{"id":"W4390821393","doi":"10.1109/tcss.2023.3344597","title":"Effect of Text Augmentation and Adversarial Training on Fake News Detection","year":2024,"lang":"en","type":"article","venue":"IEEE Transactions on Computational Social Systems","topic":"Misinformation and Its Impacts","field":"Social Sciences","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"National Research Council Canada; University of Windsor; Research and Productivity Council; University of Victoria","funders":"","keywords":"Adversarial system; Computer science; Classifier (UML); Artificial intelligence; Machine learning; Fake news; Training set; Computer security; Internet privacy","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0006507033,0.0001111998,0.0001687329,0.0002059667,0.0006694517,0.0001702614,0.00004938869,0.000153145,0.00005342275],"category_scores_gemma":[0.00002109807,0.0001074259,0.00009222318,0.000340237,0.0001005387,0.0003251504,4.224953e-7,0.0001819419,0.00004067573],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001989114,"about_ca_system_score_gemma":0.0001300315,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008486,"about_ca_topic_score_gemma":0.0001754279,"domain_scores_codex":[0.9985363,0.0003408742,0.0002826327,0.0001490521,0.0005409922,0.0001501407],"domain_scores_gemma":[0.9991318,0.0006100142,0.00008558043,0.00003335509,0.00006084899,0.00007839697],"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":[0.0003797094,0.00006190599,0.000003128119,0.0002765611,0.0001615303,0.000001931158,0.1144223,0.1340169,0.0003335049,0.0127231,0.0003952922,0.7372242],"study_design_scores_gemma":[0.01440031,0.007712958,0.003392291,0.002463662,0.0009443265,0.00005746186,0.1717655,0.726949,0.01275184,0.003662194,0.05314075,0.002759661],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4188871,0.00004738271,0.5645258,0.0007301622,0.004880837,0.000927238,0.00007942045,0.0002637931,0.009658298],"genre_scores_gemma":[0.9992786,0.000009716715,0.0000266999,0.00004940577,0.0003529638,0.00001483847,0.000007096762,0.00001066791,0.0002499985],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7344645,"threshold_uncertainty_score":0.5148949,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02867692008173512,"score_gpt":0.3300422168292402,"score_spread":0.3013652967475051,"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."}}