{"id":"W4405194255","doi":"10.5753/stil.2024.245416","title":"Toxic Text Classification in Portuguese: Is LLaMA 3.1 8B All You Need?","year":2024,"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":"Blutip (Canada)","funders":"Universidade Federal de Ouro Preto; Fundação de Amparo à Pesquisa do Estado de Minas Gerais; Conselho Nacional de Desenvolvimento Científico e Tecnológico; Coordenação de Aperfeiçoamento de Pessoal de Nível Superior","keywords":"Portuguese; Computer science; Natural language processing; Artificial intelligence; Linguistics","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":[0.001778084,0.001218459,0.0006680427,0.001005995,0.0006676036,0.00190961,0.0008047846,0.001514597,0.00385086],"category_scores_gemma":[0.008326977,0.0003482948,0.0008051566,0.0005122513,0.0005013334,0.002020871,0.000956647,0.001511732,0.004913246],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009524095,"about_ca_system_score_gemma":0.001253785,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01412261,"about_ca_topic_score_gemma":0.02110511,"domain_scores_codex":[0.9991887,0.0003366869,0.00005285321,0.0002182428,0.0001256856,0.00007775131],"domain_scores_gemma":[0.9975677,0.001445274,0.0001384919,0.0003117617,0.0003625172,0.0001743266],"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":[0.002557403,0.001345071,0.0595432,0.00108797,0.0002746669,0.001609805,0.002137114,0.1057871,0.03853269,0.005393334,0.1015447,0.680187],"study_design_scores_gemma":[0.0001068172,0.0005085086,0.02211286,0.000207124,0.0001238917,0.0006815462,0.001125389,0.9186466,0.02148708,0.004863423,0.03001125,0.0001254899],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8232833,0.002487147,0.0903047,0.007719823,0.001696421,0.0004331731,0.01195408,0.02944702,0.03267432],"genre_scores_gemma":[0.9281538,0.0005052686,0.04218476,0.0008322665,0.0001933785,0.0001614173,0.01482574,0.001052922,0.0120903],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01412261,"threshold_uncertainty_score":0.02808082,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02507409172674729,"score_gpt":0.2668513962956643,"score_spread":0.2417773045689171,"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."}}