{"id":"W4391225179","doi":"10.1093/jamia/ocad252","title":"Harnessing the potential of large language models in medical education: promise and pitfalls","year":2024,"lang":"en","type":"review","venue":"Journal of the American Medical Informatics Association","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":89,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University Health Centre","funders":"National Key Research and Development Program of China; National Institutes of Health; National Natural Science Foundation of China","keywords":"Narrative; Misconduct; Process (computing); Engineering ethics; Psychology; Medicine; Medical education; Political science; Computer science; Engineering","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":[0.004242266,0.0001557073,0.001022312,0.0002288425,0.00006465899,0.00004391683,0.0003282548,0.0002570814,0.00004002513],"category_scores_gemma":[0.004480427,0.00007600667,0.0003127941,0.0006298708,0.0001499948,0.0001498059,0.0001018498,0.001473568,0.000005700002],"about_ca_system_candidate":true,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005154262,"about_ca_system_score_gemma":0.006818801,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001348435,"about_ca_topic_score_gemma":0.00003153263,"domain_scores_codex":[0.9956893,0.0002933951,0.002085846,0.00006679429,0.001657671,0.0002070358],"domain_scores_gemma":[0.9957488,0.0004367077,0.003103314,0.0001927039,0.0003016653,0.0002168797],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00001189654,0.000152794,0.0001240021,0.004365447,0.0001383716,0.000006452955,0.004539448,0.000002194243,1.527865e-7,0.0001242372,0.003843708,0.9866913],"study_design_scores_gemma":[0.0009406332,0.001160277,0.00110925,0.2827371,0.01188242,0.004475971,0.08898488,0.07521707,0.00001999343,0.006805756,0.5255297,0.001136944],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.02137746,0.9621,0.0003198546,0.01353436,0.001743999,0.0006564867,0.000009676649,0.000008600461,0.0002495286],"genre_scores_gemma":[0.0116019,0.9854746,0.0001930154,0.00170844,0.0008468891,0.0000129849,0.00001077288,0.00001751471,0.0001339151],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.9855543,"threshold_uncertainty_score":0.9988116,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05065315228572722,"score_gpt":0.4462854854152415,"score_spread":0.3956323331295143,"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."}}