{"id":"W4412791970","doi":"10.1001/jamaophthalmol.2025.2413","title":"Ophthalmological Question Answering and Reasoning Using OpenAI o1 vs Other Large Language Models","year":2025,"lang":"en","type":"article","venue":"JAMA Ophthalmology","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"U.S. National Library of Medicine","keywords":"Metric (unit); Medicine; Macro; Artificial intelligence; Natural language processing; Machine learning; Computer science; Operations management","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":[0.007448729,0.001755408,0.0008459799,0.00234922,0.0008003768,0.00404424,0.002565642,0.001374126,0.005535351],"category_scores_gemma":[0.03056198,0.0004971853,0.002576128,0.001766544,0.0007123405,0.005029626,0.003687676,0.002337483,0.002703109],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002329031,"about_ca_system_score_gemma":0.004007237,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0216545,"about_ca_topic_score_gemma":0.02239726,"domain_scores_codex":[0.9955668,0.001811881,0.0004729361,0.001059492,0.000887762,0.0002011226],"domain_scores_gemma":[0.9777561,0.01584705,0.0009899628,0.002629974,0.002085286,0.000691595],"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":[0.004491179,0.002437296,0.08523773,0.003254368,0.001311675,0.0008184749,0.004082958,0.1630434,0.01166172,0.0174811,0.07075396,0.6354261],"study_design_scores_gemma":[0.0003947153,0.0009839113,0.008912291,0.0003519631,0.0003648866,0.0003197194,0.001159379,0.9228588,0.008548186,0.01940045,0.0365234,0.0001823453],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.4294686,0.003626255,0.4148421,0.006420871,0.0006353242,0.003051578,0.02684685,0.08201157,0.03309679],"genre_scores_gemma":[0.6273972,0.0008256332,0.3188747,0.001374257,0.0001572979,0.001268607,0.04413487,0.00109488,0.004872347],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.0216545,"threshold_uncertainty_score":0.04305691,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1228985104852794,"score_gpt":0.4578100303736842,"score_spread":0.3349115198884048,"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."}}