{"id":"W4409262152","doi":"10.1109/wacv61041.2025.00915","title":"FUN-AD: Fully Unsupervised Learning for Anomaly Detection with Noisy Training Data","year":2025,"lang":"en","type":"article","venue":"","topic":"Anomaly Detection Techniques and Applications","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"Information and Communications Technology Council; National Research Foundation of Korea","keywords":"Anomaly detection; Computer science; Unsupervised learning; Artificial intelligence; Training (meteorology); Training set; Machine learning; Pattern recognition (psychology)","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001902242,0.001882476,0.001721843,0.001730439,0.0007619936,0.001360102,0.00329025,0.001878112,0.004865842],"category_scores_gemma":[0.006240461,0.000916753,0.001353798,0.001232943,0.0008733307,0.00239146,0.002517433,0.002663532,0.003082667],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006493092,"about_ca_system_score_gemma":0.001557275,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005582392,"about_ca_topic_score_gemma":0.009693226,"domain_scores_codex":[0.9987524,0.0002725479,0.00006317954,0.0004036828,0.0003852889,0.0001228887],"domain_scores_gemma":[0.9976169,0.0008697999,0.0001079116,0.0007886828,0.000504184,0.0001125684],"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":[0.0006121623,0.0003044066,0.003067454,0.0002150136,0.0003181295,0.0001895496,0.0000836144,0.1012415,0.008752855,0.005537435,0.03190297,0.8477748],"study_design_scores_gemma":[0.00002290204,0.00004634463,0.0003432509,0.00000784029,0.00001680872,0.0000965603,0.00001091882,0.9836945,0.005032866,0.007901737,0.002809596,0.00001665425],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.006982353,0.000203199,0.9678967,0.00009749828,0.0001279737,0.00007711652,0.0006305925,0.02323879,0.0007458127],"genre_scores_gemma":[0.1425847,0.0001576354,0.8450647,0.0002563981,0.0001461681,0.0002656431,0.003883269,0.00174547,0.005896149],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005582392,"threshold_uncertainty_score":0.01627791,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04135908940490878,"score_gpt":0.2839425167875609,"score_spread":0.2425834273826521,"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."}}