1{2 "document_id": "lb_qasper_466bd29bcab1cdfdef327777808236bd2677e2a54414a32a",3 "family": "public_qasper",4 "split": "test",5 "document_hash": "89376fc48caf741a72bd103e4bc51d59ebe200284b743475e0c5c9c0a39b52b5",6 "questions": [7 {8 "question_id": "q00",9 "qtype": "qasper",10 "question": "What datasets did they use for evaluation?",11 "answer": "CSAT dataset, 20 newsgroups, Fisher Phase 1 corpus",12 "aliases": [13 "CSAT dataset , 20 newsgroups, Fisher Phase 1 corpus"14 ],15 "answer_type": "qa_f1",16 "tolerance": 0.0,17 "supporting_fact_ids": [],18 "answer_position_frac": -1.019 }20 ],21 "facts": []22}