{"id":"W4404654936","doi":"10.2196/63129","title":"Performance of ChatGPT-4o on the Japanese Medical Licensing Examination: Evalution of Accuracy in Text-Only and Image-Based Questions","year":2024,"lang":"en","type":"article","venue":"JMIR Medical Education","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Preprint; Psychology; 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":[0.001647634,0.00009745851,0.0001702489,0.0002628322,0.00006143538,0.00001348409,0.00008512087,0.0001915411,0.0007046097],"category_scores_gemma":[0.005763396,0.00006937393,0.00003801814,0.0005615409,0.00029694,0.0001367592,0.00001368924,0.0004037821,0.00001969087],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001404343,"about_ca_system_score_gemma":0.003993025,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004490095,"about_ca_topic_score_gemma":0.00006218055,"domain_scores_codex":[0.9981267,0.0001694301,0.0005401439,0.0002059621,0.000812178,0.0001455164],"domain_scores_gemma":[0.9981316,0.001092977,0.0001019608,0.0002092591,0.0002827931,0.000181363],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.000108919,0.0009868118,0.01119422,0.001364242,0.00001244679,0.000003252789,0.01057417,0.000003698164,0.0007876844,0.002343807,0.001284513,0.9713362],"study_design_scores_gemma":[0.0003327029,0.001337313,0.743364,0.01882947,0.0001166014,0.0001976326,0.01233654,0.2097976,0.0094845,0.0005947328,0.003307643,0.000301279],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9518354,0.0005474605,0.00003083453,0.04566308,0.0006057863,0.0004921814,0.000001077311,0.00002476982,0.0007994082],"genre_scores_gemma":[0.997993,0.0003178652,0.00007243951,0.0009837825,0.000402492,0.00009803827,0.00003353943,0.00001095317,0.00008786803],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9710349,"threshold_uncertainty_score":0.7714981,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05894639362884545,"score_gpt":0.4490566095975038,"score_spread":0.3901102159686584,"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."}}