{"id":"W4406385218","doi":"10.1089/fpsam.2024.0206","title":"Comparative Performance of the Leading Large Language Models in Answering Complex Rhinoplasty Consultation Questions","year":2025,"lang":"en","type":"article","venue":"Facial Plastic Surgery & Aesthetic Medicine","topic":"Nasal Surgery and Airway Studies","field":"Medicine","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; University of Ottawa","funders":"","keywords":"Rhinoplasty; Natural language processing; Computer science; Linguistics; Psychology; Medicine; Nose; Philosophy; Surgery","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":[0.0005689756,0.0002108604,0.0008719107,0.0003387307,0.0001772143,0.000004779357,0.00008313775,0.00007067432,0.00006090445],"category_scores_gemma":[0.001167441,0.0001432492,0.0000982021,0.0008764188,0.0003899873,0.00008788089,0.00005064699,0.0002756827,0.000009597718],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007073498,"about_ca_system_score_gemma":0.0001708299,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008098077,"about_ca_topic_score_gemma":0.00009912201,"domain_scores_codex":[0.9983172,0.0001154825,0.0006608896,0.000241534,0.0003375077,0.0003273333],"domain_scores_gemma":[0.9974398,0.001962444,0.0001718512,0.0002104091,0.0001477684,0.00006769769],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.001918967,0.0008767023,0.9110132,0.0009693044,0.0002319541,0.00008668975,0.02293689,0.007255962,0.01923247,0.003376201,0.007692392,0.02440924],"study_design_scores_gemma":[0.004908773,0.0003934214,0.894803,0.01022924,0.0004024041,0.00009253262,0.009945837,0.06585682,0.004029355,0.0002089667,0.008596262,0.0005333861],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9938522,0.0004813914,0.0007258214,0.0007705363,0.0009254023,0.0003266643,0.00002081267,0.00004337508,0.002853808],"genre_scores_gemma":[0.9989749,0.000123795,0.00003254938,0.0002229345,0.0000499936,0.00002734271,0.00003120804,0.000009174107,0.0005281274],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.05860086,"threshold_uncertainty_score":0.5841531,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04048520541200501,"score_gpt":0.3090169258263897,"score_spread":0.2685317204143847,"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."}}