{"id":"W4385571961","doi":"10.18653/v1/2023.acl-short.17","title":"A Weakly Supervised Classifier and Dataset of White Supremacist Language","year":2023,"lang":"en","type":"article","venue":"","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"York University; Carnegie Mellon University; University of Pittsburgh","keywords":"Classifier (UML); Computer science; Artificial intelligence; Counterexample; Generalization; Speech recognition; Natural language processing; Mathematics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001710911,0.001196687,0.0005355868,0.003294371,0.001527603,0.001099781,0.001460247,0.002433943,0.002709838],"category_scores_gemma":[0.008154204,0.0002323009,0.0006757437,0.001567265,0.001253571,0.00143477,0.001535093,0.002086412,0.004355648],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009090188,"about_ca_system_score_gemma":0.001381048,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005711927,"about_ca_topic_score_gemma":0.01086451,"domain_scores_codex":[0.9978023,0.000613955,0.0002319114,0.0005405661,0.0006318385,0.000179394],"domain_scores_gemma":[0.9933904,0.002343419,0.0006592748,0.001275371,0.00171777,0.0006137133],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001630636,0.004516931,0.1064192,0.001834415,0.0003019375,0.003642059,0.002559313,0.02039254,0.05720984,0.005973031,0.4565367,0.3389835],"study_design_scores_gemma":[0.0006377483,0.001980308,0.2105918,0.0006137793,0.0002320974,0.006679299,0.006077951,0.3221354,0.08910225,0.01230342,0.3491511,0.0004948614],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"dataset","genre_scores_codex":[0.7369323,0.002393458,0.04198406,0.003700845,0.001392983,0.001908356,0.1798311,0.008044905,0.02381197],"genre_scores_gemma":[0.5244654,0.0004626252,0.06312641,0.001080098,0.0004557848,0.001676156,0.3955337,0.0003076518,0.01289221],"genre_candidate":"dataset","genre_consensus":null,"teacher_disagreement_score":0.005711927,"threshold_uncertainty_score":0.01135731,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01541823408194179,"score_gpt":0.2504659125987392,"score_spread":0.2350476785167974,"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."}}