{"id":"W4405843673","doi":"10.1038/s41598-024-82223-y","title":"Enhancing practicality and efficiency of deepfake detection","year":2024,"lang":"en","type":"article","venue":"Scientific Reports","topic":"Digital Media Forensic Detection","field":"Computer Science","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"Computer Research Institute of Montréal","funders":"Natural Sciences and Engineering Research Council of Canada; Ministère de l'Économie, de l’Innovation et des Exportations du Québec","keywords":"Computer science; Data science","routes":{"ca_aff":true,"ca_fund":true,"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":[0.001814918,0.00008199563,0.0001057345,0.0002194154,0.0001076525,0.0007319421,0.0001093206,0.00004172248,0.000004246118],"category_scores_gemma":[0.0006246572,0.00007157194,0.00004664854,0.0009677877,0.0002086565,0.001082151,0.000135605,0.0000939012,0.00001336628],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003779444,"about_ca_system_score_gemma":0.000109183,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001393901,"about_ca_topic_score_gemma":0.00002900972,"domain_scores_codex":[0.9982733,0.00002986726,0.0003712861,0.0006688319,0.000474247,0.0001825306],"domain_scores_gemma":[0.9990087,0.0001054059,0.0001229768,0.0005655605,0.0001127103,0.00008466761],"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":[0.000003188064,0.00006992042,0.0001386072,0.0002103351,0.00001658518,0.0004081835,0.001477935,0.00001682473,0.431284,0.005383416,0.0002483602,0.5607426],"study_design_scores_gemma":[0.00002835156,0.0000706367,0.0003480376,0.0000882919,0.00001020859,0.001123412,0.00004312774,0.01652797,0.9297293,0.04296371,0.008940148,0.0001267708],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5200884,0.0002152023,0.4609895,0.00008348012,0.0161857,0.0001415733,2.369248e-7,0.0002334778,0.002062496],"genre_scores_gemma":[0.9964355,0.00000131867,0.003214412,0.000005149022,0.00003319687,0.000006154852,7.068383e-7,0.000004945879,0.000298645],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5606158,"threshold_uncertainty_score":0.7058134,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008676008576843905,"score_gpt":0.2437657504445885,"score_spread":0.2350897418677446,"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."}}