{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002395403,0.0009761433,0.0004362077,0.0009496554,0.0005895484,0.0007358849,0.0007613726,0.0008728236,0.001594951],"category_scores_gemma":[0.007246328,0.0001786182,0.0005992771,0.0005462345,0.0006725906,0.0009496269,0.0008870414,0.001229983,0.001275624],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005683749,"about_ca_system_score_gemma":0.0003708224,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002768975,"about_ca_topic_score_gemma":0.00769197,"domain_scores_codex":[0.997466,0.0005959731,0.0002296988,0.0004900965,0.001010175,0.0002081127],"domain_scores_gemma":[0.9967862,0.0007206918,0.0003888391,0.0009725629,0.001024754,0.0001070015],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.002732316,0.002092733,0.1281968,0.001282365,0.0006361529,0.001317816,0.001306158,0.02063757,0.1266754,0.005094771,0.215289,0.4947388],"study_design_scores_gemma":[0.0002887365,0.001745864,0.5003062,0.0003976022,0.0003119616,0.007049713,0.00237474,0.2282252,0.1337951,0.007274123,0.117906,0.0003247703],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8665203,0.001809325,0.07260323,0.002221778,0.001854241,0.001781028,0.02869264,0.005817275,0.01870012],"genre_scores_gemma":[0.8591912,0.0006836529,0.06437457,0.001196096,0.0002601036,0.001164354,0.06539059,0.0003932962,0.007346255],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.002768975,"threshold_uncertainty_score":0.01266825,"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."}}