{"id":"W4415620452","doi":"10.2196/81807","title":"Large Language Models for the National Radiological Technologist Licensure Examination in Japan: Cross-Sectional Comparative Benchmarking and Evaluation of Model-Generated Items Study","year":2025,"lang":"en","type":"article","venue":"JMIR Medical Education","topic":"Innovations in Medical Education","field":"Medicine","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"CLARITY; Formative assessment; Benchmarking; Licensure; Workflow; Blueprint; Radiological weapon; Quality (philosophy)","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.02473573,0.0008182485,0.0008458606,0.002592158,0.0008669788,0.00119404,0.001182636,0.0008485246,0.001169362],"category_scores_gemma":[0.06345153,0.0006405282,0.002352444,0.001560444,0.001001337,0.002159813,0.002366729,0.0009821914,0.000522423],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002453379,"about_ca_system_score_gemma":0.001945088,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008437879,"about_ca_topic_score_gemma":0.007888644,"domain_scores_codex":[0.9887566,0.005718255,0.00145696,0.001211108,0.00241914,0.0004379251],"domain_scores_gemma":[0.9560122,0.01691778,0.00915876,0.005928349,0.01016647,0.001816338],"domain_codex":null,"domain_gemma":"evaluation","domain_candidate":"evaluation","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.000321421,0.0004486617,0.9890072,0.00006239553,0.0001577178,0.0001014442,0.002478651,0.0004428886,0.0002223013,0.00005762537,0.000256066,0.006443447],"study_design_scores_gemma":[0.00005916165,0.001397531,0.9882876,0.00005681528,0.0001827356,0.0002586452,0.003339575,0.005283143,0.0004201113,0.0001099374,0.000566752,0.00003804973],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9987785,0.00006956742,0.0007037692,0.0000262142,0.000005646674,0.00009286485,0.000109618,0.00001037806,0.0002033069],"genre_scores_gemma":[0.9985091,0.00004127815,0.0008231888,0.00001783907,0.000004091052,0.0001259971,0.00037446,0.00001070487,0.00009330134],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9752643,"threshold_uncertainty_score":0.1308166,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08605173198643123,"score_gpt":0.476747212296595,"score_spread":0.3906954803101638,"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."}}