1.0,4.25
151+2019,baseline,llm,1857,0.1341,0.0964,0.0598,0.0894,0.0549,0.3409,3.073
152+2019,baseline,ml_theory,17024,0.1535,0.2274,0.0773,0.1634,0.3046,0.5953,3.149
153+2019,pre,ml_theory,21,0.1905,0.4286,0.0952,0.4286,1.0,0.9524,3.143
154+2019,post,ml_theory,11,0.0909,0.3636,0.3636,0.3636,1.0,1.0,3.455
155+2019,baseline,multimodal_ai,632,0.394,0.1883,0.1108,0.1978,0.1883,0.4225,3.908
156+2019,baseline,nlp,25725,0.1328,0.1341,0.0681,0.1228,0.136,0.5435,3.154
157+2019,pre,nlp,2,0.0,0.0,0.5,0.0,1.0,1.0,2.5
158+2019,post,nlp,11,0.0,0.3636,0.0909,0.3636,1.0,1.0,3.364
159+2019,baseline,operating_systems,1578,0.1394,0.1584,0.0735,0.1255,0.0342,0.3878,2.999
160+2019,pre,operating_systems,1,0.0,1.0,1.0,0.0,1.0,1.0,5.0
161+2019,baseline,privacy_preserving_computation,3256,0.1895,0.1987,0.0501,0.1827,0.0934,0.4487,3.325
162+2019,pre,programming_languages,3,0.0,0.3333,0.0,0.3333,1.0,1.0,2.667
163+2019,baseline,programming_languages,4471,0.0512,0.1436,0.0282,0.1454,0.1118,0.5328,2.365
164+2019,pre,quantum_computing,6,0.1667,0.1667,0.1667,0.5,1.0,1.0,3.667
165+2019,baseline,quantum_computing,7473,0.195,0.1551,0.0541,0.2061,0.055,0.5535,3.755
166+2019,post,quantum_computing,3,1.0,0.0,0.0,0.3333,1.0,1.0,5.0
167+2019,baseline,reinforcement_learning,4735,0.1757,0.237,0.1012,0.1415,0.2769,0.5166,3.289
168+2019,pre,reinforcement_learning,3,0.0,0.3333,0.3333,0.6667,1.0,1.0,4.0
169+2019,post,reinforcement_learning,3,0.0,0.3333,0.3333,0.0,1.0,1.0,1.333
170+2019,post,robotics,1,0.0,0.0,0.0,0.0,1.0,1.0,2.0
171+2019,pre,robotics,9,0.1111,0.5556,0.1111,0.4444,1.0,0.6667,2.778
172+2019,baseline,robotics,20045,0.2113,0.1761,0.0529,0.1707,0.1106,0.3911,3.786
173+2019,post,scientific_computing,3,0.0,0.0,0.0,0.0,1.0,1.0,2.333
174+2019,pre,scientific_computing,102,0.1373,0.2353,0.0196,0.3529,1.0,0.951,2.598
175+2019,baseline,scientific_computing,18219,0.2475,0.1713,0.0354,0.2323,0.1239,0.5402,3.135
176+2019,pre,software_engineering,1,0.0,0.0,0.0,0.0,1.0,1.0,2.0
177+2019,post,software_engineering,4,0.0,0.0,0.0,0.25,1.0,1.0,3.75
178+2019,baseline,software_engineering,28930,0.0577,0.0777,0.037,0.1046,0.0358,0.3823,2.566
179+2019,baseline,speech_audio,12780,0.1381,0.1002,0.065,0.1126,0.0982,0.4074,3.1
180+2019,pre,speech_audio,3,0.0,0.3333,0.0,0.3333,1.0,1.0,1.667
181+2019,baseline,sustainable_computing,667,0.1664,0.1739,0.0405,0.1979,0.027,0.4018,3.261
182+2020,baseline,ai_accelerators,1085,0.2507,0.2406,0.0608,0.2461,0.1134,0.4101,4.532
183+2020,post,ai_agents,3,0.0,0.0,0.0,0.0,1.0,1.0,4.0
184+2020,baseline,ai_agents,5126,0.1851,0.1092,0.0308,0.1773,0.1178,0.4992,2.967
185+2020,post,ai_safety,1,0.0,0.0,0.0,0.0,1.0,1.0,2.0
186+2020,baseline,ai_safety,2153,0.0376,0.2606,0.0646,0.1319,0.1709,0.5625,2.7
187+2020,baseline,algorithms,29342,0.1856,0.1522,0.0313,0.2337,0.1347,0.554,2.888
188+2020,post,algorithms,23,0.0435,0.2174,0.0435,0.3913,1.0,1.0,2.913
189+2020,pre,algorithms,11,0.0,0.2727,0.0,0.1818,0.9091,1.0,2.636
190+2020,pre,applied_cryptography,1,1.0,0.0,0.0,0.0,1.0,1.0,1.0
191+2020,post,applied_cryptography,2,0.0,0.0,0.0,0.0,1.0,1.0,2.5
192+2020,baseline,applied_cryptography,11276,0.1642,0.0868,0.0438,0.1563,0.0468,0.4327,2.971
193+2020,baseline,autonomous_scientific_discovery,193,0.1088,0.3212,0.0933,0.1813,0.1606,0.6684,4.425
194+2020,baseline,blockchain,9688,0.1226,0.0824,0.0297,0.1361,0.0325,0.4047,2.794
195+2020,post,blockchain,3,0.0,0.0,0.0,0.0,1.0,1.0,4.333
196+2020,baseline,cloud_computing,8184,0.1477,0.0842,0.0442,0.1299,0.0208,0.3877,2.905
197+2020,baseline,compilers,737,0.0909,0.19,0.0529,0.1357,0.057,0.5427,3.033
198+2020,baseline,computational_biology,18078,0.1859,0.2428,0.0534,0.2199,0.0296,0.6409,5.128
199+2020,post,computational_complexity,6,0.0,0.3333,0.0,0.1667,1.0,1.0,3.167
200+2020,pre,computational_complexity,1,0.0,0.0,0.0,1.0,1.0,1.0,4.0
201+2020,baseline,computational_complexity,3865,0.1056,0.1415,0.0168,0.1946,0.1213,0.5679,2.566
202+2020,baseline,computer_architecture,17491,0.1377,0.1962,0.0932,0.1596,0.0414,0.3303,3.686
203+2020,post,computer_graphics,1,0.0,0.0,0.0,0.0,1.0,1.0,4.0
204+2020,baseline,computer_graphics,9444,0.1364,0.144,0.0497,0.142,0.0684,0.4842,2.916
205+2020,baseline,computer_networks,54382,0.2429,0.1209,0.0662,0.2249,0.0764,0.3882,3.482
206+2020,post,computer_networks,3,0.0,0.0,0.0,0.3333,1.0,1.0,3.0
207+2020,pre,computer_networks,4,0.0,0.25,0.25,0.5,1.0,1.0,3.0
208+2020,pre,computer_vision,4,0.5,0.0,0.0,0.25,1.0,1.0,3.75
209+2020,baseline,computer_vision,84014,0.2919,0.1198,0.0653,0.1652,0.1432,0.4669,3.822
210+2020,post,computer_vision,28,0.2857,0.25,0.1786,0.2143,1.0,1.0,4.464
211+2020,baseline,cybersecurity,32697,0.1191,0.13,0.0355,0.1253,0.0651,0.4388,2.897
212+2020,post,cybersecurity,8,0.0,0.375,0.125,0.375,1.0,1.0,4.375
213+2020,post,database_systems,2,0.0,0.0,0.0,0.0,1.0,1.0,10.0
214+2020,baseline,database_systems,15057,0.1216,0.1113,0.0414,0.1343,0.0434,0.495,3.057
215+2020,pre,database_systems,2,0.0,1.0,0.0,0.5,1.0,1.0,3.0
216+2020,pre,deep_learning_architectures,2,0.0,0.0,0.0,0.0,1.0,1.0,1.5
217+2020,baseline,deep_learning_architectures,17902,0.2731,0.1615,0.0678,0.1913,0.1986,0.5126,3.727
218+2020,post,deep_learning_architectures,5,0.0,0.4,0.2,0.4,1.0,1.0,3.0
219+2020,pre,distributed_computing,2,0.5,0.5,0.0,0.5,1.0,1.0,3.5
220+2020,post,distributed_computing,1,0.0,1.0,1.0,1.0,1.0,1.0,2.0
221+2020,baseline,distributed_computing,10898,0.194,0.1899,0.0584,0.1823,0.1012,0.4681,3.447
222+2020,post,edge_computing,1,0.0,0.0,0.0,0.0,1.0,1.0,3.0
223+2020,baseline,edge_computing,7114,0.1751,0.107,0.0447,0.211,0.0283,0.3663,3.436
224+2020,baseline,federated_learning,304,0.2072,0.2072,0.0757,0.2434,0.1743,0.5461,3.905
225+2020,pre,formal_verification,2,0.5,0.0,0.0,0.5,1.0,1.0,4.5
226+2020,baseline,formal_verification,4530,0.1044,0.1415,0.0391,0.243,0.2243,0.6351,2.859
227+2020,post,formal_verification,3,0.0,0.0,0.0,0.6667,1.0,1.0,2.333
