{"id":"W4412945576","doi":"10.18653/v1/2025.acl-long.407","title":"Data Laundering: Artificially Boosting Benchmark Results through Knowledge Distillation","year":2025,"lang":"en","type":"article","venue":"","topic":"Imbalanced Data Classification Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Boosting (machine learning); Benchmark (surveying); Computer science; Data mining; Machine learning; Artificial intelligence; Geology","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":[],"consensus_categories":[],"category_scores_codex":[0.0005978548,0.0001269162,0.0001305542,0.0000947288,0.000198783,0.0002916617,0.00217576,0.00007005417,0.000007714037],"category_scores_gemma":[0.0006952081,0.0001213688,0.00002070076,0.0007347414,0.00004920207,0.001409467,0.00146168,0.0001183191,0.00005742397],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006849166,"about_ca_system_score_gemma":0.0001708174,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002464722,"about_ca_topic_score_gemma":0.0000734592,"domain_scores_codex":[0.9983954,0.0000628212,0.0004438292,0.0007178102,0.0001573848,0.0002227499],"domain_scores_gemma":[0.9968474,0.0002783134,0.0001270803,0.00259024,0.0001253482,0.0000316031],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.000007761564,0.00005904357,0.00007660801,0.00002101149,0.00001096895,0.000001322423,0.0003478124,0.00001059286,0.002015631,0.8365563,0.09857307,0.06231992],"study_design_scores_gemma":[0.0003438406,0.00003058745,0.004170337,0.0001310122,0.00001106822,0.000003208195,0.000105901,0.3208291,0.02201921,0.05998879,0.5919725,0.0003944905],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00008466121,0.00005569381,0.7891362,0.001771008,0.0003066727,0.0001566448,0.00004863289,0.0007620818,0.2076783],"genre_scores_gemma":[0.5352206,0.00002798922,0.4597762,0.0003618811,0.00007707713,0.00001589574,0.0006026503,0.000008220803,0.003909444],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7765675,"threshold_uncertainty_score":0.4949276,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1078842696340384,"score_gpt":0.372544980241367,"score_spread":0.2646607106073286,"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."}}