{"id":"W4414410969","doi":"10.1038/s44325-025-00083-5","title":"Fine-tuning LLMs in behavioral psychology for scalable health coaching","year":2025,"lang":"en","type":"article","venue":"npj Cardiovascular Health","topic":"Behavioral Health and Interventions","field":"Psychology","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institute of Arthritis and Musculoskeletal and Skin Diseases; School of Medicine, Stanford University; National Institutes of Health; National Institute on Aging; Imperial College London; American Heart Association; American Diabetes Association; University of Washington; National Heart, Lung, and Blood Institute; National Center for Advancing Translational Sciences; Canada Excellence Research Chairs, Government of Canada; Stanford Bio-X","keywords":"Transtheoretical model; Operationalization; Coaching; eHealth; Health psychology; Health coaching; Psychological intervention; Behavior change; Physical activity; Behavioural sciences","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004349618,0.000587741,0.0003141841,0.000579011,0.0003213238,0.001265642,0.001071057,0.0009210162,0.006268506],"category_scores_gemma":[0.01580108,0.0003902385,0.0007066079,0.0002604093,0.0008659703,0.001597214,0.00207838,0.001753,0.002269569],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001027941,"about_ca_system_score_gemma":0.001768632,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002298522,"about_ca_topic_score_gemma":0.003429798,"domain_scores_codex":[0.9978521,0.001213932,0.00008673465,0.0003571966,0.0003925481,0.00009751885],"domain_scores_gemma":[0.9922209,0.005910954,0.0003259758,0.0008927585,0.0003629386,0.0002865123],"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":[0.0007294646,0.002831915,0.01076971,0.001532841,0.0001630968,0.000124621,0.003043432,0.03324496,0.07568729,0.0388246,0.01276704,0.820281],"study_design_scores_gemma":[0.0008395683,0.002966987,0.02033954,0.001397684,0.0003391463,0.0004103151,0.001542694,0.5661083,0.04935787,0.1842839,0.1721354,0.0002785989],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05397768,0.000990205,0.9184406,0.003398353,0.000191152,0.0008480341,0.0003149367,0.009938626,0.01190047],"genre_scores_gemma":[0.3270384,0.0005582288,0.6638463,0.001628777,0.00008518996,0.001783892,0.0003106506,0.0005886579,0.004159996],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006268506,"threshold_uncertainty_score":0.02300328,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09591221042924133,"score_gpt":0.4696993880497813,"score_spread":0.37378717762054,"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."}}