{"id":"W3155638757","doi":"","title":"Zero-shot Cross-lingual Content Filtering: Offensive Language and Hate Speech Detection","year":2021,"lang":"en","type":"article","venue":"Queen Mary Research Online (Queen Mary University of London)","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Engineering and Physical Sciences Research Council; European Commission; Javna Agencija za Raziskovalno Dejavnost RS; Institute for Catastrophic Loss Reduction","keywords":"Offensive; Computer science; Natural language processing; Speech recognition; Zero (linguistics); Classifier (UML); Artificial intelligence; Voice activity detection; Task (project management); Linguistics; Speech processing; Mathematics; Engineering","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":[0.001006971,0.000970025,0.001048987,0.001653548,0.0007807435,0.001117127,0.001037543,0.001277992,0.003694078],"category_scores_gemma":[0.002214282,0.0003760011,0.0009272511,0.0005943536,0.0004711346,0.001417404,0.002214416,0.000920115,0.005677526],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003818295,"about_ca_system_score_gemma":0.000730036,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00651034,"about_ca_topic_score_gemma":0.01025204,"domain_scores_codex":[0.9989237,0.0001294489,0.00005457939,0.0004071464,0.0002715291,0.000213648],"domain_scores_gemma":[0.9987569,0.0002484039,0.00006933826,0.0002973386,0.0005207551,0.0001073179],"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":[0.0008898759,0.0007919057,0.01818302,0.000354794,0.0003328591,0.001078896,0.001149976,0.003064611,0.1865756,0.001589649,0.02449442,0.7614944],"study_design_scores_gemma":[0.00008863436,0.0009803117,0.07695441,0.0001630425,0.0004680751,0.005083459,0.00145694,0.480652,0.3805846,0.004578573,0.04862264,0.0003672468],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4556989,0.001818165,0.4666736,0.0005379053,0.001022946,0.0004388641,0.005151157,0.04442773,0.02423075],"genre_scores_gemma":[0.7789714,0.000429059,0.1849844,0.000551726,0.0002555771,0.0002971268,0.01047805,0.001086774,0.02294587],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00651034,"threshold_uncertainty_score":0.01294488,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05353052957588103,"score_gpt":0.3054008333783639,"score_spread":0.2518703038024829,"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."}}