{"id":"W4413457979","doi":"10.2139/ssrn.5394370","title":"Correntropy Meets Cross-Entropy: A Robust Loss Against Noisy Labels","year":2025,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Optimal Experimental Design Methods","field":"Decision Sciences","cited_by":1,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"","keywords":"Cross entropy; Artificial intelligence; Entropy (arrow of time); Computer science; Mathematics; Pattern recognition (psychology); 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","scholarly_communication","open_science","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.02116663,0.00108972,0.001710331,0.00119092,0.0008487783,0.002885311,0.005735107,0.0008803641,0.0004897827],"category_scores_gemma":[0.004277891,0.0008884118,0.001309116,0.001257313,0.0005050034,0.0006408671,0.002697998,0.01126896,0.0005305283],"about_ca_system_candidate":true,"about_ca_system_consensus":true,"about_ca_system_score_codex":0.004844405,"about_ca_system_score_gemma":0.01283314,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008886404,"about_ca_topic_score_gemma":0.0001310512,"domain_scores_codex":[0.9832547,0.002357461,0.002733589,0.002037011,0.003905077,0.005712127],"domain_scores_gemma":[0.9926041,0.001421951,0.002065205,0.002005025,0.001373166,0.0005305917],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.003093747,0.002147773,0.02039979,0.000152833,0.003751377,0.0006703382,0.001893597,0.1630532,0.013761,0.3590755,0.01784439,0.4141564],"study_design_scores_gemma":[0.00393612,0.0007565916,0.0008986218,0.0005121941,0.0002596151,0.0008614189,0.001990986,0.01410596,0.005168657,0.9415808,0.02780393,0.002125141],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.243297,0.04735714,0.6629603,0.005588681,0.01895129,0.002258771,0.0003664262,0.0004202008,0.01880017],"genre_scores_gemma":[0.8065203,0.02266615,0.06779552,0.001683926,0.003521923,0.0001720913,0.0001003071,0.0002732922,0.09726646],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5951648,"threshold_uncertainty_score":0.9996443,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07808477875120536,"score_gpt":0.4289299879324521,"score_spread":0.3508452091812467,"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."}}