{"id":"W7130629240","doi":"","title":"End-to-End Training of Neural Retrievers for Open-Domain Question Answering","year":2021,"lang":"","type":"other","venue":"Open MIND","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Question answering; Salient; Task (project management); Artificial neural network; Supervised learning; Training set; Labrador Retriever; Unsupervised learning","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002954632,0.001857223,0.001063638,0.001187132,0.0005427591,0.001197075,0.002776433,0.002496062,0.0106622],"category_scores_gemma":[0.008281237,0.0005574232,0.001154474,0.000792825,0.0007152683,0.003490951,0.001886786,0.002745248,0.008201901],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008919673,"about_ca_system_score_gemma":0.0008984152,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003408767,"about_ca_topic_score_gemma":0.00737081,"domain_scores_codex":[0.9986123,0.0004177862,0.00009730338,0.0005470787,0.0001841178,0.0001414336],"domain_scores_gemma":[0.9970265,0.001613084,0.0001555941,0.0005894481,0.0004812606,0.000134123],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.000980353,0.001219423,0.005420605,0.0008286568,0.000305402,0.0003646841,0.0006209979,0.0631252,0.0331512,0.003308246,0.0313029,0.8593723],"study_design_scores_gemma":[0.000171075,0.0008610421,0.002437434,0.00009007862,0.0001358772,0.0003356502,0.0003735204,0.9316335,0.04223108,0.007266364,0.01441304,0.00005127173],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.262951,0.004647494,0.6322102,0.001183799,0.000466136,0.0008081415,0.002734547,0.07484822,0.02015043],"genre_scores_gemma":[0.6131878,0.0007253049,0.3455247,0.00130022,0.00019046,0.0007638001,0.01472377,0.001775771,0.02180807],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.0106622,"threshold_uncertainty_score":0.03566861,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1039148193804308,"score_gpt":0.3692658689008578,"score_spread":0.265351049520427,"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."}}