{"id":"W3214546336","doi":"","title":"End-to-End Training of Multi-Document Reader and Retriever for Open-Domain Question Answering","year":2021,"lang":"en","type":"article","venue":"arXiv (Cornell University)","topic":"Intelligent Tutoring Systems and Adaptive Learning","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":false,"ca_institutions":"McGill University","funders":"","keywords":"Question answering; Computer science; Information retrieval; Domain (mathematical analysis); World Wide Web; Open domain; Mathematics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004826231,0.0001100433,0.0001869959,0.0000902928,0.0001421843,0.0001013831,0.000393665,0.00004949882,0.00001209437],"category_scores_gemma":[0.0000747326,0.0001260991,0.00005080945,0.0003685513,0.00002722516,0.0005024752,0.0003485495,0.00008906841,0.000004366287],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007605827,"about_ca_system_score_gemma":0.00007437739,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001156793,"about_ca_topic_score_gemma":0.00003137985,"domain_scores_codex":[0.9989866,0.00009353931,0.000152589,0.0004884351,0.00006127526,0.0002175508],"domain_scores_gemma":[0.999271,0.0001095418,0.000096485,0.0002730508,0.0001534936,0.00009639307],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003014484,0.00003826055,0.002327049,0.0000436959,0.0000477227,0.00009997145,0.00341616,0.01014223,0.006352727,0.973639,0.00002344901,0.00383964],"study_design_scores_gemma":[0.0116258,0.002141258,0.04220274,0.003233646,0.0002255195,0.0002015425,0.03061406,0.5027989,0.1424491,0.04144803,0.2192861,0.003773325],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2072119,0.00003982589,0.7919388,0.00008359669,0.000174654,0.0001798529,0.000002306052,0.00002914334,0.0003398955],"genre_scores_gemma":[0.9584956,0.00001072723,0.03417923,0.00005331051,0.00003018145,9.678105e-7,0.000001787788,0.000008439184,0.007219722],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9321909,"threshold_uncertainty_score":0.5142174,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.116351746990044,"score_gpt":0.2378018237176534,"score_spread":0.1214500767276093,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}