{"id":"W4229440380","doi":"10.18653/v1/2022.naacl-main.192","title":"Necessity and Sufficiency for Explaining Text Classifiers: A Case Study in Hate Speech Detection","year":2022,"lang":"en","type":"article","venue":"Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":27,"is_retracted":false,"has_abstract":true,"ca_institutions":"National Research Council Canada","funders":"","keywords":"Computer science; Computational linguistics; Voice activity detection; Artificial intelligence; Linguistics; Speech recognition; Natural language processing; Speech processing; Philosophy","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":[0.02092578,0.0008222798,0.00065006,0.002553368,0.0025763,0.004355538,0.002043327,0.005321208,0.006246618],"category_scores_gemma":[0.1366175,0.0008672267,0.000789384,0.001727363,0.003619396,0.01278131,0.00225828,0.003603398,0.001261284],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001910228,"about_ca_system_score_gemma":0.001608403,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002319501,"about_ca_topic_score_gemma":0.002844749,"domain_scores_codex":[0.9862694,0.009157169,0.001196774,0.001372366,0.001621124,0.0003831696],"domain_scores_gemma":[0.6903337,0.2827417,0.006353762,0.008294853,0.01148499,0.0007910295],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00134368,0.0003409096,0.0649723,0.001693234,0.0001924966,0.006868726,0.02848339,0.0154672,0.01981616,0.4807053,0.01753551,0.3625811],"study_design_scores_gemma":[0.0001807545,0.0003303596,0.01377609,0.001051462,0.0003487722,0.00792482,0.00797301,0.313296,0.02792235,0.5771393,0.04988426,0.0001727132],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.161673,0.002440742,0.8040208,0.01242454,0.0002198243,0.0003649384,0.0007212149,0.001207789,0.0169271],"genre_scores_gemma":[0.7863095,0.0004738734,0.2089718,0.0006106236,0.0002121202,0.0002040442,0.0007893885,0.0003483188,0.002080235],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02092578,"threshold_uncertainty_score":0.1106674,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0191876154053738,"score_gpt":0.2586639038828571,"score_spread":0.2394762884774833,"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."}}