{"id":"W4414994693","doi":"10.1007/978-3-032-01399-6_5","title":"Evaluating Prompt Engineering Techniques for Accuracy and Confidence Elicitation in Medical LLMs","year":2025,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Quality and Safety in Healthcare","field":"Health Professions","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Ontario Tech University","funders":"","keywords":"Correctness; Low Confidence; Confidence interval; Calibration; Brier score; Overconfidence effect","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":[],"consensus_categories":[],"category_scores_codex":[0.01309448,0.0008927585,0.0006746764,0.001388457,0.0003803339,0.001862752,0.001130001,0.001496887,0.004125473],"category_scores_gemma":[0.1345149,0.0003595121,0.0003996009,0.0009484654,0.0003984171,0.002092217,0.001928027,0.001047528,0.001388783],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006964646,"about_ca_system_score_gemma":0.001088547,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001155479,"about_ca_topic_score_gemma":0.001585364,"domain_scores_codex":[0.9866353,0.008160632,0.001330669,0.0009195692,0.00259668,0.0003570746],"domain_scores_gemma":[0.667641,0.3087147,0.00640006,0.004829321,0.01139994,0.00101497],"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":[0.007981949,0.0009950948,0.02324763,0.001975256,0.0001764325,0.0001914931,0.003630912,0.01493184,0.04591949,0.002448602,0.00385746,0.8946439],"study_design_scores_gemma":[0.001122537,0.007683632,0.05954131,0.001036938,0.0008178326,0.001172467,0.005220452,0.7042041,0.1902199,0.01434635,0.01435524,0.0002791672],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.4983908,0.001683428,0.4846031,0.0005637154,0.0001924496,0.0009652498,0.001056659,0.007676676,0.004868002],"genre_scores_gemma":[0.748679,0.0003766483,0.2476814,0.0001871359,0.0000619625,0.0003773362,0.0007642964,0.0002815606,0.001590654],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01309448,"threshold_uncertainty_score":0.06925106,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1194715902993269,"score_gpt":0.4954619294699572,"score_spread":0.3759903391706302,"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."}}