{"id":"W3025313450","doi":"10.48550/arxiv.2005.06608","title":"Understanding and Detecting Dangerous Speech in Social Media","year":2020,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Offensive; Exploit; Social media; Macro; Baseline (sea); Computer science; Internet privacy; Computer security; World Wide Web; Political science; Engineering; Operations research","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.000253712,0.000228812,0.0002755023,0.0002917803,0.0002247879,0.0001750841,0.0006421074,0.0002862747,0.000005691128],"category_scores_gemma":[0.00008145109,0.0002934762,0.00008498685,0.0007091023,0.00006854773,0.0002588965,0.001258811,0.0007827759,0.00001828036],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004095136,"about_ca_system_score_gemma":0.00008215781,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001093211,"about_ca_topic_score_gemma":0.0003717247,"domain_scores_codex":[0.9983946,0.0001256134,0.0001597698,0.000900344,0.0000970941,0.0003225775],"domain_scores_gemma":[0.9993029,0.0001391324,0.0001474949,0.0002502863,0.0000288903,0.0001312821],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0004016974,0.0003116627,0.0175599,0.001070277,0.0004965193,0.0173449,0.03503651,0.02819519,0.004991721,0.8076511,0.000920715,0.08601981],"study_design_scores_gemma":[0.001948941,0.0001237314,0.00458293,0.0002859711,0.00009886834,0.00006660017,0.002932476,0.466455,0.001730355,0.5197992,0.0003332008,0.00164273],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4102367,0.00002937032,0.5869082,0.0001886322,0.0004329893,0.0001563578,0.000003401965,0.0002668567,0.001777419],"genre_scores_gemma":[0.9978541,0.00006130737,0.001814276,0.00004820621,0.0001614538,4.169143e-7,0.000002567676,0.0000148243,0.00004282803],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5876174,"threshold_uncertainty_score":0.9999517,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2127232738839858,"score_gpt":0.2023706039008456,"score_spread":0.01035266998314016,"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."}}