{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002364327,0.002156098,0.001784608,0.001032358,0.0007751738,0.00139516,0.002988877,0.004441448,0.01502847],"category_scores_gemma":[0.005491352,0.0008302417,0.001031901,0.0006172077,0.0003272475,0.002681126,0.00180891,0.002358597,0.01984116],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006472453,"about_ca_system_score_gemma":0.00121382,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006735649,"about_ca_topic_score_gemma":0.01028357,"domain_scores_codex":[0.9985853,0.0002892444,0.0000912543,0.0006676474,0.0001629162,0.0002035678],"domain_scores_gemma":[0.9967956,0.001719901,0.00008959552,0.0003920959,0.0007051523,0.0002976438],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001523422,0.001271359,0.00414315,0.0003098508,0.0002162057,0.0004848172,0.0003585007,0.0115089,0.08020771,0.0003939576,0.0203601,0.879222],"study_design_scores_gemma":[0.0002867969,0.001720676,0.009822775,0.00006721715,0.0003098929,0.0009067785,0.000561227,0.8399414,0.1320029,0.001677245,0.01256487,0.0001382768],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3708753,0.003713966,0.5165703,0.0008719235,0.001122534,0.0007076334,0.002533874,0.09416978,0.009434603],"genre_scores_gemma":[0.6192007,0.0005714722,0.338398,0.0008862338,0.0002446273,0.0005108251,0.01105216,0.001373172,0.02776271],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01502847,"threshold_uncertainty_score":0.05027527,"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."}}