{"id":"W4406070280","doi":"10.1007/s10115-024-02321-1","title":"Fake news detection: comparative evaluation of BERT-like models and large language models with generative AI-annotated data","year":2025,"lang":"en","type":"article","venue":"Knowledge and Information Systems","topic":"Misinformation and Its Impacts","field":"Social Sciences","cited_by":43,"is_retracted":false,"has_abstract":false,"ca_institutions":"Toronto Metropolitan University; Vector Institute","funders":"","keywords":"Computer science; Generative grammar; Language model; Fake news; Artificial intelligence; Natural language processing; Generative model; Information retrieval","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":[0.02314073,0.002669788,0.002119899,0.004082035,0.001654141,0.004608538,0.004659823,0.004596998,0.004127358],"category_scores_gemma":[0.07072268,0.001182483,0.00180768,0.002378218,0.00184868,0.008757744,0.002682497,0.004013559,0.002443053],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003735532,"about_ca_system_score_gemma":0.003025491,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03869615,"about_ca_topic_score_gemma":0.03139057,"domain_scores_codex":[0.988739,0.007384152,0.000699501,0.001665825,0.001121309,0.0003902461],"domain_scores_gemma":[0.8040937,0.179518,0.002450871,0.006114888,0.00570482,0.002117758],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.01129777,0.004307502,0.0269958,0.002233483,0.002369637,0.0005777514,0.001710602,0.6459736,0.002761446,0.01000018,0.0185307,0.2732415],"study_design_scores_gemma":[0.0001222335,0.0001778089,0.0008871219,0.00003980775,0.0001265264,0.00005569824,0.0001292367,0.9947038,0.0007170009,0.002536715,0.0004690857,0.00003489913],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7562237,0.00705089,0.1855296,0.008416146,0.001103988,0.0009776815,0.006865086,0.01887406,0.01495866],"genre_scores_gemma":[0.9308776,0.0008616416,0.05618494,0.0005697434,0.0002324105,0.000231988,0.007903374,0.0007026882,0.002435714],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.03869615,"threshold_uncertainty_score":0.1223813,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1036317786528191,"score_gpt":0.3879677179702772,"score_spread":0.2843359393174581,"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."}}