{"id":"W4385570093","doi":"10.18653/v1/2023.sicon-1.4","title":"What Makes a Good Counter-Stereotype? Evaluating Strategies for Automated Responses to Stereotypical Text","year":2023,"lang":"en","type":"article","venue":"","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"National Research Council Canada","funders":"","keywords":"Stereotype (UML); Correctness; Set (abstract data type); Tone (literature); Computer science; Stereotype threat; Social media; Social psychology; Psychology; Internet privacy; World Wide Web; Algorithm; Linguistics","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":[0.005075742,0.001178644,0.0005830164,0.001281566,0.0003814425,0.001491237,0.0009592175,0.001179841,0.002023246],"category_scores_gemma":[0.04431009,0.000300141,0.0002620014,0.0002547952,0.0005226207,0.001199455,0.0008089977,0.0005665683,0.0009500645],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004221713,"about_ca_system_score_gemma":0.0005670557,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0009985432,"about_ca_topic_score_gemma":0.001410307,"domain_scores_codex":[0.9951189,0.003049569,0.0003110246,0.0006999999,0.0006214031,0.0001992063],"domain_scores_gemma":[0.9169949,0.0715996,0.00488614,0.002355799,0.00302761,0.001136047],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.008743011,0.004760913,0.1906559,0.003256751,0.0005347259,0.000961226,0.01119004,0.0137151,0.1515901,0.001570087,0.002991255,0.610031],"study_design_scores_gemma":[0.0009663225,0.01243205,0.2696067,0.0005144309,0.001017471,0.001762083,0.009144553,0.5087081,0.1852206,0.004456389,0.005783232,0.000388104],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.969044,0.0001439984,0.02502856,0.0001170472,0.00005428021,0.0006554237,0.0002010632,0.001968321,0.002787325],"genre_scores_gemma":[0.9729864,0.00005927236,0.02551004,0.00009563597,0.00002152982,0.0002685072,0.0002126422,0.00008735294,0.0007585401],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.005075742,"threshold_uncertainty_score":0.02684337,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05505928854682075,"score_gpt":0.3571107279461463,"score_spread":0.3020514393993256,"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."}}