{"id":"W4410711823","doi":"10.2196/75103","title":"Evaluating and Improving Syndrome Differentiation Thinking Ability in Large Language Models: Method Development Study","year":2025,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Traditional Chinese Medicine Studies","field":"Medicine","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Preprint; Computer science; Natural language processing; World Wide Web","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":[],"consensus_categories":[],"category_scores_codex":[0.002795133,0.0001774207,0.0004683429,0.0002736904,0.0001407322,0.00002195018,0.000107429,0.00009778459,0.00003721062],"category_scores_gemma":[0.001050582,0.0001299299,0.00003327258,0.0003270532,0.00004733493,0.0002061456,0.0002317089,0.0005161866,0.000002419604],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001785651,"about_ca_system_score_gemma":0.0002976419,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002762734,"about_ca_topic_score_gemma":0.00008011278,"domain_scores_codex":[0.9974671,0.0001026157,0.000902559,0.0001590513,0.001109862,0.0002588303],"domain_scores_gemma":[0.9990854,0.0003830412,0.0001239223,0.0001938946,0.00008668822,0.0001270516],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"observational","study_design_scores_codex":[0.0002129,0.002771397,0.365117,0.006498816,0.0005342925,0.0001782285,0.3954657,0.00009756861,0.0001152861,0.00257841,0.000122376,0.226308],"study_design_scores_gemma":[0.004428588,0.000226131,0.7151683,0.0008516187,0.00008066653,0.0000501088,0.02909246,0.2490357,0.00001654421,0.0009038656,0.00001337804,0.000132665],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9728438,0.0001556289,0.02458906,0.0004180289,0.0001009383,0.001078806,0.000002590787,0.00008161635,0.0007294798],"genre_scores_gemma":[0.9749686,0.000005104633,0.02406714,0.0006788422,0.00003147703,0.0001678905,0.00003575434,0.000008021005,0.00003717751],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3663732,"threshold_uncertainty_score":0.529839,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03897205438098617,"score_gpt":0.3985427546078242,"score_spread":0.359570700226838,"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."}}