1{2 "document_id": "lb_qasper_8b6bf313950a892cbda035f2c7b3d8b01472ff34749f028d",3 "family": "public_qasper",4 "split": "test",5 "document_hash": "67631a9f6fd1ea44c64cb611f6f31a9dd6e7f1f7f36d6ce53dfb272880db88ff",6 "questions": [7 {8 "question_id": "q00",9 "qtype": "qasper",10 "question": "How does their model improve interpretability compared to softmax transformers?",11 "answer": "the attention heads in the proposed adaptively sparse Transformer can specialize more and with higher confidence",12 "aliases": [13 "We introduce sparse attention into the Transformer architecture"14 ],15 "answer_type": "qa_f1",16 "tolerance": 0.0,17 "supporting_fact_ids": [],18 "answer_position_frac": -1.019 }20 ],21 "facts": []22}