{"id":"W7130587969","doi":"10.1109/fllm67465.2025.11391216","title":"Auditing Facial Emotion Recognition Datasets for Posed Expressions and Racial Bias","year":2025,"lang":"","type":"article","venue":"","topic":"Emotion and Mood Recognition","field":"Psychology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University","funders":"","keywords":"Sadness; Facial expression; Anger; Audit; Harm; Face (sociological concept); Emotional expression","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","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0009884902,0.0003714006,0.0004058032,0.000529693,0.0009102835,0.0001833738,0.0001323058,0.0006075458,0.004428196],"category_scores_gemma":[0.001265863,0.0003905506,0.0001896221,0.0003488382,0.0001606875,0.0003146654,0.0001141806,0.0003516529,0.0003517092],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007765764,"about_ca_system_score_gemma":0.000142132,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007124399,"about_ca_topic_score_gemma":0.0001597842,"domain_scores_codex":[0.9969811,0.0004924962,0.0008445221,0.0009039262,0.0002258285,0.0005520781],"domain_scores_gemma":[0.9982304,0.0005948043,0.0003492266,0.0003174544,0.0003281473,0.0001799783],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0007689088,0.0007003946,0.0001014275,0.00030807,0.0002026424,0.000004711565,0.001871904,9.534828e-7,0.007704133,0.002252009,0.1146188,0.871466],"study_design_scores_gemma":[0.04054971,0.001978697,0.02128658,0.004123738,0.002335335,0.00009714211,0.01944195,0.004620199,0.03167868,0.05448961,0.8164983,0.00290011],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.4827526,0.00098552,0.1792925,0.009564338,0.0219551,0.007807289,0.02280642,0.000656501,0.2741797],"genre_scores_gemma":[0.9695366,0.0004185676,0.003655584,0.002913782,0.00140002,0.0002726517,0.01232495,0.00005123986,0.009426616],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8685659,"threshold_uncertainty_score":0.9998546,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1041308981096688,"score_gpt":0.377649117197802,"score_spread":0.2735182190881332,"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."}}