228+2020,baseline,foundation_models,137,0.3139,0.219,0.073,0.146,0.2993,0.5255,3.803
229+2020,baseline,generative_ai,1498,0.245,0.1889,0.0794,0.1462,0.2036,0.4653,3.633
230+2020,baseline,hci,36247,0.063,0.1427,0.037,0.1265,0.0322,0.4348,3.126
231+2020,post,hci,4,0.0,0.25,0.0,0.25,1.0,1.0,2.0
232+2020,post,information_retrieval,3,0.0,0.0,0.0,0.0,1.0,1.0,4.333
233+2020,baseline,information_retrieval,11774,0.214,0.1395,0.0854,0.1618,0.1181,0.4676,3.247
234+2020,baseline,llm,2461,0.1451,0.1121,0.0813,0.1077,0.0947,0.4275,3.242
235+2020,baseline,ml_theory,19292,0.1493,0.2407,0.0809,0.1813,0.3396,0.6305,3.221
236+2020,pre,ml_theory,3,0.0,0.0,0.3333,0.6667,1.0,1.0,3.667
237+2020,post,ml_theory,14,0.0,0.2143,0.0,0.1429,1.0,1.0,3.214
238+2020,baseline,multimodal_ai,755,0.4053,0.2079,0.1311,0.2291,0.1841,0.4278,4.013
239+2020,post,nlp,23,0.1739,0.1739,0.2609,0.3043,1.0,1.0,4.13
240+2020,baseline,nlp,28724,0.1429,0.1459,0.0794,0.1374,0.1809,0.5686,3.294
241+2020,baseline,operating_systems,1454,0.1534,0.1589,0.0722,0.1437,0.0303,0.3982,3.071
242+2020,post,privacy_preserving_computation,2,0.0,0.0,0.5,0.5,1.0,1.0,2.0
243+2020,baseline,privacy_preserving_computation,4155,0.1641,0.2113,0.0602,0.194,0.1232,0.5083,3.463
244+2020,post,programming_languages,3,0.0,0.0,0.0,0.3333,1.0,1.0,4.667
245+2020,baseline,programming_languages,4123,0.0589,0.155,0.0303,0.146,0.1179,0.5511,2.439
246+2020,pre,quantum_computing,3,0.0,0.6667,0.6667,0.6667,1.0,1.0,2.0
247+2020,baseline,quantum_computing,8096,0.1902,0.1784,0.0615,0.2134,0.0725,0.5983,3.697
248+2020,baseline,reinforcement_learning,5674,0.1882,0.2652,0.1038,0.1768,0.2978,0.5583,3.438
249+2020,post,reinforcement_learning,1,0.0,0.0,0.0,0.0,1.0,1.0,2.0
250+2020,post,robotics,2,0.0,0.5,0.0,0.5,1.0,1.0,2.0
251+2020,pre,robotics,5,0.2,0.8,0.4,0.4,1.0,0.8,4.0
252+2020,baseline,robotics,20346,0.1966,0.2101,0.0598,0.1794,0.1495,0.4674,3.798
253+2020,baseline,scientific_computing,19719,0.2425,0.1752,0.0312,0.2442,0.1673,0.5889,3.243
254+2020,post,scientific_computing,5,0.2,0.4,0.0,0.8,1.0,1.0,2.8
255+2020,pre,scientific_computing,16,0.0,0.125,0.0,0.375,1.0,0.9375,2.188
256+2020,post,software_engineering,1,0.0,1.0,0.0,0.0,1.0,1.0,3.0
257+2020,pre,software_engineering,1,0.0,0.0,1.0,1.0,1.0,1.0,3.0
258+2020,baseline,software_engineering,27754,0.0673,0.0862,0.0382,0.1131,0.0464,0.4258,2.662
259+2020,baseline,speech_audio,12490,0.1448,0.1029,0.0749,0.1233,0.1324,0.4532,3.205
260+2020,post,speech_audio,3,0.0,0.3333,0.3333,0.3333,1.0,1.0,2.667
261+2020,baseline,sustainable_computing,634,0.2145,0.1562,0.0568,0.2618,0.0505,0.4811,3.554
262+2021,baseline,ai_accelerators,1249,0.2858,0.2314,0.0769,0.2346,0.1153,0.4139,4.869
263+2021,post,ai_agents,2,0.0,0.5,0.0,0.5,1.0,1.0,3.0
264+2021,baseline,ai_agents,4890,0.1945,0.1415,0.0315,0.192,0.1247,0.5284,3.1
265+2021,baseline,ai_safety,2493,0.0461,0.2944,0.0542,0.1777,0.2166,0.6193,2.908
266+2021,pre,ai_safety,1,0.0,1.0,1.0,1.0,1.0,1.0,6.0
267+2021,baseline,algorithms,28971,0.2026,0.1532,0.0315,0.237,0.143,0.5746,2.96
268+2021,pre,algorithms,5,0.0,0.4,0.0,0.2,0.6,1.0,2.0
269+2021,post,algorithms,47,0.0,0.3617,0.0426,0.3191,0.9362,0.9362,3.106
270+2021,baseline,applied_cryptography,10920,0.1968,0.0904,0.0473,0.1662,0.0542,0.4442,3.1
271+2021,post,applied_cryptography,4,0.0,0.25,0.0,0.25,0.5,1.0,3.25
272+2021,baseline,autonomous_scientific_discovery,221,0.0588,0.3258,0.0814,0.1674,0.2443,0.7104,4.643
273+2021,baseline,blockchain,10644,0.1487,0.0821,0.0336,0.1458,0.0316,0.4334,2.946
274+2021,post,blockchain,3,0.3333,0.3333,0.0,0.3333,0.6667,1.0,5.333
275+2021,baseline,cloud_computing,7735,0.1694,0.0829,0.0512,0.1396,0.0244,0.4105,2.996
276+2021,baseline,compilers,709,0.1086,0.2116,0.0917,0.1636,0.0465,0.5416,3.169
277+2021,baseline,computational_biology,19218,0.2026,0.2293,0.0558,0.2368,0.0312,0.6674,5.123
278+2021,baseline,computational_complexity,3753,0.1159,0.1367,0.0189,0.1937,0.121,0.5838,2.678
279+2021,post,computational_complexity,10,0.0,0.3,0.0,0.0,0.9,0.9,2.9
280+2021,pre,computational_complexity,2,0.0,0.5,0.0,0.5,1.0,1.0,3.0
281+2021,post,computer_architecture,4,0.0,0.5,0.0,0.5,1.0,1.0,4.75
282+2021,baseline,computer_architecture,16396,0.156,0.1994,0.1045,0.1649,0.0504,0.3453,3.86
283+2021,baseline,computer_graphics,8966,0.1694,0.137,0.0552,0.1374,0.0751,0.5152,3.031
284+2021,post,computer_graphics,3,0.0,0.0,0.0,0.3333,0.6667,0.6667,2.667
285+2021,baseline,computer_networks,51635,0.2651,0.1166,0.0781,0.2306,0.083,0.4144,3.575
286+2021,pre,computer_networks,1,0.0,0.0,0.0,0.0,1.0,1.0,2.0
287+2021,post,computer_networks,8,0.125,0.0,0.125,0.25,1.0,1.0,4.125
288+2021,post,computer_vision,53,0.1887,0.1132,0.0377,0.1887,0.9245,0.9434,4.151
289+2021,baseline,computer_vision,93042,0.3358,0.1179,0.0736,0.1712,0.1556,0.4797,4.015
290+2021,pre,computer_vision,1,0.0,1.0,0.0,1.0,1.0,1.0,3.0
291+2021,baseline,cybersecurity,31770,0.138,0.1298,0.0413,0.1408,0.0716,0.4731,3.092
292+2021,post,cybersecurity,9,0.0,0.0,0.1111,0.1111,1.0,1.0,4.333
293+2021,baseline,database_systems,14331,0.1502,0.1043,0.0518,0.1423,0.0498,0.5393,3.296
294+2021,post,database_systems,1,0.0,0.0,0.0,1.0,1.0,1.0,5.0
295+2021,post,deep_learning_architectures,12,0.0833,0.1667,0.1667,0.1667,1.0,1.0,3.75
296+2021,pre,deep_learning_architectures,1,1.0,0.0,0.0,0.0,1.0,0.0,3.0
297+2021,baseline,deep_learning_architectures,21211,0.3058,0.1691,0.0764,0.2017,0.2191,0.5294,3.794
298+2021,baseline,distributed_computing,10088,0.2127,0.191,0.0643,0.1884,0.1107,0.4908,3.54
299+2021,pre,distributed_computing,1,0.0,1.0,0.0,0.0,0.0,1.0,3.0
300+2021,post,distributed_computing,7,0.1429,0.2857,0.0,0.5714,1.0,1.0,3.143
301+2021,baseline,edge_computing,7910,0.1958,0.0952,0.0497,0.2155,0.0279,0.4023,3.519
302+2021,post,federated_learning,1,0.0,0.0,0.0,0.0,1.0,1.0,4.0
