{"id":"W2954811041","doi":"10.18653/v1/s19-2092","title":"UNBNLP at SemEval-2019 Task 5 and 6: Using Language Models to Detect Hate Speech and Offensive Language","year":2019,"lang":"en","type":"article","venue":"","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of New Brunswick","funders":"","keywords":"Offensive; Computer science; Language model; Language identification; Task (project management); Natural language processing; Artificial intelligence; SemEval; Character (mathematics); Language understanding; Word (group theory); Speech recognition; Linguistics; Natural language","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":[0.006744954,0.00203527,0.001367248,0.001257963,0.001453102,0.002372252,0.001742427,0.00378406,0.01519664],"category_scores_gemma":[0.01159836,0.0004478406,0.0009258426,0.0005510621,0.0004757824,0.003050179,0.002769345,0.002990128,0.01815176],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008379778,"about_ca_system_score_gemma":0.001126966,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007954266,"about_ca_topic_score_gemma":0.01298315,"domain_scores_codex":[0.9964975,0.001776496,0.000220206,0.0006405538,0.0005595483,0.0003057286],"domain_scores_gemma":[0.992351,0.003074261,0.000377285,0.001794238,0.00163486,0.0007683779],"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":[0.005801214,0.006295852,0.04679938,0.001013374,0.0005937336,0.001319879,0.0008836184,0.01533935,0.03827211,0.002596909,0.3192535,0.5618311],"study_design_scores_gemma":[0.001739333,0.006546158,0.1226243,0.0004091831,0.0004349559,0.002724716,0.002898002,0.606802,0.09978537,0.01102882,0.1445011,0.0005060182],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7230721,0.002743956,0.08226609,0.004509862,0.005184271,0.00301523,0.08491626,0.02855222,0.06573995],"genre_scores_gemma":[0.7443804,0.0003449185,0.06932218,0.001539306,0.0005788965,0.001504291,0.1387091,0.001025057,0.04259588],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01519664,"threshold_uncertainty_score":0.05083787,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01165980983610699,"score_gpt":0.2339918116001839,"score_spread":0.2223320017640769,"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."}}