{"id":"W3109752747","doi":"10.18653/v1/2020.coling-main.230","title":"Improving Conversational Question Answering Systems after Deployment using Feedback-Weighted Learning","year":2020,"lang":"en","type":"preprint","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"CHIST-ERA; Ministerio de Educación, Cultura y Deporte; Nvidia; Samsung Advanced Institute of Technology; Canadian Institute for Advanced Research; Eusko Jaurlaritza; Agencia Estatal de Investigación; Samsung; Agence Nationale de la Recherche","keywords":"Computer science; Software deployment; Exploit; Domain (mathematical analysis); Binary number; Binary classification; Recommender system; Question answering; Matching (statistics); Artificial intelligence; Machine learning; Information retrieval; Human–computer interaction; Software engineering; Support vector machine","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004360468,0.0003398001,0.0003514998,0.0001667114,0.0001338684,0.000769783,0.0007049591,0.0002488185,0.0000235788],"category_scores_gemma":[0.00004777619,0.0003528991,0.0001176936,0.0001434714,0.00002127262,0.000434661,0.002008926,0.0007993432,0.00002913365],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004676647,"about_ca_system_score_gemma":0.0003078267,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00137733,"about_ca_topic_score_gemma":0.000007925326,"domain_scores_codex":[0.9972539,0.0001907643,0.0005581991,0.001059351,0.0005838363,0.0003539418],"domain_scores_gemma":[0.9987193,0.00005800569,0.0003285443,0.0005598008,0.0001747469,0.0001596073],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003368086,0.00005136068,0.008024695,0.001434812,0.0002115177,0.0001488161,0.003147238,0.9250301,0.006970155,0.03399793,0.0000366708,0.02091307],"study_design_scores_gemma":[0.000172642,0.00001628946,0.0002926596,0.0003036066,0.00002205232,0.00001499013,0.00006051123,0.9977449,0.0003374296,0.0004728832,0.000174491,0.0003875069],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07496369,0.0003157114,0.9209838,0.0003565936,0.002033716,0.0004054765,0.000001853985,0.0007200582,0.000219115],"genre_scores_gemma":[0.801034,0.000007594122,0.1982371,0.0001181189,0.0003996244,0.00004665909,0.00001190503,0.00002900621,0.000116015],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7260703,"threshold_uncertainty_score":0.9998923,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02939263299873394,"score_gpt":0.2517073156640895,"score_spread":0.2223146826653555,"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."}}