303+2021,baseline,federated_learning,650,0.2677,0.2554,0.1092,0.2215,0.2385,0.5092,3.963
304+2021,post,formal_verification,4,0.0,0.0,0.0,0.5,1.0,1.0,2.5
305+2021,baseline,formal_verification,4507,0.1041,0.1378,0.0375,0.2443,0.2403,0.6357,2.842
306+2021,baseline,foundation_models,249,0.3775,0.241,0.1084,0.2209,0.3293,0.5301,4.418
307+2021,baseline,generative_ai,1959,0.2838,0.1868,0.0812,0.1741,0.2006,0.4609,3.646
308+2021,post,generative_ai,1,0.0,0.0,0.0,1.0,1.0,1.0,4.0
309+2021,post,hci,10,0.1,0.0,0.0,0.5,1.0,1.0,3.9
310+2021,baseline,hci,34419,0.0893,0.151,0.0371,0.1405,0.039,0.4914,3.338
311+2021,baseline,information_retrieval,11536,0.2499,0.1328,0.0911,0.1738,0.1351,0.4956,3.407
312+2021,post,information_retrieval,7,0.1429,0.0,0.2857,0.0,1.0,1.0,2.714
313+2021,post,llm,2,0.5,0.0,0.5,0.0,1.0,1.0,5.0
314+2021,baseline,llm,2847,0.209,0.1507,0.1117,0.1236,0.1324,0.4939,3.597
315+2021,pre,ml_theory,2,0.0,0.5,0.0,1.0,1.0,1.0,2.0
316+2021,baseline,ml_theory,19943,0.1689,0.2307,0.0824,0.1926,0.3116,0.6341,3.358
317+2021,post,ml_theory,11,0.1818,0.2727,0.0909,0.2727,1.0,1.0,3.091
318+2021,baseline,multimodal_ai,946,0.4693,0.1744,0.1332,0.2452,0.185,0.4313,4.196
319+2021,post,multimodal_ai,2,0.0,0.5,0.5,0.0,1.0,1.0,1.5
320+2021,baseline,nlp,29555,0.1756,0.1559,0.0914,0.1505,0.1871,0.6122,3.443
321+2021,post,nlp,38,0.1053,0.2632,0.1579,0.2105,0.9474,1.0,4.289
322+2021,post,operating_systems,1,0.0,0.0,0.0,0.0,0.0,0.0,1.0
323+2021,baseline,operating_systems,1322,0.18,0.1445,0.0983,0.1377,0.0272,0.4584,3.332
324+2021,post,privacy_preserving_computation,1,0.0,0.0,0.0,0.0,1.0,1.0,4.0
325+2021,baseline,privacy_preserving_computation,4240,0.1995,0.216,0.0726,0.2078,0.1363,0.5389,3.705
326+2021,baseline,programming_languages,3972,0.0622,0.1433,0.0305,0.1568,0.1365,0.5997,2.489
327+2021,post,programming_languages,6,0.0,0.0,0.0,0.1667,1.0,1.0,1.5
328+2021,baseline,quantum_computing,9273,0.1785,0.1857,0.0771,0.2061,0.0764,0.626,3.997
329+2021,pre,quantum_computing,1,0.0,0.0,0.0,0.0,1.0,1.0,2.0
330+2021,post,quantum_computing,6,0.0,0.3333,0.0,0.1667,1.0,1.0,2.333
331+2021,baseline,reinforcement_learning,6587,0.2408,0.269,0.1084,0.1861,0.3095,0.5737,3.6
332+2021,post,reinforcement_learning,5,0.0,0.0,0.2,0.0,0.8,1.0,3.0
333+2021,baseline,robotics,22059,0.2466,0.1899,0.0596,0.1815,0.1688,0.4756,3.924
334+2021,post,robotics,7,0.0,0.2857,0.0,0.0,1.0,1.0,4.143
335+2021,pre,scientific_computing,2,0.0,0.5,0.0,0.5,1.0,1.0,2.5
336+2021,post,scientific_computing,8,0.0,0.0,0.0,0.25,0.875,1.0,2.375
337+2021,baseline,scientific_computing,20289,0.2473,0.171,0.0344,0.2401,0.1854,0.5948,3.247
338+2021,post,software_engineering,7,0.1429,0.2857,0.1429,0.2857,0.8571,0.8571,4.143
339+2021,baseline,software_engineering,24664,0.0827,0.0936,0.047,0.1295,0.0722,0.4869,2.894
340+2021,baseline,speech_audio,12783,0.1848,0.1181,0.0952,0.1327,0.1519,0.476,3.468
341+2021,post,speech_audio,4,0.5,0.25,0.0,0.5,1.0,1.0,4.75
342+2021,baseline,sustainable_computing,717,0.2218,0.1855,0.0502,0.2427,0.0516,0.5202,3.661
343+2022,baseline,ai_accelerators,1341,0.2677,0.2424,0.0895,0.2528,0.1111,0.4534,5.238
344+2022,post,ai_accelerators,80,0.4875,0.225,0.0625,0.225,0.0125,0.4375,4.888
345+2022,baseline,ai_agents,4131,0.252,0.1268,0.0315,0.1915,0.1365,0.5556,3.269
346+2022,post,ai_agents,390,0.2436,0.0821,0.0308,0.1513,0.1385,0.6051,3.072
347+2022,post,ai_safety,205,0.0634,0.2585,0.0293,0.1561,0.2,0.6146,3.127
348+2022,baseline,ai_safety,2749,0.0531,0.2699,0.0691,0.1848,0.1961,0.6501,3.057
349+2022,post,algorithms,2237,0.2713,0.1203,0.0246,0.2083,0.0997,0.5431,3.077
350+2022,pre,algorithms,2,0.0,0.0,0.0,0.5,1.0,1.0,2.0
351+2022,baseline,algorithms,25963,0.2425,0.1585,0.0349,0.2437,0.1415,0.583,3.061
352+2022,baseline,applied_cryptography,9409,0.2389,0.0983,0.0498,0.186,0.0543,0.4911,3.237
353+2022,post,applied_cryptography,890,0.2742,0.0539,0.0461,0.1371,0.0371,0.5146,3.217
354+2022,baseline,autonomous_scientific_discovery,215,0.1256,0.3767,0.093,0.2093,0.1953,0.6512,4.126
355+2022,post,autonomous_scientific_discovery,21,0.0952,0.381,0.0476,0.1905,0.2857,0.7143,3.905
356+2022,post,blockchain,1169,0.1814,0.071,0.0359,0.1454,0.0257,0.4953,3.163
357+2022,baseline,blockchain,10601,0.1663,0.0874,0.0326,0.1647,0.0322,0.4817,3.052
358+2022,baseline,cloud_computing,6306,0.1917,0.0864,0.0584,0.143,0.0251,0.4565,3.124
359+2022,post,cloud_computing,797,0.1982,0.079,0.0615,0.1129,0.0163,0.4291,3.031
360+2022,post,compilers,58,0.0517,0.0345,0.0172,0.1207,0.0,0.7069,3.397
361+2022,baseline,compilers,584,0.0993,0.2072,0.0822,0.1781,0.0599,0.5976,3.433
362+2022,post,computational_biology,2018,0.2914,0.2279,0.052,0.225,0.0238,0.5956,5.172
363+2022,baseline,computational_biology,17518,0.2238,0.2211,0.0528,0.2374,0.0347,0.6935,5.352
364+2022,baseline,computational_complexity,3266,0.131,0.1562,0.0233,0.214,0.1191,0.5842,2.755
365+2022,post,computational_complexity,275,0.1309,0.1127,0.0145,0.1673,0.1382,0.6436,2.665
366+2022,post,computer_architecture,1089,0.214,0.1579,0.0817,0.1524,0.045,0.3636,3.834
367+2022,baseline,computer_architecture,14085,0.1847,0.206,0.1199,0.1693,0.0552,0.3663,4.099
368+2022,post,computer_graphics,778,0.1799,0.1015,0.063,0.1054,0.0758,0.5938,3.085
369+2022,baseline,computer_graphics,7509,0.1963,0.1472,0.0587,0.1429,0.0993,0.5683,3.186
370+2022,pre,computer_networks,1,1.0,0.0,0.0,0.0,1.0,0.0,3.0
371+2022,baseline,computer_networks,44215,0.2916,0.1113,0.0822,0.2366,0.0894,0.4315,3.75
372+2022,post,computer_networks,4925,0.317,0.0993,0.0828,0.2337,0.0404,0.3789,3.787
373+2022,post,computer_vision,8585,0.3726,0.0874,0.0515,0.1449,0.1123,0.4953,4.089
374+2022,baseline,computer_vision,90119,0.3793,0.1063,0.0745,0.1753,0.1554,0.5056,4.18
375+2022,post,cybersecurity,3238,0.1868,0.1035,0.0389,0.1242,0.0537,0.491,3.283
376+2022,baseline,cybersecurity,28316,0.157,0.1322,0.0463,0.1548,0.0784,0.513,3.24
