{"id":"W3015276915","doi":"10.1145/3318464.3389696","title":"Complaint-driven Training Data Debugging for Query 2.0","year":2020,"lang":"en","type":"article","venue":"","topic":"Machine Learning and Data Classification","field":"Computer Science","cited_by":34,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"Innovative Research Group Project of the National Natural Science Foundation of China","keywords":"Debugging; Retraining; Heuristic; Set (abstract data type); Training set; Inference; Training (meteorology)","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":[],"consensus_categories":[],"category_scores_codex":[0.0002157362,0.00006800532,0.00009485736,0.000024135,0.0001251468,0.0001493967,0.001433324,0.00002153295,0.00001674968],"category_scores_gemma":[0.000257357,0.00006085347,0.00001978479,0.0001308456,0.00001274076,0.0005232434,0.0004055897,0.00008553544,0.00004843379],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000006440299,"about_ca_system_score_gemma":0.00004714385,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000185459,"about_ca_topic_score_gemma":0.0000111386,"domain_scores_codex":[0.9991578,0.00003457201,0.0001353952,0.000413119,0.0001035605,0.0001554936],"domain_scores_gemma":[0.9990072,0.000166467,0.00005185308,0.000658711,0.00002313923,0.00009268517],"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":[0.00001098771,0.00002467329,0.001078784,0.00005157138,0.00002783731,0.000004187794,0.002908063,0.0007345667,0.001801647,0.3202114,0.0610769,0.6120694],"study_design_scores_gemma":[0.0001374561,0.00002455256,0.0004980188,0.000004413781,0.000002895478,0.000002722122,0.00006197289,0.8504168,0.00002836517,0.0002244463,0.1485215,0.00007686062],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0001512154,0.00001799587,0.9696946,0.02580452,0.00007595222,0.0001022228,0.00001975613,0.0004320231,0.003701677],"genre_scores_gemma":[0.5460098,0.000002002586,0.4507036,0.002778622,0.0001300077,0.000006748588,0.0002846212,0.000006412316,0.00007819258],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8496823,"threshold_uncertainty_score":0.2663497,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3264365626267824,"score_gpt":0.3428036694085752,"score_spread":0.01636710678179271,"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."}}