{"id":"W4385478179","doi":"10.1109/compsac57700.2023.00049","title":"Performance Evaluation of Transformer-based NLP Models on Fake News Detection Datasets","year":2023,"lang":"en","type":"article","venue":"","topic":"Misinformation and Its Impacts","field":"Social Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Cistel Technology (Canada); Carleton University","funders":"","keywords":"Transformer; Computer science; Natural language processing; Artificial intelligence; Engineering","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":[0.004431617,0.002295985,0.00125969,0.002288188,0.0007445351,0.001407969,0.001664283,0.001693294,0.001883184],"category_scores_gemma":[0.0111155,0.0003892252,0.001064794,0.001319614,0.0006489531,0.002964505,0.00121733,0.002458915,0.002003054],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002131113,"about_ca_system_score_gemma":0.001633496,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02561,"about_ca_topic_score_gemma":0.02105819,"domain_scores_codex":[0.997863,0.0008362129,0.0002223576,0.0004763593,0.0003914105,0.0002107851],"domain_scores_gemma":[0.9947485,0.003224381,0.0002756696,0.000598414,0.0009511423,0.0002018437],"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":[0.002424758,0.001323968,0.01865738,0.0009812228,0.0007567923,0.0005236625,0.0004216325,0.3871096,0.008718903,0.003519767,0.0474172,0.5281451],"study_design_scores_gemma":[0.00004678806,0.000184321,0.001280796,0.00003125579,0.00005036677,0.0001019995,0.0001281462,0.9903267,0.005047516,0.001216023,0.001554915,0.00003111728],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8182388,0.01389123,0.1028133,0.004684016,0.001389462,0.0004290998,0.01120172,0.02891748,0.01843489],"genre_scores_gemma":[0.9215383,0.001690823,0.04850077,0.0006804756,0.0001604633,0.0001593227,0.02124336,0.0003311943,0.005695295],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02561,"threshold_uncertainty_score":0.05092186,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.155940578426593,"score_gpt":0.3838809708651885,"score_spread":0.2279403924385954,"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."}}