377+2022,baseline,database_systems,12221,0.1795,0.0996,0.0493,0.15,0.0489,0.5964,3.379
378+2022,post,database_systems,1404,0.1538,0.0926,0.0513,0.1132,0.0264,0.589,3.162
379+2022,baseline,deep_learning_architectures,20266,0.3449,0.1523,0.0673,0.2006,0.2212,0.5569,3.934
380+2022,post,deep_learning_architectures,2004,0.3468,0.1208,0.0509,0.1477,0.1816,0.5334,3.88
381+2022,post,distributed_computing,855,0.3111,0.1193,0.0421,0.1509,0.076,0.4538,3.591
382+2022,baseline,distributed_computing,8524,0.2475,0.1763,0.0644,0.1953,0.1098,0.5016,3.634
383+2022,baseline,edge_computing,7666,0.2224,0.0892,0.0536,0.2237,0.0256,0.4126,3.72
384+2022,post,edge_computing,888,0.2252,0.0856,0.0417,0.2005,0.0158,0.3784,3.67
385+2022,baseline,federated_learning,880,0.3227,0.2193,0.1068,0.2568,0.2159,0.5227,4.658
386+2022,post,federated_learning,94,0.4149,0.1489,0.0426,0.2234,0.1489,0.4362,3.979
387+2022,baseline,formal_verification,3703,0.118,0.1431,0.0416,0.2684,0.2609,0.6603,3.012
388+2022,post,formal_verification,292,0.161,0.1096,0.0479,0.1952,0.1781,0.6164,3.065
389+2022,baseline,foundation_models,313,0.3674,0.2013,0.099,0.1661,0.4026,0.6006,4.629
390+2022,post,foundation_models,37,0.6216,0.0811,0.1622,0.2162,0.2162,0.6216,5.297
391+2022,post,generative_ai,208,0.2788,0.101,0.0529,0.1442,0.1394,0.4663,3.659
392+2022,baseline,generative_ai,1952,0.2874,0.144,0.0625,0.1773,0.1901,0.5195,3.781
393+2022,post,hci,2949,0.1061,0.0953,0.0173,0.1261,0.0264,0.6355,3.263
394+2022,baseline,hci,31852,0.0967,0.1548,0.0376,0.1461,0.0451,0.5358,3.491
395+2022,post,information_retrieval,908,0.2841,0.0837,0.0518,0.1355,0.0826,0.5143,3.331
396+2022,baseline,information_retrieval,10335,0.3016,0.1272,0.1087,0.1722,0.1533,0.5442,3.553
397+2022,baseline,llm,2943,0.2314,0.1699,0.1223,0.1359,0.208,0.5963,3.879
398+2022,post,llm,323,0.2322,0.1858,0.096,0.1331,0.2601,0.6347,3.96
399+2022,post,ml_theory,1918,0.1794,0.1741,0.0553,0.1496,0.2242,0.5782,3.395
400+2022,baseline,ml_theory,18213,0.1898,0.2007,0.0703,0.1872,0.2975,0.6562,3.479
401+2022,post,multimodal_ai,97,0.4845,0.1649,0.0928,0.2577,0.2165,0.4948,4.567
402+2022,baseline,multimodal_ai,1181,0.5258,0.1304,0.1431,0.1846,0.2134,0.4674,4.456
403+2022,pre,nlp,1,0.0,0.0,0.0,0.0,1.0,1.0,4.0
404+2022,baseline,nlp,26840,0.2025,0.151,0.0918,0.1623,0.1914,0.6484,3.672
405+2022,post,nlp,2837,0.203,0.1146,0.0663,0.1445,0.1981,0.6105,3.477
406+2022,baseline,operating_systems,965,0.2093,0.143,0.0777,0.1565,0.0321,0.513,3.456
407+2022,post,operating_systems,118,0.1356,0.1186,0.0339,0.1017,0.0254,0.5593,2.992
408+2022,baseline,privacy_preserving_computation,3940,0.2246,0.2152,0.0756,0.2211,0.1561,0.5817,3.769
409+2022,post,privacy_preserving_computation,436,0.3716,0.1491,0.0688,0.1881,0.1124,0.4472,3.706
410+2022,post,programming_languages,288,0.0903,0.0417,0.0208,0.1042,0.0972,0.7361,2.688
411+2022,baseline,programming_languages,3376,0.0705,0.1605,0.0427,0.1804,0.1348,0.6099,2.609
412+2022,post,quantum_computing,799,0.2115,0.199,0.0876,0.2278,0.0713,0.7134,3.946
413+2022,baseline,quantum_computing,8417,0.2053,0.2223,0.0985,0.2231,0.0926,0.672,3.959
414+2022,post,reinforcement_learning,666,0.3063,0.2042,0.0841,0.1682,0.2553,0.5225,3.728
415+2022,baseline,reinforcement_learning,6351,0.2903,0.2121,0.0931,0.1735,0.2907,0.5579,3.733
416+2022,baseline,robotics,21166,0.2672,0.1764,0.06,0.1801,0.1875,0.4943,4.074
417+2022,post,robotics,1937,0.4177,0.1239,0.0413,0.1791,0.1136,0.4399,4.08
418+2022,post,scientific_computing,1760,0.2767,0.1517,0.029,0.225,0.1261,0.5608,3.417
419+2022,pre,scientific_computing,2,0.5,0.0,0.0,0.0,1.0,1.0,1.0
420+2022,baseline,scientific_computing,18170,0.2804,0.1718,0.0369,0.251,0.2024,0.6222,3.534
421+2022,baseline,software_engineering,21619,0.0929,0.0994,0.055,0.1359,0.0741,0.5436,3.017
422+2022,post,software_engineering,2106,0.1187,0.0655,0.0418,0.0959,0.0465,0.5793,2.952
423+2022,post,speech_audio,1078,0.2746,0.0724,0.0529,0.1197,0.0891,0.551,3.432
424+2022,baseline,speech_audio,11620,0.2333,0.1095,0.0998,0.1434,0.1675,0.5202,3.643
425+2022,post,sustainable_computing,78,0.2308,0.141,0.0641,0.2308,0.0769,0.6538,3.59
426+2022,baseline,sustainable_computing,752,0.246,0.1755,0.0585,0.2447,0.0465,0.4973,3.914
427+2023,baseline,ai_accelerators,32,0.1875,0.4688,0.1875,0.3125,0.7188,0.8125,5.5
428+2023,post,ai_accelerators,1567,0.3101,0.2246,0.0862,0.238,0.0893,0.4416,5.294
429+2023,post,ai_agents,4912,0.2673,0.115,0.0364,0.1861,0.1423,0.5867,3.342
430+2023,baseline,ai_agents,79,0.3038,0.3671,0.1266,0.3924,0.8228,0.8987,4.114
431+2023,post,ai_safety,3960,0.0548,0.2593,0.0606,0.179,0.2225,0.7081,3.08
432+2023,baseline,ai_safety,55,0.0727,0.4727,0.0909,0.3273,0.8,0.9455,4.273
433+2023,pre,algorithms,1,0.0,0.0,0.0,0.0,1.0,1.0,1.0
434+2023,post,algorithms,27335,0.2562,0.1418,0.0332,0.2338,0.1265,0.6019,3.12
435+2023,baseline,algorithms,1195,0.0954,0.3247,0.0444,0.4209,0.6126,0.8435,3.09
436+2023,baseline,applied_cryptography,131,0.2366,0.1985,0.0916,0.3359,0.7405,0.8855,3.206
437+2023,post,applied_cryptography,11087,0.2385,0.0922,0.0515,0.174,0.0486,0.5315,3.256
438+2023,pre,applied_cryptography,1,0.0,0.0,0.0,0.0,1.0,1.0,2.0
439+2023,baseline,autonomous_scientific_discovery,12,0.0,0.5,0.0833,0.25,0.75,1.0,4.25
440+2023,post,autonomous_scientific_discovery,388,0.1005,0.3325,0.0696,0.2088,0.2655,0.75,4.809
441+2023,post,blockchain,14260,0.1501,0.0745,0.0304,0.1508,0.0262,0.5109,3.089
442+2023,baseline,blockchain,89,0.1573,0.2584,0.0674,0.4382,0.4045,0.7303,3.584
443+2023,post,cloud_computing,7616,0.1756,0.0865,0.0685,0.1292,0.0213,0.5169,3.11
444+2023,baseline,cloud_computing,27,0.2593,0.1852,0.0741,0.3333,0.3704,0.6667,4.111
445+2023,baseline,compilers,5,0.2,0.6,0.2,0.6,0.6,0.8,3.8
446+2023,post,compilers,687,0.131,0.1674,0.0684,0.1368,0.0509,0.6769,3.48
447+2023,baseline,computational_biology,175,0.1314,0.3886,0.1257,0.3657,0.4629,0.9029,5.869
