{"id":"W3115763169","doi":"10.18653/v1/2020.coling-main.456","title":"Explain by Evidence: An Explainable Memory-based Neural Network for Question Answering","year":2020,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Interpretability; Computer science; Artificial intelligence; TRACE (psycholinguistics); Machine learning; Tracing; Artificial neural network; Question answering","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":[0.0007385144,0.0007933162,0.0005718115,0.0007116813,0.0003188067,0.000901692,0.002053743,0.001425123,0.002404799],"category_scores_gemma":[0.003720003,0.0003928813,0.0007825281,0.0006929956,0.0006211447,0.002289007,0.001259114,0.001972187,0.0004730746],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008505535,"about_ca_system_score_gemma":0.0007548123,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007330193,"about_ca_topic_score_gemma":0.01372675,"domain_scores_codex":[0.9997948,0.00004837329,0.00001282212,0.00008242617,0.00003408293,0.00002750613],"domain_scores_gemma":[0.9991055,0.0005044988,0.0001045622,0.0001538716,0.00009664286,0.00003493546],"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":[0.0005577047,0.000259105,0.006557517,0.0003128588,0.0003303722,0.0002952326,0.0005318228,0.37729,0.009378595,0.02855311,0.00922168,0.5667121],"study_design_scores_gemma":[0.00001488127,0.00004568405,0.0005140512,0.00002592783,0.00005558154,0.00003572139,0.00001930085,0.974766,0.001506683,0.02174474,0.001258416,0.00001297358],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.101651,0.003312293,0.8822848,0.002770625,0.0001962455,0.0001523717,0.001545322,0.004096436,0.003990896],"genre_scores_gemma":[0.8399916,0.0009639871,0.1515375,0.0006844531,0.0001271613,0.000202213,0.001662786,0.0001251982,0.004705171],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007330193,"threshold_uncertainty_score":0.014575,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05916506338980908,"score_gpt":0.2892817277727288,"score_spread":0.2301166643829197,"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."}}