{"id":"W4413448599","doi":"10.1101/2025.08.14.25332839","title":"Automated Evaluation of Large Language Model Response Concordance with Human Specialist Responses on Physician-to-Physician eConsult Cases","year":2025,"lang":"en","type":"preprint","venue":"medRxiv","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Concordance; Medicine; Family medicine; Internal medicine","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":[0.04123502,0.001398372,0.001028853,0.002201224,0.0007051932,0.002240192,0.001808062,0.002090254,0.003337979],"category_scores_gemma":[0.15463,0.0003868865,0.00093522,0.001047611,0.001230173,0.001472783,0.003147817,0.001441159,0.002089608],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001254428,"about_ca_system_score_gemma":0.002006616,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002619215,"about_ca_topic_score_gemma":0.003644814,"domain_scores_codex":[0.9343339,0.05018969,0.003151562,0.006366743,0.005214215,0.0007439903],"domain_scores_gemma":[0.7888874,0.1749765,0.008652911,0.01280897,0.01252478,0.002149437],"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":[0.0107562,0.003540033,0.2003921,0.005067012,0.001645092,0.002145705,0.01902133,0.09612848,0.05234263,0.005797362,0.05031973,0.5528443],"study_design_scores_gemma":[0.001534919,0.002701886,0.09037827,0.0004753392,0.0003730165,0.001512422,0.003920419,0.8076434,0.06041598,0.007987152,0.02268533,0.0003717707],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8638495,0.001269582,0.113883,0.001311285,0.0004546912,0.001329024,0.003734748,0.008341052,0.005827181],"genre_scores_gemma":[0.9176102,0.0001581936,0.07345682,0.0004327872,0.0001275345,0.0006432519,0.005714342,0.0003181206,0.001538742],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.04123502,"threshold_uncertainty_score":0.2180742,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.160679898729795,"score_gpt":0.483121711576446,"score_spread":0.322441812846651,"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."}}