448+2023,post,computational_biology,20401,0.2255,0.2021,0.051,0.2295,0.0383,0.6763,5.322
449+2023,baseline,computational_complexity,81,0.037,0.2593,0.0123,0.3333,0.6667,0.9259,2.247
450+2023,post,computational_complexity,3418,0.1451,0.1243,0.0237,0.2039,0.1211,0.6085,2.907
451+2023,baseline,computer_architecture,101,0.1584,0.4257,0.1881,0.3267,0.7426,0.8614,5.366
452+2023,post,computer_architecture,16268,0.2064,0.1934,0.1127,0.1679,0.0562,0.4078,4.182
453+2023,pre,computer_graphics,1,1.0,0.0,0.0,1.0,1.0,1.0,4.0
454+2023,post,computer_graphics,9201,0.2016,0.1387,0.0619,0.1362,0.1048,0.6405,3.25
455+2023,baseline,computer_graphics,95,0.2421,0.2947,0.1368,0.3684,0.5263,0.7684,4.126
456+2023,post,computer_networks,51205,0.305,0.1044,0.0872,0.2313,0.0719,0.4381,3.854
457+2023,baseline,computer_networks,676,0.2988,0.2396,0.1272,0.4911,0.8121,0.895,4.087
458+2023,baseline,computer_vision,2547,0.36,0.256,0.1841,0.3286,0.8041,0.883,5.142
459+2023,post,computer_vision,111714,0.3907,0.095,0.0672,0.164,0.1516,0.522,4.276
460+2023,baseline,cybersecurity,368,0.1929,0.3478,0.106,0.3723,0.7853,0.894,4.315
461+2023,post,cybersecurity,36113,0.1543,0.1243,0.0435,0.1435,0.0674,0.5534,3.26
462+2023,baseline,database_systems,95,0.1789,0.2842,0.1474,0.4526,0.8,0.9474,4.695
463+2023,post,database_systems,14392,0.1695,0.0933,0.0538,0.1336,0.0456,0.6591,3.392
464+2023,baseline,deep_learning_architectures,629,0.2814,0.3084,0.1113,0.3037,0.8124,0.9094,4.161
465+2023,post,deep_learning_architectures,24978,0.3643,0.1375,0.068,0.1931,0.1961,0.5627,3.985
466+2023,post,distributed_computing,9664,0.2739,0.1739,0.0691,0.1911,0.0972,0.5465,3.73
467+2023,baseline,distributed_computing,134,0.1567,0.3955,0.1269,0.3881,0.7239,0.8806,4.94
468+2023,post,edge_computing,9080,0.2148,0.0878,0.051,0.2214,0.0246,0.43,3.721
469+2023,baseline,edge_computing,40,0.25,0.4,0.1,0.475,0.7,0.825,5.125
470+2023,post,federated_learning,1467,0.3565,0.182,0.0832,0.2522,0.1677,0.5058,4.235
471+2023,baseline,federated_learning,30,0.4667,0.2667,0.0667,0.5333,0.7667,0.8333,4.867
472+2023,baseline,formal_verification,137,0.0511,0.2774,0.0438,0.4088,0.7664,0.9489,3.051
473+2023,post,formal_verification,4248,0.1236,0.137,0.0391,0.2651,0.2651,0.6883,2.972
474+2023,post,foundation_models,563,0.3783,0.1829,0.0853,0.1741,0.444,0.6643,5.025
475+2023,baseline,foundation_models,8,0.25,0.25,0.125,0.375,0.625,0.625,4.0
476+2023,baseline,generative_ai,49,0.1429,0.3061,0.1633,0.2653,0.7143,0.7755,3.796
477+2023,post,generative_ai,3597,0.2616,0.1507,0.0556,0.1529,0.1724,0.5635,3.571
478+2023,post,hci,40423,0.0994,0.1392,0.0352,0.1424,0.0483,0.6369,3.489
479+2023,baseline,hci,164,0.1037,0.3598,0.0915,0.3476,0.5793,0.8902,4.622
480+2023,post,information_retrieval,12421,0.3156,0.1169,0.1065,0.1647,0.1479,0.6093,3.601
481+2023,baseline,information_retrieval,147,0.3673,0.2925,0.2585,0.3673,0.7959,0.9048,5.224
482+2023,post,llm,10332,0.1815,0.2002,0.0938,0.1495,0.2537,0.7129,3.78
483+2023,baseline,llm,57,0.3158,0.4737,0.2982,0.2982,0.7018,0.8947,12.579
484+2023,baseline,ml_theory,733,0.1514,0.3779,0.1091,0.3302,0.7422,0.9018,3.626
485+2023,post,ml_theory,22160,0.1991,0.1787,0.0602,0.1764,0.2568,0.6606,3.526
486+2023,baseline,multimodal_ai,18,0.4444,0.2222,0.1667,0.1111,0.7222,0.9444,5.5
487+2023,post,multimodal_ai,1914,0.5486,0.1332,0.1254,0.187,0.2665,0.5946,4.753
488+2023,post,nlp,33400,0.2043,0.1387,0.0764,0.1607,0.2012,0.6816,3.664
489+2023,baseline,nlp,565,0.2142,0.3504,0.2265,0.2991,0.6708,0.9593,5.228
490+2023,baseline,operating_systems,3,0.6667,0.0,0.3333,0.0,1.0,1.0,6.0
491+2023,post,operating_systems,1170,0.2308,0.1521,0.0897,0.141,0.0504,0.5718,3.696
492+2023,baseline,privacy_preserving_computation,104,0.1923,0.4231,0.1442,0.5096,0.7115,0.9231,4.394
493+2023,post,privacy_preserving_computation,4906,0.2662,0.1936,0.0671,0.2224,0.1449,0.5946,3.818
494+2023,post,programming_languages,3822,0.0801,0.1308,0.0322,0.1533,0.1227,0.6669,2.567
495+2023,baseline,programming_languages,54,0.037,0.3519,0.0556,0.3704,0.7037,0.9444,3.593
496+2023,post,quantum_computing,10237,0.1889,0.2151,0.0962,0.2131,0.0795,0.6837,4.082
497+2023,baseline,quantum_computing,307,0.1629,0.316,0.1531,0.3974,0.4235,0.9283,4.798
498+2023,post,reinforcement_learning,8203,0.3344,0.1872,0.0745,0.1681,0.2604,0.5574,3.832
499+2023,baseline,reinforcement_learning,219,0.1644,0.3744,0.137,0.2877,0.79,0.8767,3.699
500+2023,baseline,robotics,505,0.1505,0.4376,0.1168,0.2653,0.8733,0.9267,4.364
501+2023,post,robotics,25254,0.3114,0.1607,0.0569,0.1764,0.1845,0.5093,4.152
502+2023,post,scientific_computing,19480,0.2953,0.1676,0.0401,0.2462,0.1657,0.6018,3.429
503+2023,pre,scientific_computing,3,0.0,0.0,0.0,0.6667,1.0,1.0,2.0
504+2023,baseline,scientific_computing,940,0.1628,0.334,0.0245,0.3543,0.834,0.9489,3.152
505+2023,baseline,software_engineering,187,0.1711,0.1872,0.1337,0.4225,0.5348,0.8235,4.294
506+2023,post,software_engineering,26058,0.0995,0.0924,0.0499,0.1292,0.0722,0.6213,3.019
507+2023,baseline,speech_audio,355,0.1972,0.3577,0.338,0.2958,0.8254,0.8958,4.766
508+2023,post,speech_audio,14333,0.2172,0.0996,0.0837,0.1322,0.1704,0.5598,3.626
509+2023,post,sustainable_computing,1073,0.2432,0.1622,0.0801,0.2209,0.0718,0.5778,3.954
510+2023,baseline,sustainable_computing,5,0.0,0.6,0.2,0.0,1.0,1.0,3.8
511+2024,baseline,ai_accelerators,5,0.0,0.2,0.2,0.4,1.0,1.0,5.8
512+2024,post,ai_accelerators,1715,0.3376,0.193,0.0746,0.2117,0.1073,0.4239,5.27
513+2024,post,ai_agents,5888,0.2661,0.1165,0.0343,0.1619,0.1686,0.5922,3.481
514+2024,baseline,ai_agents,11,0.3636,0.1818,0.0909,0.2727,0.9091,1.0,4.091
515+2024,baseline,ai_safety,16,0.1875,0.5625,0.125,0.25,0.625,0.75,3.75
516+2024,post,ai_safety,6201,0.0555,0.2348,0.0437,0.168,0.2092,0.7002,3.166
517+2024,baseline,algorithms,301,0.1063,0.3023,0.0399,0.4219,0.6113,0.8937,3.086
518+2024,pre,algorithms,1,0.0,0.0,0.0,0.0,1.0,1.0,2.0
