{"id":"W6910509767","doi":"10.48448/cg6v-4y14","title":"Explain by Evidence: An Explainable Memory-based Neural Network for Question Answering","year":2020,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Interpretability; Question answering; TRACE (psycholinguistics); Artificial neural network; Process (computing); Quality (philosophy); Generalization; Deep neural networks","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.003213354,0.0007831821,0.0007159341,0.0006923064,0.0007803987,0.0005866937,0.002048963,0.0003710701,0.0006261574],"category_scores_gemma":[0.001063434,0.0007900918,0.0001355098,0.002467952,0.001209441,0.001342848,0.0002072289,0.0005305314,0.0003910921],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006385003,"about_ca_system_score_gemma":0.001256592,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008364008,"about_ca_topic_score_gemma":0.0009767405,"domain_scores_codex":[0.9939423,0.0002679938,0.0006157632,0.001983035,0.001550071,0.001640902],"domain_scores_gemma":[0.9969176,0.0003031192,0.0006822426,0.001108524,0.0002518393,0.0007366539],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001648956,0.000138657,0.0001335114,0.0002810264,0.00002103664,0.00002475173,0.0001733353,0.03513893,0.01762923,0.0007734149,0.9392351,0.006286119],"study_design_scores_gemma":[0.001004219,0.0009467312,0.0000147732,0.001266673,0.0001047381,0.0000100868,0.0003546687,0.73288,0.002322251,0.0003606031,0.2593066,0.001428678],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.006608992,0.02632347,0.7542339,0.01221465,0.02027581,0.03161011,0.0060734,0.02588082,0.1167788],"genre_scores_gemma":[0.1658008,0.0002192676,0.5766985,0.008413726,0.01709687,0.002658152,0.005569428,0.01014326,0.2133999],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.697741,"threshold_uncertainty_score":0.999455,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04717623683347327,"score_gpt":0.3336192800388499,"score_spread":0.2864430432053766,"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."}}