{"id":"W4412475991","doi":"10.3390/math13142285","title":"Toward Adaptive Unsupervised and Blind Image Forgery Localization with ViT-VAE and a Gaussian Mixture Model","year":2025,"lang":"en","type":"article","venue":"Mathematics","topic":"Digital Media Forensic Detection","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"123 Certification (Canada)","funders":"","keywords":"Mixture model; Computer science; Artificial intelligence; Pattern recognition (psychology); Gaussian; Convolutional neural network; Divergence (linguistics); Autoencoder; Benchmark (surveying); Expectation–maximization algorithm; Artificial neural network; Mathematics","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":[0.001649672,0.001450288,0.001931885,0.001219185,0.0005267599,0.001127838,0.002743358,0.001840994,0.001746135],"category_scores_gemma":[0.004449314,0.001039923,0.001491973,0.0008084325,0.001187676,0.002053509,0.002888794,0.002247374,0.001389415],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007344664,"about_ca_system_score_gemma":0.001500986,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004374031,"about_ca_topic_score_gemma":0.005927023,"domain_scores_codex":[0.9989097,0.0002489388,0.0000672818,0.0003364065,0.0003122815,0.0001254733],"domain_scores_gemma":[0.9987663,0.0004714282,0.0001208129,0.000287995,0.0002909952,0.00006248605],"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":[0.0003876025,0.0001498789,0.0013804,0.000182163,0.0002119055,0.0001869102,0.0002260553,0.4034006,0.02832504,0.01504958,0.004416585,0.5460833],"study_design_scores_gemma":[0.00001095406,0.00003038159,0.0001557967,0.000009991772,0.00001359554,0.0001079208,0.00001036226,0.9865601,0.007584884,0.00460767,0.0008891343,0.00001930921],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.004189398,0.0002639483,0.9940584,0.00005535221,0.00002216796,0.00002367576,0.00003048719,0.001011498,0.0003451176],"genre_scores_gemma":[0.2222044,0.0004172212,0.7715846,0.0003267996,0.00006310113,0.0001291456,0.0004406953,0.000278637,0.004555449],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004374031,"threshold_uncertainty_score":0.008724451,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01584347309708426,"score_gpt":0.2226926103081617,"score_spread":0.2068491372110775,"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."}}