{"id":"W4413277779","doi":"10.1109/taffc.2025.3599859","title":"Leveraging Eye Movement for Instructing Robust Video-Based Facial Expression Recognition","year":2025,"lang":"en","type":"article","venue":"IEEE Transactions on Affective Computing","topic":"Gaze Tracking and Assistive Technology","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"Hubei Key Laboratory of Intelligent Geo-Information Processing; Natural Science Foundation of Hubei Province; National Natural Science Foundation of China","keywords":"Facial expression; Computer science; Facial expression recognition; Computer vision; Eye movement; Artificial intelligence; Emotion recognition; Expression (computer science); Speech recognition; Human–computer interaction; Facial recognition system; Feature extraction","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":[0.0004130513,0.0002488339,0.0002548956,0.0005199657,0.0009110286,0.0001297297,0.0003852332,0.0001332328,0.000004003178],"category_scores_gemma":[0.00004704811,0.0002636597,0.0001775384,0.0006201491,0.00006035443,0.0002051004,0.00001003609,0.0004020116,0.000007621618],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002667628,"about_ca_system_score_gemma":0.00008198287,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001561808,"about_ca_topic_score_gemma":0.000006244941,"domain_scores_codex":[0.9982797,0.0001364675,0.0003010622,0.00071161,0.0001765373,0.0003946406],"domain_scores_gemma":[0.9986632,0.0006161085,0.0001505447,0.0003300556,0.0001916433,0.00004844723],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00004740819,0.0002123918,0.0002048082,0.00006941373,0.00004940013,0.000002863933,0.0002314639,0.1270312,0.02353936,0.0002644762,0.00005614995,0.8482911],"study_design_scores_gemma":[0.001462966,0.0001793453,0.001436514,0.0005490541,0.00002510877,0.000001284994,0.0001302677,0.4636919,0.529653,0.002516628,0.00006306125,0.0002908536],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1716171,0.000009026844,0.8252823,0.0003548564,0.001317792,0.0004702267,0.000007423577,0.0006580534,0.0002832614],"genre_scores_gemma":[0.8924426,7.941715e-7,0.1070329,0.0003575611,0.00004142911,0.00007617528,0.000002581658,0.00001411147,0.00003185048],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8480002,"threshold_uncertainty_score":0.9999816,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02881493789973664,"score_gpt":0.2737631537182268,"score_spread":0.2449482158184902,"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."}}