{"id":"W4319460874","doi":"10.2196/45312","title":"How Does ChatGPT Perform on the United States Medical Licensing Examination (USMLE)? The Implications of Large Language Models for Medical Education and Knowledge Assessment","year":2023,"lang":"en","type":"article","venue":"JMIR Medical Education","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":2028,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institute of Diabetes and Digestive and Kidney Diseases; National Heart, Lung, and Blood Institute; National Center for Advancing Translational Sciences; Yale University","keywords":"Interpretability; United States Medical Licensing Examination; Computer science; Set (abstract data type); Medical education; Artificial intelligence; Medical school; Medicine","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.01367351,0.001517669,0.0008219362,0.001205851,0.0006195303,0.002669434,0.001571049,0.001932722,0.002898467],"category_scores_gemma":[0.06081767,0.0005042432,0.00103886,0.0005671629,0.0007134879,0.004203397,0.002235115,0.002275468,0.003284976],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001698004,"about_ca_system_score_gemma":0.001619462,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01788922,"about_ca_topic_score_gemma":0.02018998,"domain_scores_codex":[0.9923024,0.005157992,0.0002673524,0.001476162,0.0005696145,0.0002265417],"domain_scores_gemma":[0.9450803,0.04804195,0.001109278,0.002499263,0.002306538,0.0009626661],"domain_codex":null,"domain_gemma":"evaluation","domain_candidate":"evaluation","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.003623944,0.000974555,0.165722,0.0007298975,0.0008776613,0.0006409349,0.004277904,0.1794835,0.01252736,0.001595724,0.01470269,0.6148438],"study_design_scores_gemma":[0.0001593751,0.0009795465,0.0315001,0.0001059026,0.0002458671,0.0003658162,0.001328489,0.9472417,0.01076907,0.004028624,0.003142386,0.0001330449],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9032248,0.0007930435,0.07502103,0.002170488,0.0002229673,0.0003952623,0.001997687,0.00928633,0.0068884],"genre_scores_gemma":[0.9686548,0.0001580466,0.02575186,0.0004294671,0.00004359985,0.0001249076,0.00271313,0.0002754739,0.001848582],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9863265,"threshold_uncertainty_score":0.07231331,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07788344718376497,"score_gpt":0.4810281358814464,"score_spread":0.4031446886976815,"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."}}