{"id":"W4411983184","doi":"10.1007/s43678-025-00951-0","title":"How well does GPT-4 perform on an emergency medicine board exam? A comparative assessment","year":2025,"lang":"en","type":"article","venue":"Canadian Journal of Emergency Medicine","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":2,"is_retracted":false,"has_abstract":false,"ca_institutions":"McGill University Health Centre; McGill University","funders":"","keywords":"Medicine; Emergency medicine; Medical education; Medical physics; Medical emergency","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.0040499,0.0003900123,0.0007252746,0.001930867,0.0006992725,0.001275451,0.001025489,0.0009420176,0.00270921],"category_scores_gemma":[0.02395342,0.0002332615,0.001222254,0.001130216,0.001066326,0.002477756,0.001421585,0.0007698984,0.0009761778],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001252191,"about_ca_system_score_gemma":0.0008511522,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006183307,"about_ca_topic_score_gemma":0.008688072,"domain_scores_codex":[0.9981841,0.0006011989,0.0003027022,0.0001842775,0.0005033697,0.000224275],"domain_scores_gemma":[0.9847938,0.006801599,0.002695799,0.0004713599,0.003422044,0.001815374],"domain_codex":null,"domain_gemma":"evaluation","domain_candidate":"evaluation","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0007509094,0.0004782196,0.9823186,0.00008085302,0.0002647711,0.0002433616,0.002454204,0.0001571757,0.0002061758,0.0001791654,0.000845765,0.0120209],"study_design_scores_gemma":[0.00005188316,0.001692224,0.9894549,0.00006927033,0.0001646384,0.0005960935,0.005639168,0.0004271663,0.0001471331,0.0002128884,0.001521194,0.00002342434],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9956011,0.0004358878,0.0001152789,0.0003196215,0.00004220277,0.0000310224,0.0002118295,0.00000408456,0.003239039],"genre_scores_gemma":[0.9978327,0.0005377543,0.0002640215,0.0001215068,0.00004719553,0.00003370764,0.0005058023,0.000007357945,0.0006498035],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9959501,"threshold_uncertainty_score":0.02141815,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3062881425983279,"score_gpt":0.500642595247326,"score_spread":0.1943544526489981,"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."}}