{"id":"W4402540843","doi":"10.2196/63430","title":"ChatGPT-4 Omni Performance in USMLE Disciplines and Clinical Skills: Comparative Analysis","year":2024,"lang":"en","type":"article","venue":"JMIR Medical Education","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":89,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Preprint; Medical education; Psychology; Medicine; Computer science; World Wide Web","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0008581128,0.0001017743,0.0002974708,0.0003534459,0.00005644949,0.00003284954,0.00005980725,0.0001750406,0.0004979458],"category_scores_gemma":[0.0004632257,0.00007979781,0.00007897498,0.001058227,0.0001647889,0.0001517595,0.00002234805,0.0004049813,0.0001072288],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009980376,"about_ca_system_score_gemma":0.001453828,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002135965,"about_ca_topic_score_gemma":0.0003779375,"domain_scores_codex":[0.9984979,0.00008994687,0.0005877177,0.0003269569,0.0003178644,0.0001796062],"domain_scores_gemma":[0.9990067,0.0003466758,0.00004785804,0.0001701685,0.0001048925,0.0003237196],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00003634573,0.0008600592,0.6219739,0.0002064715,0.00008082991,0.000003264566,0.008690058,0.00000210523,0.000002840995,0.0004109627,0.002137903,0.3655953],"study_design_scores_gemma":[0.00005936371,0.0001943916,0.9415879,0.0006511277,0.0002041926,0.00001871957,0.004206109,0.03849139,0.00007780327,0.0003775993,0.01400595,0.0001254954],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9824802,0.00117957,0.0001035944,0.0135101,0.001165631,0.0003511521,0.000001052564,0.00004352118,0.001165188],"genre_scores_gemma":[0.9959171,0.001006241,0.0001694993,0.0008486696,0.001014036,0.0001282656,0.00008564115,0.000006667499,0.0008239216],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3654698,"threshold_uncertainty_score":0.5452157,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1129561708718853,"score_gpt":0.5515908393364383,"score_spread":0.438634668464553,"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."}}