{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002035846,0.001929474,0.00104607,0.001138966,0.0004863504,0.001449696,0.001785901,0.0015508,0.003060516],"category_scores_gemma":[0.005830025,0.0003938017,0.001299196,0.0006956454,0.000914852,0.001673621,0.001791384,0.002011028,0.001799856],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009912906,"about_ca_system_score_gemma":0.001085037,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009656572,"about_ca_topic_score_gemma":0.01277012,"domain_scores_codex":[0.9990008,0.0002672445,0.00004216245,0.0004041044,0.0001722474,0.0001135054],"domain_scores_gemma":[0.9984666,0.0006272815,0.00006736648,0.0005739178,0.0001773745,0.00008748869],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001063023,0.0004495835,0.01020388,0.0006250748,0.0003713933,0.000309972,0.0001292088,0.4805598,0.01760315,0.005529298,0.04202529,0.4411303],"study_design_scores_gemma":[0.00005425159,0.0001267298,0.0011023,0.00003502081,0.00002284442,0.0001505469,0.00003468465,0.9833266,0.009386574,0.0027684,0.002969929,0.00002204358],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.4865333,0.009979878,0.4310199,0.002064304,0.001390287,0.0004945667,0.00816238,0.04241745,0.01793793],"genre_scores_gemma":[0.8705953,0.0007669523,0.1091206,0.0006884873,0.0001250383,0.000117926,0.01199244,0.001001141,0.005592011],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.009656572,"threshold_uncertainty_score":0.01920074,"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."}}