{"id":"W7138308848","doi":"10.18653/v1/2025.sciprodllm-1.2","title":"Human-Centered Disability Bias Detection in Large Language Models","year":2025,"lang":"","type":"article","venue":"","topic":"Text Readability and Simplification","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université du Québec à Montréal","funders":"","keywords":"Language model; Noise (video); Feature (linguistics); Data collection; Natural language","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":[0.001466702,0.000251419,0.0003261654,0.0001772689,0.000283015,0.0003066484,0.0008124814,0.0002333927,0.0001356356],"category_scores_gemma":[0.0001339002,0.0002518005,0.0001603357,0.001262521,0.0001554227,0.001044638,0.0004440752,0.0003681167,0.0000404],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004873381,"about_ca_system_score_gemma":0.0000924815,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001938265,"about_ca_topic_score_gemma":0.01540574,"domain_scores_codex":[0.9969189,0.0004762994,0.0008031277,0.001000266,0.0002747074,0.0005266735],"domain_scores_gemma":[0.9980385,0.0001342101,0.0001161306,0.001536419,0.000086837,0.00008785031],"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":[0.00007226264,0.004948428,0.03723164,0.0004757838,0.00004560878,0.000007096764,0.01150699,0.001412699,0.008265457,0.2969143,0.0001068881,0.6390129],"study_design_scores_gemma":[0.001044775,0.00007625676,0.1945886,0.00008637984,0.00001325933,0.00000132329,0.0009890756,0.763541,0.008857996,0.03024874,0.000239739,0.0003128301],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5195337,0.0001261684,0.4738108,0.0008599042,0.0003373516,0.0004616267,0.000009021408,0.0001169585,0.004744466],"genre_scores_gemma":[0.9975837,0.000009819496,0.0005329255,0.0002647274,0.00003124294,0.00003905455,0.000006746607,0.00000647391,0.001525265],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7621284,"threshold_uncertainty_score":0.9999934,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05454424794295367,"score_gpt":0.3174023805049431,"score_spread":0.2628581325619895,"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."}}