519+2024,post,algorithms,29655,0.2628,0.1426,0.0331,0.2325,0.1478,0.6008,3.178
520+2024,baseline,applied_cryptography,18,0.3333,0.2778,0.0,0.5556,0.6111,0.9444,3.889
521+2024,post,applied_cryptography,12408,0.2445,0.0892,0.0497,0.172,0.0516,0.4971,3.358
522+2024,post,autonomous_scientific_discovery,456,0.1316,0.2368,0.0658,0.2061,0.25,0.7193,4.524
523+2024,baseline,blockchain,13,0.1538,0.3077,0.0,0.3846,0.3846,0.8462,4.0
524+2024,post,blockchain,15363,0.1381,0.0738,0.0235,0.142,0.0258,0.5146,3.143
525+2024,post,cloud_computing,8597,0.1599,0.0976,0.0776,0.128,0.0264,0.5386,3.166
526+2024,baseline,cloud_computing,12,0.1667,0.5,0.0,0.3333,0.4167,0.6667,4.0
527+2024,baseline,compilers,1,0.0,0.0,0.0,0.0,1.0,1.0,7.0
528+2024,post,compilers,767,0.146,0.1669,0.09,0.1669,0.0808,0.6454,3.568
529+2024,post,computational_biology,23499,0.2475,0.1911,0.0466,0.2245,0.0485,0.6748,5.41
530+2024,baseline,computational_biology,23,0.1739,0.4348,0.2609,0.6087,0.5217,0.9565,5.696
531+2024,baseline,computational_complexity,37,0.0811,0.2432,0.027,0.3243,0.5405,0.8108,2.757
532+2024,post,computational_complexity,3787,0.1627,0.122,0.019,0.1954,0.1336,0.6237,2.962
533+2024,post,computer_architecture,17149,0.2345,0.1865,0.1051,0.1558,0.0735,0.3984,4.488
534+2024,baseline,computer_architecture,10,0.1,0.4,0.0,0.4,0.8,1.0,7.7
535+2024,baseline,computer_graphics,21,0.1905,0.1429,0.1429,0.3333,0.5238,0.7619,4.19
536+2024,post,computer_graphics,10590,0.216,0.128,0.0579,0.1317,0.1256,0.6182,3.385
537+2024,baseline,computer_networks,103,0.2621,0.2136,0.1456,0.4757,0.6505,0.7961,4.049
538+2024,post,computer_networks,54239,0.3051,0.1057,0.0762,0.2299,0.0843,0.4255,3.934
539+2024,baseline,computer_vision,375,0.4613,0.2533,0.1573,0.3787,0.7173,0.84,5.067
540+2024,post,computer_vision,132942,0.4018,0.0919,0.0593,0.1611,0.1741,0.52,4.4
541+2024,baseline,cybersecurity,69,0.3043,0.3188,0.1159,0.3913,0.6667,0.8841,4.812
542+2024,post,cybersecurity,45479,0.1501,0.1174,0.0405,0.1389,0.0753,0.5558,3.311
543+2024,post,database_systems,15014,0.1659,0.097,0.0515,0.1251,0.0565,0.6648,3.497
544+2024,baseline,database_systems,10,0.2,0.3,0.0,0.3,0.7,0.9,3.9
545+2024,post,deep_learning_architectures,28853,0.3714,0.1316,0.0589,0.1904,0.2177,0.5622,4.154
546+2024,baseline,deep_learning_architectures,132,0.3182,0.303,0.0909,0.3258,0.7576,0.9015,4.129
547+2024,baseline,distributed_computing,27,0.1481,0.2222,0.1111,0.3333,0.5556,0.8148,3.963
548+2024,post,distributed_computing,10574,0.2745,0.1701,0.0681,0.1838,0.1154,0.526,3.864
549+2024,baseline,edge_computing,7,0.1429,0.5714,0.2857,0.4286,0.5714,0.7143,4.286
550+2024,post,edge_computing,10203,0.2225,0.0828,0.0465,0.2039,0.0297,0.4136,3.778
551+2024,post,federated_learning,2065,0.4232,0.1554,0.0634,0.2615,0.1758,0.4552,4.311
552+2024,baseline,federated_learning,5,0.0,0.4,0.4,0.2,0.6,0.8,3.8
553+2024,pre,formal_verification,1,0.0,0.0,0.0,0.0,1.0,0.0,1.0
554+2024,post,formal_verification,4194,0.1319,0.1328,0.042,0.2737,0.253,0.695,3.069
555+2024,baseline,formal_verification,38,0.0263,0.1842,0.0263,0.4737,0.8421,0.9737,2.842
556+2024,baseline,foundation_models,1,1.0,1.0,0.0,1.0,1.0,1.0,4.0
557+2024,post,foundation_models,1045,0.3407,0.2364,0.0947,0.2335,0.4871,0.6861,5.201
558+2024,baseline,generative_ai,11,0.2727,0.0909,0.0909,0.3636,0.8182,1.0,3.636
559+2024,post,generative_ai,7136,0.2379,0.1359,0.0425,0.1382,0.1392,0.5646,3.873
560+2024,baseline,hci,32,0.1875,0.2813,0.125,0.5625,0.4688,0.8438,4.906
561+2024,post,hci,48042,0.1042,0.1352,0.0298,0.1373,0.0617,0.6278,3.52
562+2024,post,information_retrieval,14120,0.3161,0.1125,0.0934,0.1618,0.1756,0.5898,3.642
563+2024,baseline,information_retrieval,26,0.3846,0.2308,0.2308,0.3846,0.8077,0.8462,4.577
564+2024,post,llm,24861,0.173,0.1976,0.0856,0.1443,0.2912,0.7103,4.11
565+2024,baseline,llm,12,0.3333,0.3333,0.25,0.4167,0.3333,0.6667,5.167
566+2024,baseline,ml_theory,179,0.162,0.3911,0.067,0.3855,0.6983,0.8994,3.626
567+2024,post,ml_theory,26531,0.1971,0.1731,0.0525,0.1778,0.2528,0.6623,3.573
568+2024,post,multimodal_ai,3402,0.525,0.137,0.0964,0.1999,0.3136,0.5632,5.121
569+2024,baseline,multimodal_ai,6,1.0,0.1667,0.1667,0.6667,0.6667,0.8333,5.833
570+2024,baseline,nlp,74,0.3108,0.2703,0.2027,0.2568,0.5946,0.8514,5.122
571+2024,post,nlp,38757,0.1934,0.1256,0.0565,0.1473,0.208,0.6729,3.796
572+2024,post,operating_systems,1259,0.2367,0.1374,0.0874,0.1398,0.0445,0.5409,3.742
573+2024,baseline,operating_systems,1,0.0,0.0,0.0,0.0,1.0,1.0,2.0
574+2024,baseline,privacy_preserving_computation,18,0.1111,0.6111,0.2778,0.3889,0.6111,1.0,4.444
575+2024,post,privacy_preserving_computation,6668,0.2722,0.1752,0.0604,0.2037,0.1435,0.5775,3.819
576+2024,baseline,programming_languages,8,0.0,0.125,0.125,0.375,0.75,0.75,4.25
577+2024,post,programming_languages,3838,0.08,0.149,0.0323,0.1628,0.1397,0.6709,2.706
578+2024,baseline,quantum_computing,75,0.1467,0.48,0.1733,0.4267,0.4133,0.92,4.44
579+2024,post,quantum_computing,12647,0.1791,0.215,0.0924,0.2163,0.0991,0.6726,4.147
580+2024,baseline,reinforcement_learning,46,0.1739,0.4348,0.1304,0.3043,0.7609,0.8261,3.957
581+2024,post,reinforcement_learning,9846,0.3523,0.1701,0.0639,0.172,0.2556,0.5581,3.828
582+2024,post,robotics,29916,0.3173,0.1614,0.0531,0.1716,0.2201,0.5058,4.302
583+2024,baseline,robotics,70,0.1429,0.4,0.1,0.2714,0.8286,0.9286,4.586
584+2024,baseline,scientific_computing,161,0.2609,0.3168,0.0435,0.3975,0.7081,0.8882,3.081
585+2024,post,scientific_computing,21726,0.3014,0.1647,0.0369,0.2337,0.2144,0.6183,3.49
586+2024,post,software_engineering,30489,0.1064,0.0944,0.0501,0.1222,0.0907,0.636,3.183
587+2024,baseline,software_engineering,44,0.0682,0.1591,0.1136,0.4318,0.5455,0.7727,4.159
588+2024,baseline,speech_audio,23,0.2174,0.1739,0.087,0.3478,0.4783,0.7826,4.435
589+2024,post,speech_audio,16404,0.2245,0.0943,0.0739,0.1312,0.1822,0.5574,3.721
