{"id":"W4396680020","doi":"10.2196/56342","title":"Assessing the Ability of a Large Language Model to Score Free-Text Medical Student Clinical Notes: Quantitative Study","year":2024,"lang":"en","type":"article","venue":"JMIR Medical Education","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":27,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Preprint; Computer science; Natural language processing; Psychology; Medical education; Medicine; 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":["metaresearch","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.005181074,0.000152095,0.0003648798,0.0001389157,0.0001202652,0.00006257776,0.0003770672,0.0002489658,0.0009187789],"category_scores_gemma":[0.01797056,0.00009865672,0.0001357413,0.0005863709,0.000210757,0.0001589099,0.0001322352,0.0008406704,0.0001023433],"about_ca_system_candidate":true,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001713957,"about_ca_system_score_gemma":0.007231049,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0007866418,"about_ca_topic_score_gemma":0.0004059096,"domain_scores_codex":[0.9959952,0.0004242062,0.001128689,0.0004724036,0.001692733,0.0002867466],"domain_scores_gemma":[0.9965393,0.001611199,0.0001113537,0.0006747506,0.0003804384,0.0006829465],"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":[0.0001484007,0.01469612,0.1334562,0.0006802068,0.0001458504,0.00003124889,0.1795984,0.00001042972,0.0001546892,0.003861541,0.01409433,0.6531225],"study_design_scores_gemma":[0.0004727164,0.002380751,0.495992,0.003859567,0.0003236625,0.00004360147,0.4019577,0.08947871,0.000288659,0.003012343,0.001797213,0.0003930198],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9428059,0.0006667383,0.004999496,0.04780532,0.001814517,0.001566186,0.000004659615,0.00006849231,0.0002686541],"genre_scores_gemma":[0.9935627,0.00003881674,0.0005670749,0.004408839,0.0009201739,0.0003765279,0.00002615466,0.00002197535,0.00007774656],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6527295,"threshold_uncertainty_score":0.9999945,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2698714885083644,"score_gpt":0.6452357604818295,"score_spread":0.3753642719734651,"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."}}