{"id":"W4409155725","doi":"10.2196/65726","title":"Using a Hybrid of AI and Template-Based Method in Automatic Item Generation to Create Multiple-Choice Questions in Medical Education: Hybrid AIG","year":2025,"lang":"en","type":"article","venue":"JMIR Formative Research","topic":"Psychometric Methodologies and Testing","field":"Decision Sciences","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Summative assessment; Formative assessment; Field (mathematics); Artificial intelligence; Template; Subject-matter expert; Machine learning; Expert system; Programming language; Psychology","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"],"consensus_categories":[],"category_scores_codex":[0.02163142,0.0001314284,0.0004226134,0.004245714,0.0002030184,0.0001754948,0.0005337601,0.00007575114,0.000104693],"category_scores_gemma":[0.2585786,0.0001038724,0.0000476689,0.00620178,0.0001509474,0.0004269293,0.0002792791,0.0005700702,0.000007422577],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002398824,"about_ca_system_score_gemma":0.001366812,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001205304,"about_ca_topic_score_gemma":0.0002950377,"domain_scores_codex":[0.9928897,0.003551953,0.001105308,0.0004319079,0.001618701,0.0004024126],"domain_scores_gemma":[0.9240765,0.07411651,0.0001927843,0.0004756999,0.0009307108,0.0002077783],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00009090679,0.0006419345,0.4048661,0.0002133647,0.00001881251,0.00001382586,0.002312394,0.002094107,0.001546804,0.001264183,0.009915644,0.5770219],"study_design_scores_gemma":[0.0006548719,0.00009603654,0.2205383,0.0004994066,0.000002000394,0.000009969572,0.000738495,0.7698631,0.001778556,0.005042576,0.0006808665,0.00009584054],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8510514,0.0001516196,0.1447124,0.002420234,0.0002319108,0.0006771929,0.00001154268,0.00001822013,0.0007254698],"genre_scores_gemma":[0.9496148,0.000007629405,0.04978179,0.0002910041,0.00004870101,0.0001768823,0.000005271119,0.000006244642,0.00006762998],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.767769,"threshold_uncertainty_score":0.7497063,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.5171630694068262,"score_gpt":0.6303167085209163,"score_spread":0.11315363911409,"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."}}