590+2024,post,sustainable_computing,1271,0.2557,0.1338,0.0795,0.1943,0.0787,0.5578,3.889
591+2025,post,ai_accelerators,1906,0.351,0.1406,0.0714,0.1768,0.1081,0.4481,5.341
592+2025,post,ai_agents,8815,0.2288,0.1105,0.04,0.1416,0.2358,0.6535,3.683
593+2025,baseline,ai_agents,4,0.0,0.75,0.0,0.5,1.0,1.0,2.5
594+2025,post,ai_safety,10881,0.0543,0.1753,0.0297,0.1442,0.202,0.709,3.128
595+2025,baseline,ai_safety,4,0.0,0.5,0.25,0.5,1.0,1.0,2.25
596+2025,post,algorithms,33709,0.252,0.1255,0.0281,0.2162,0.1636,0.6326,3.179
597+2025,baseline,algorithms,86,0.093,0.2442,0.0233,0.4186,0.8488,1.0,2.965
598+2025,baseline,applied_cryptography,5,0.2,0.2,0.0,0.2,1.0,1.0,2.4
599+2025,post,applied_cryptography,14607,0.2125,0.0813,0.0509,0.1545,0.0637,0.5226,3.311
600+2025,post,autonomous_scientific_discovery,764,0.1414,0.1872,0.0537,0.1767,0.3141,0.7853,5.212
601+2025,post,blockchain,18011,0.1128,0.0705,0.0241,0.1341,0.0253,0.5348,3.069
602+2025,baseline,blockchain,1,0.0,0.0,0.0,1.0,1.0,1.0,3.0
603+2025,post,cloud_computing,10973,0.1211,0.1111,0.0722,0.1008,0.0327,0.5728,2.922
604+2025,baseline,cloud_computing,1,1.0,0.0,0.0,1.0,1.0,1.0,4.0
605+2025,post,compilers,767,0.1408,0.1499,0.0508,0.1499,0.0782,0.6597,3.58
606+2025,baseline,computational_biology,2,0.5,0.5,0.0,0.0,1.0,1.0,3.0
607+2025,post,computational_biology,28152,0.2449,0.1696,0.0385,0.2155,0.0575,0.6719,5.511
608+2025,baseline,computational_complexity,12,0.0833,0.25,0.0833,0.4167,1.0,1.0,2.667
609+2025,post,computational_complexity,4197,0.1384,0.1044,0.0157,0.1978,0.1592,0.6652,2.894
610+2025,baseline,computer_architecture,1,0.0,1.0,0.0,0.0,1.0,1.0,6.0
611+2025,post,computer_architecture,20208,0.2336,0.1645,0.0941,0.1425,0.0947,0.4181,4.374
612+2025,post,computer_graphics,12439,0.2169,0.1149,0.053,0.1274,0.1329,0.6286,3.498
613+2025,baseline,computer_graphics,1,0.0,0.0,0.0,1.0,1.0,1.0,6.0
614+2025,post,computer_networks,59546,0.3019,0.094,0.0708,0.2197,0.086,0.4309,3.999
615+2025,baseline,computer_networks,9,0.4444,0.3333,0.0,0.4444,1.0,1.0,3.333
616+2025,post,computer_vision,150701,0.384,0.0821,0.0528,0.1519,0.1899,0.5276,4.528
617+2025,baseline,computer_vision,56,0.3929,0.3214,0.1964,0.4643,1.0,1.0,4.804
618+2025,post,cybersecurity,59026,0.1374,0.1044,0.0351,0.1254,0.0853,0.5625,3.281
619+2025,baseline,cybersecurity,18,0.2778,0.2778,0.0556,0.3889,1.0,1.0,3.778
620+2025,baseline,database_systems,7,0.2857,0.4286,0.1429,0.2857,1.0,1.0,4.571
621+2025,post,database_systems,16674,0.1328,0.0945,0.0506,0.1136,0.0678,0.7097,3.437
622+2025,post,deep_learning_architectures,34110,0.3518,0.105,0.0488,0.1734,0.2299,0.58,3.985
623+2025,baseline,deep_learning_architectures,31,0.2903,0.1935,0.0645,0.4839,0.9677,1.0,3.677
624+2025,post,distributed_computing,12541,0.2456,0.1487,0.0657,0.1604,0.1293,0.5647,4.068
625+2025,pre,distributed_computing,1,0.0,0.0,0.0,0.0,1.0,1.0,1.0
626+2025,baseline,distributed_computing,8,0.375,0.25,0.0,0.25,1.0,1.0,3.25
627+2025,post,edge_computing,11554,0.2095,0.082,0.0413,0.1895,0.0286,0.4313,3.748
628+2025,baseline,federated_learning,1,0.0,0.0,0.0,0.0,1.0,1.0,6.0
629+2025,post,federated_learning,2543,0.4019,0.1074,0.0448,0.2206,0.1479,0.44,4.244
630+2025,baseline,formal_verification,10,0.1,0.2,0.1,0.4,0.9,1.0,3.6
631+2025,post,formal_verification,4613,0.1108,0.1288,0.0334,0.253,0.3213,0.7175,3.036
632+2025,baseline,foundation_models,1,1.0,0.0,0.0,0.0,1.0,1.0,3.0
633+2025,post,foundation_models,1853,0.3092,0.1641,0.0912,0.1878,0.4873,0.6999,7.008
634+2025,baseline,generative_ai,2,0.0,0.0,0.0,0.0,1.0,1.0,2.0
635+2025,post,generative_ai,12741,0.232,0.1183,0.0352,0.1352,0.1188,0.5894,3.315
636+2025,baseline,hci,4,0.5,0.0,0.25,1.0,1.0,1.0,5.25
637+2025,post,hci,58144,0.0952,0.1144,0.0237,0.1274,0.0696,0.6531,3.478
638+2025,baseline,information_retrieval,5,0.2,0.4,0.2,0.6,1.0,1.0,3.8
639+2025,post,information_retrieval,17374,0.283,0.0894,0.0706,0.1377,0.1806,0.6215,3.698
640+2025,baseline,llm,1,0.0,1.0,0.0,0.0,1.0,1.0,3.0
641+2025,post,llm,45157,0.1716,0.1428,0.0652,0.1272,0.2818,0.7223,4.274
642+2025,post,ml_theory,32743,0.1822,0.1381,0.0401,0.1614,0.246,0.672,3.503
643+2025,baseline,ml_theory,55,0.1636,0.4727,0.0909,0.4727,0.8364,1.0,3.273
644+2025,post,multimodal_ai,6749,0.4749,0.1193,0.0794,0.1886,0.3063,0.5709,5.353
645+2025,baseline,nlp,13,0.2308,0.1538,0.1538,0.3077,0.9231,1.0,6.308
646+2025,post,nlp,42901,0.1673,0.1029,0.0449,0.1313,0.1974,0.6974,3.847
647+2025,post,operating_systems,1363,0.2201,0.1409,0.102,0.1394,0.0741,0.5708,3.864
648+2025,baseline,privacy_preserving_computation,4,0.25,0.5,0.5,0.5,1.0,1.0,3.25
649+2025,post,privacy_preserving_computation,8813,0.248,0.145,0.0478,0.1896,0.1317,0.5814,3.765
650+2025,post,programming_languages,4190,0.0766,0.1344,0.0258,0.1492,0.1618,0.6912,2.693
651+2025,baseline,programming_languages,3,0.0,0.0,0.0,0.0,0.6667,1.0,1.0
652+2025,post,quantum_computing,17457,0.1445,0.1819,0.0782,0.1878,0.1051,0.6794,4.136
653+2025,baseline,quantum_computing,14,0.1429,0.3571,0.0,0.5714,0.5,1.0,4.071
654+2025,post,reinforcement_learning,13202,0.3453,0.1331,0.0507,0.1556,0.2486,0.564,3.872
655+2025,baseline,reinforcement_learning,8,0.375,0.5,0.375,0.25,0.875,1.0,2.875
656+2025,baseline,robotics,13,0.2308,0.3846,0.1538,0.3846,1.0,1.0,4.077
657+2025,post,robotics,33451,0.334,0.1384,0.0459,0.1619,0.2615,0.5187,4.502
658+2025,baseline,scientific_computing,25,0.2,0.44,0.04,0.48,0.8,1.0,3.04
659+2025,post,scientific_computing,23163,0.277,0.1471,0.0314,0.2194,0.2629,0.6314,3.489
660+2025,pre,scientific_computing,3,0.3333,0.0,0.0,0.3333,1.0,1.0,2.667
661+2025,baseline,software_engineering,5,0.0,0.0,0.0,0.6,1.0,1.0,3.4
662+2025,post,software_engineering,36376,0.0962,0.0879,0.0416,0.1078,0.1122,0.6499,3.196
663+2025,post,speech_audio,18372,0.2127,0.0741,0.0554,0.1185,0.1869,0.5608,3.854
664+2025,baseline,speech_audio,1,0.0,1.0,0.0,0.0,1.0,1.0,4.0
