{"id":"W4410701915","doi":"10.1145/3715669.3726785","title":"Learning Disorder Detection Using Eye Tracking: Are Large Language Models Better Than Machine Learning?","year":2025,"lang":"en","type":"article","venue":"","topic":"Gaze Tracking and Assistive Technology","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":false,"ca_institutions":"Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Computer science; Artificial intelligence; Eye tracking; Machine learning; Tracking (education); Natural language processing; Psychology","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":[0.000288204,0.0001904983,0.0002114337,0.0003066626,0.0004288904,0.0001493391,0.0004621044,0.0001466622,0.00001793689],"category_scores_gemma":[0.00008688169,0.0001726027,0.00008866253,0.0006108164,0.00003830931,0.0003709512,0.0002855503,0.0006812958,0.00001716964],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005410969,"about_ca_system_score_gemma":0.00001973289,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002729396,"about_ca_topic_score_gemma":0.0003015461,"domain_scores_codex":[0.9986197,0.0001307038,0.0002008439,0.0004803039,0.0001607606,0.0004076581],"domain_scores_gemma":[0.9993932,0.0000581259,0.0001144423,0.0003278038,0.00006956264,0.00003686701],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0000191072,0.0003105412,0.668054,0.00007860546,0.0001233658,0.00007275012,0.002326959,0.03481981,0.01928098,0.01758422,0.00003437007,0.2572953],"study_design_scores_gemma":[0.0004112952,0.00005604819,0.02649297,0.00004756829,0.00001444109,0.000006178029,0.0005089748,0.9632502,0.00614253,0.001503893,0.001334526,0.00023134],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4561406,0.0001265611,0.5417183,0.0004738689,0.00009816989,0.00005099359,4.692851e-7,0.0006793676,0.000711632],"genre_scores_gemma":[0.9911553,0.00000553773,0.006561531,0.0002270335,0.00002618946,0.000007107885,0.000002888368,0.0000151333,0.001999247],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9284304,"threshold_uncertainty_score":0.7038535,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01232598692517258,"score_gpt":0.2681987318658679,"score_spread":0.2558727449406953,"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."}}