{"id":"W4403220723","doi":"10.1148/radiol.241489","title":"Retrieval-Augmented Generation for Large Language Models in Radiology: Another Leap Forward in Board Examination Performance","year":2024,"lang":"en","type":"article","venue":"Radiology","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":22,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto General Hospital","funders":"","keywords":"Medicine; Medical physics; Radiology","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009345072,0.0001028329,0.0002360221,0.0003439194,0.00003720781,0.000009107552,0.00004836396,0.0002328732,0.00005939068],"category_scores_gemma":[0.0001586797,0.00009439212,0.0000418338,0.0002792587,0.00003076598,0.0001484065,0.000007930325,0.0001971114,0.00002555666],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003224357,"about_ca_system_score_gemma":0.0001412228,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001144451,"about_ca_topic_score_gemma":0.0003848874,"domain_scores_codex":[0.9988233,0.0001174561,0.0003806134,0.000300738,0.00006705122,0.0003108695],"domain_scores_gemma":[0.9995641,0.0001629173,0.00003822682,0.0001484654,0.00005122722,0.00003509652],"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":[0.002347849,0.0007870297,0.1194065,0.002221068,0.0001646139,0.0001643517,0.06870147,0.006176204,0.2001406,0.02054447,0.01731126,0.5620346],"study_design_scores_gemma":[0.0004091815,0.0007579394,0.01445031,0.000168024,0.00002783358,0.00009172266,0.001334757,0.959463,0.01863798,0.0009363379,0.003542795,0.000180094],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9800175,0.002698157,0.01356162,0.001891079,0.0007797046,0.0007220316,0.000007991837,0.00004993344,0.0002719334],"genre_scores_gemma":[0.9966958,0.0006120707,0.0004369412,0.0006226907,0.0005795127,0.0001177884,0.000222325,0.00002069865,0.000692212],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9532868,"threshold_uncertainty_score":0.3849199,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1076246466781111,"score_gpt":0.3942330226564129,"score_spread":0.2866083759783018,"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."}}