{"id":"W4411120416","doi":"10.18653/v1/2025.naacl-long.92","title":"AfriHate: A Multilingual Collection of Hate Speech and Abusive Language Datasets for African Languages","year":2025,"lang":"en","type":"article","venue":"","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"European Commission; DeepMind; Universität Hamburg; International Development Research Centre; Rockefeller Foundation","keywords":"Computer science; Natural language processing; Artificial intelligence; Linguistics; Speech recognition","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002039967,0.00009086965,0.000139639,0.0001803954,0.000100247,0.00006446981,0.0002037451,0.00005233383,0.000008040234],"category_scores_gemma":[0.0001115674,0.00008522707,0.00003535103,0.0004081074,0.0000358587,0.0001688201,0.0001023761,0.00005492425,0.000002574581],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002062574,"about_ca_system_score_gemma":0.0000472577,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006092058,"about_ca_topic_score_gemma":0.0003957073,"domain_scores_codex":[0.9992768,0.00002980741,0.0001558514,0.0002722266,0.00009763856,0.0001677279],"domain_scores_gemma":[0.9994481,0.0001171033,0.00006006614,0.0002648427,0.00007044295,0.00003941751],"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":[0.0001852054,0.0001873177,0.0002284525,0.0002933425,0.0001359999,0.00004856553,0.006371359,0.00002568897,0.0950203,0.01311202,0.008233563,0.8761582],"study_design_scores_gemma":[0.001410125,0.0002647257,0.001539024,0.00007592092,0.00003353569,0.00003036934,0.003266155,0.01949347,0.9670938,0.001136767,0.005407547,0.0002485114],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6725584,0.0006109361,0.315097,0.0007856872,0.0005038963,0.001046627,0.0001797949,0.0003637469,0.008853948],"genre_scores_gemma":[0.9504213,0.00002204249,0.04708671,0.0001161684,0.00002609665,0.00002803854,0.00003073774,0.000005353873,0.002263518],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8759097,"threshold_uncertainty_score":0.3475459,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.00752431814208908,"score_gpt":0.2785270477125188,"score_spread":0.2710027295704297,"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."}}