{"id":"W4417090763","doi":"10.1109/icmla66185.2025.00041","title":"Sliced-Wasserstein Distance-based Data Selection","year":2025,"lang":"en","type":"article","venue":"","topic":"Anomaly Detection Techniques and Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"Natural Sciences and Engineering Research Council of Canada; Mitacs","keywords":"Benchmark (surveying); Anomaly detection; Selection (genetic algorithm); Scalability; Data point; Model selection; Feature selection; Euclidean distance; Data modeling","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":[0.0001154051,0.00005803707,0.00005604611,0.00007824208,0.0001463496,0.000103395,0.0009251819,0.00003417767,0.00002318317],"category_scores_gemma":[0.000009361872,0.00005359367,0.00002015199,0.0007863736,0.00001462733,0.0002709724,0.0001847243,0.00006192469,0.00002399353],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000358129,"about_ca_system_score_gemma":0.00007247589,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005841146,"about_ca_topic_score_gemma":0.00007720739,"domain_scores_codex":[0.9993756,0.00001490819,0.0001133617,0.0003205185,0.00007469251,0.0001009414],"domain_scores_gemma":[0.998997,0.00002867402,0.00003039691,0.0008755986,0.00004357427,0.00002474603],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00000248255,0.00005970819,0.0005958215,0.000008872499,0.000007739946,2.466241e-7,0.000005295424,0.00005193097,0.002222501,0.9013612,0.03449756,0.0611867],"study_design_scores_gemma":[0.0001234072,0.00002286486,0.001023528,0.000009339631,0.000004947033,7.956902e-7,0.000007518784,0.6204034,0.03648544,0.005869192,0.3359388,0.000110796],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0002010777,0.00001515999,0.97329,0.002877313,0.00005334399,0.0001187,0.000002299255,0.0007115426,0.02273057],"genre_scores_gemma":[0.7993487,0.000004299675,0.1945565,0.001021757,0.00001816427,0.00004780664,0.00001000729,0.000002956323,0.004989748],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.895492,"threshold_uncertainty_score":0.2185487,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02196130828809907,"score_gpt":0.2936971163154363,"score_spread":0.2717358080273372,"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."}}