665+2025,post,sustainable_computing,1639,0.1751,0.1086,0.0586,0.1733,0.0763,0.6223,3.628
666+2026,post,ai_accelerators,1649,0.336,0.174,0.0491,0.1734,0.1552,0.5221,5.324
667+2026,baseline,ai_agents,2,0.0,0.0,0.0,0.5,1.0,1.0,2.5
668+2026,post,ai_agents,12207,0.2079,0.1603,0.0646,0.156,0.3532,0.7345,3.959
669+2026,post,ai_safety,12544,0.0608,0.1738,0.0316,0.1383,0.2433,0.7451,3.152
670+2026,post,algorithms,28744,0.2104,0.1472,0.029,0.2084,0.2672,0.7478,3.014
671+2026,baseline,algorithms,32,0.0625,0.3125,0.0938,0.375,0.9375,1.0,3.438
672+2026,post,applied_cryptography,12176,0.2034,0.0786,0.042,0.1438,0.0954,0.5779,3.224
673+2026,baseline,applied_cryptography,2,0.0,0.0,0.0,0.0,1.0,1.0,2.5
674+2026,post,autonomous_scientific_discovery,910,0.178,0.2176,0.0769,0.1615,0.4846,0.8253,5.401
675+2026,post,blockchain,13276,0.0923,0.0696,0.0158,0.1114,0.0316,0.5505,2.83
676+2026,post,cloud_computing,8774,0.1039,0.0871,0.052,0.0888,0.0482,0.5391,2.806
677+2026,post,compilers,621,0.124,0.1787,0.0709,0.1417,0.1514,0.7134,3.467
678+2026,pre,computational_biology,1,0.0,0.0,0.0,1.0,1.0,1.0,2.0
679+2026,post,computational_biology,23786,0.2089,0.1665,0.0377,0.2018,0.0782,0.7036,5.376
680+2026,post,computational_complexity,4066,0.0878,0.1151,0.015,0.1626,0.2445,0.8195,2.526
681+2026,baseline,computational_complexity,1,0.0,0.0,0.0,0.0,1.0,1.0,1.0
682+2026,baseline,computer_architecture,1,0.0,1.0,0.0,1.0,1.0,1.0,4.0
683+2026,post,computer_architecture,15793,0.2125,0.1722,0.0806,0.1376,0.1615,0.5432,4.239
684+2026,post,computer_graphics,10258,0.1866,0.0979,0.0441,0.1096,0.1741,0.7021,3.48
685+2026,baseline,computer_graphics,2,0.5,0.5,0.0,0.0,1.0,1.0,5.0
686+2026,post,computer_networks,42412,0.2719,0.0893,0.0573,0.2024,0.1242,0.4981,3.849
687+2026,baseline,computer_networks,4,0.0,0.5,0.0,0.5,1.0,1.0,2.25
688+2026,post,computer_vision,117896,0.3551,0.0818,0.049,0.1483,0.2417,0.5989,4.564
689+2026,baseline,computer_vision,15,0.4,0.1333,0.0,0.2,1.0,1.0,3.867
690+2026,post,cybersecurity,51616,0.1206,0.1037,0.0311,0.1147,0.1273,0.6154,3.168
691+2026,baseline,cybersecurity,2,1.0,0.0,0.0,0.5,1.0,1.0,6.0
692+2026,baseline,database_systems,1,1.0,1.0,0.0,1.0,1.0,1.0,7.0
693+2026,post,database_systems,14094,0.1137,0.0828,0.0474,0.1003,0.0968,0.7181,3.269
694+2026,baseline,deep_learning_architectures,10,0.3,0.2,0.0,0.4,1.0,1.0,3.4
695+2026,post,deep_learning_architectures,26143,0.2987,0.1228,0.0433,0.1686,0.3059,0.6849,3.797
696+2026,baseline,distributed_computing,6,0.1667,0.3333,0.1667,1.0,1.0,1.0,6.0
697+2026,post,distributed_computing,9996,0.2136,0.1585,0.0657,0.1464,0.2187,0.6822,3.691
698+2026,post,edge_computing,8718,0.1969,0.0724,0.0349,0.1734,0.0373,0.464,3.659
699+2026,post,federated_learning,2841,0.3752,0.1059,0.0338,0.2214,0.1309,0.4368,4.139
700+2026,baseline,formal_verification,1,0.0,0.0,0.0,0.0,1.0,1.0,2.0
701+2026,pre,formal_verification,1,0.0,0.0,0.0,0.0,1.0,1.0,3.0
702+2026,post,formal_verification,4864,0.1059,0.1295,0.0315,0.2243,0.3951,0.7932,2.89
703+2026,post,foundation_models,2688,0.2913,0.1972,0.0792,0.2102,0.5547,0.7879,5.634
704+2026,post,generative_ai,14511,0.1911,0.1083,0.0256,0.1209,0.1264,0.6498,3.177
705+2026,baseline,hci,3,0.0,0.6667,0.0,1.0,1.0,1.0,5.0
706+2026,post,hci,47216,0.0924,0.1211,0.022,0.1243,0.0952,0.7226,3.415
707+2026,post,information_retrieval,16901,0.2574,0.116,0.0983,0.1348,0.2389,0.7213,3.92
708+2026,baseline,llm,2,0.0,0.0,0.0,0.0,1.0,1.0,1.0
709+2026,post,llm,51452,0.1978,0.1728,0.0803,0.1447,0.3587,0.8004,4.187
710+2026,post,ml_theory,29441,0.178,0.1632,0.0424,0.164,0.3013,0.7548,3.52
711+2026,baseline,ml_theory,16,0.1875,0.4375,0.3125,0.375,1.0,1.0,3.625
712+2026,post,multimodal_ai,8806,0.433,0.1296,0.0781,0.191,0.3541,0.6757,5.221
713+2026,post,nlp,34784,0.1522,0.1131,0.0478,0.1346,0.2455,0.7464,3.857
714+2026,baseline,nlp,1,0.0,1.0,0.0,0.0,1.0,1.0,1.0
715+2026,post,operating_systems,1247,0.2109,0.1195,0.0834,0.1171,0.1171,0.66,3.48
716+2026,post,privacy_preserving_computation,7624,0.2152,0.1516,0.0403,0.1838,0.1613,0.6381,3.716
717+2026,baseline,privacy_preserving_computation,1,1.0,0.0,0.0,0.0,1.0,1.0,4.0
718+2026,post,programming_languages,3689,0.0716,0.1342,0.0279,0.1458,0.2372,0.791,2.586
719+2026,baseline,programming_languages,2,0.5,0.0,0.0,0.5,1.0,1.0,4.5
720+2026,baseline,quantum_computing,4,0.0,0.75,0.25,0.25,0.5,1.0,4.75
721+2026,post,quantum_computing,19268,0.1454,0.1911,0.0757,0.1967,0.1269,0.7802,3.953
722+2026,post,reinforcement_learning,12551,0.3249,0.1582,0.0658,0.1591,0.3488,0.6796,3.962
723+2026,baseline,reinforcement_learning,4,0.25,0.0,0.0,0.25,1.0,1.0,3.75
724+2026,baseline,robotics,2,0.5,0.0,0.0,0.0,1.0,1.0,4.5
725+2026,post,robotics,27672,0.3049,0.1535,0.0534,0.1627,0.4274,0.6469,4.875
726+2026,baseline,scientific_computing,7,0.2857,0.4286,0.0,0.2857,0.7143,1.0,2.714
727+2026,post,scientific_computing,18731,0.2406,0.1702,0.0299,0.217,0.4174,0.7527,3.33
728+2026,post,software_engineering,34259,0.1017,0.1019,0.046,0.112,0.1823,0.7272,3.16
729+2026,baseline,software_engineering,2,0.0,0.0,0.0,0.0,1.0,1.0,3.5
730+2026,post,speech_audio,15624,0.2118,0.1025,0.0739,0.1411,0.2551,0.6826,3.905
731+2026,post,sustainable_computing,1328,0.1416,0.1152,0.0361,0.1521,0.1273,0.6905,3.414
added results/bibliometrics/category_impact_yearly.csv +354 −0
Large diff (354 lines) is not loaded by default to keep the page fast.Load diff
modified results/bibliometrics/category_share_ci_yearly.csv +320 −320
Large diff (641 lines) is not loaded by default to keep the page fast.Load diff
modified results/bibliometrics/diversity_yearly.csv +8 −8
Large diff (17 lines) is not loaded by default to keep the page fast.Load diff
modified tables/make_tables.py +58 −10
Large diff (109 lines) is not loaded by default to keep the page fast.Load diff
Menu

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