{"id":"W4406358932","doi":"10.36227/techrxiv.173396113.31607552/v2","title":"Impact of Inaccurate Contamination Ratios on Robust Unsupervised Anomaly Detection: Experimental Investigation","year":2025,"lang":"en","type":"preprint","venue":"","topic":"Anomaly Detection Techniques and Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Sherbrooke","funders":"","keywords":"Contamination; Anomaly detection; Anomaly (physics); Computer science; Environmental science; Artificial intelligence; Pattern recognition (psychology); Biology; Physics","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0001921451,0.0003002762,0.0003098112,0.0004261459,0.0001560875,0.00015713,0.0007067596,0.0002907733,0.00004666132],"category_scores_gemma":[0.00002486244,0.0002813,0.0002636323,0.0005732194,0.00006396339,0.0003172387,0.0004616138,0.0003150157,0.00001032189],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004151297,"about_ca_system_score_gemma":0.0003108364,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004866936,"about_ca_topic_score_gemma":0.00003162677,"domain_scores_codex":[0.9982782,0.0001101603,0.0005443877,0.0006381571,0.0002633129,0.0001658062],"domain_scores_gemma":[0.9982352,0.00007062803,0.0004023534,0.0009026608,0.0003080949,0.00008111204],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0003027438,0.002638326,0.01131619,0.0006600671,0.0009346986,0.00001033664,0.00487782,0.06781823,0.4490935,0.3223237,0.005203552,0.1348208],"study_design_scores_gemma":[0.0004174965,0.0007076365,0.03315915,0.0001192969,0.00002019894,0.000004227099,0.00005761631,0.1729192,0.7897524,0.002357695,0.00006685915,0.000418236],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3493582,0.00002912392,0.6462866,0.0001461283,0.0001700271,0.0007447101,0.00001807928,0.0003966485,0.002850511],"genre_scores_gemma":[0.9835308,0.0000150294,0.01527832,0.0001007086,0.00005157885,0.0004484493,0.00003856565,0.0000097913,0.0005267995],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6341726,"threshold_uncertainty_score":0.9999639,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03582104492765254,"score_gpt":0.3056240092742931,"score_spread":0.2698029643466406,"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."}}