{"id":"W4417477967","doi":"10.1007/s44267-025-00103-z","title":"Benchmarking Drag⋆ for eye direction transformation and beyond","year":2025,"lang":"en","type":"article","venue":"Visual Intelligence","topic":"Face recognition and analysis","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Info-communications Media Development Authority; National Research Foundation Singapore; National Research Foundation","keywords":"Benchmarking; Benchmark (surveying); Task (project management); Construct (python library); Transformation (genetics); Quality (philosophy); Point (geometry); Human eye","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":[],"consensus_categories":[],"category_scores_codex":[0.0002032208,0.00007448771,0.00009199143,0.0001853867,0.0001566167,0.0001470421,0.0001343318,0.00004075465,0.00001387787],"category_scores_gemma":[0.00002992446,0.00007150834,0.00005453102,0.0004709974,0.00002597212,0.0003908594,0.00002877513,0.00005056806,0.000009669736],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002100668,"about_ca_system_score_gemma":0.00002065702,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001218812,"about_ca_topic_score_gemma":0.00001530353,"domain_scores_codex":[0.9993886,0.00002192282,0.0001762192,0.0002059295,0.00008342627,0.0001239328],"domain_scores_gemma":[0.9996747,0.00009148741,0.00003461101,0.00008338426,0.00008171247,0.00003405063],"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":[0.000003988019,0.00002791225,0.00007805126,0.00003437285,0.00001611824,1.757373e-7,0.0004747237,0.0000732722,0.001120894,0.04619324,0.0000815127,0.9518957],"study_design_scores_gemma":[0.00006999737,0.00008582296,0.00034147,0.00004926565,0.00001963094,0.000001103353,0.0002218162,0.9114557,0.06784566,0.01387356,0.005902618,0.0001333864],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007680587,0.0001360448,0.9876065,0.0008020969,0.0002456533,0.0001341577,0.000001550802,0.00007501226,0.003318439],"genre_scores_gemma":[0.9867975,0.0001973775,0.01206475,0.0004379367,0.00002684854,0.00003235351,0.0000068677,0.000002691338,0.0004337527],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9791169,"threshold_uncertainty_score":0.2916026,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01519616167552419,"score_gpt":0.3253687731621749,"score_spread":0.3101726114866507,"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."}}