{"id":"W4386788271","doi":"10.1101/2023.09.14.23295571","title":"Comparing the Performance of ChatGPT and GPT-4 versus a Cohort of Medical Students on an Official University of Toronto Undergraduate Medical Education Progress Test","year":2023,"lang":"en","type":"preprint","venue":"medRxiv","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Guelph; University of Toronto","funders":"","keywords":"McNemar's test; Cohort; Test (biology); Medical education; Medicine; Medical school; Cohort study; Psychology; Mathematics education; Internal medicine; Statistics; Mathematics; Biology","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001685807,0.0001403053,0.0004424087,0.00006377474,0.00007738676,0.000007662396,0.0005399827,0.0002888606,0.00007685165],"category_scores_gemma":[0.0007776687,0.0001146355,0.00005962821,0.0001021548,0.0004639759,0.00004398937,0.0003161556,0.0004318298,0.00000340296],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001367393,"about_ca_system_score_gemma":0.001929361,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01286871,"about_ca_topic_score_gemma":0.008231935,"domain_scores_codex":[0.9972236,0.0001261583,0.0005047024,0.000296616,0.001688381,0.0001605215],"domain_scores_gemma":[0.9983395,0.0003353518,0.0003554126,0.0004300104,0.0003155055,0.000224233],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0005050819,0.0009265971,0.9756825,0.0007938393,0.00008777477,0.000002889113,0.002963452,0.0000122531,0.000009402575,0.0001800484,0.0001185585,0.01871763],"study_design_scores_gemma":[0.0002451246,0.001164524,0.980903,0.002470348,0.0001993438,0.00000437055,0.004001923,0.009950206,0.0007304801,0.00005961798,0.0001520363,0.0001190329],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9936672,0.0003108057,0.0000148056,0.003997899,0.001271302,0.0005873036,0.000005047831,0.00002289911,0.0001227525],"genre_scores_gemma":[0.9970963,0.00243552,0.00003856022,0.0000366967,0.000263635,0.000009027182,0.00005275289,0.00001473091,0.00005277393],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.0185986,"threshold_uncertainty_score":0.9937047,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1164460701808062,"score_gpt":0.4285416007085918,"score_spread":0.3120955305277856,"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."}}