{"id":"W4392605825","doi":"10.2196/55988","title":"Assessing the Alignment of Large Language Models With Human Values for Mental Health Integration: Cross-Sectional Study Using Schwartz’s Theory of Basic Values","year":2024,"lang":"en","type":"article","venue":"JMIR Mental Health","topic":"Mental Health via Writing","field":"Psychology","cited_by":47,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Psychology; Value (mathematics); Social psychology; Population; Mental health; Demography; Psychiatry; Sociology; Statistics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01133423,0.000309012,0.0003300898,0.001134978,0.001192811,0.001777244,0.0005412481,0.000687268,0.002242905],"category_scores_gemma":[0.02220857,0.0006953742,0.0006535414,0.0006834919,0.001328406,0.001712781,0.001612497,0.001804817,0.0003854638],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00091687,"about_ca_system_score_gemma":0.001126781,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004164126,"about_ca_topic_score_gemma":0.00551959,"domain_scores_codex":[0.9969969,0.001725043,0.0001912645,0.0003228274,0.000574045,0.0001899398],"domain_scores_gemma":[0.9858382,0.005576762,0.004120807,0.001369678,0.002174006,0.0009205901],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00005123913,0.0006052376,0.9798723,0.00002189282,0.00004807502,0.00006685678,0.01410066,0.00009754564,0.0001659166,0.0002467755,0.000252611,0.004470633],"study_design_scores_gemma":[0.00001528493,0.001055313,0.9661519,0.00005902656,0.00005252495,0.0004089861,0.02822407,0.00165575,0.0003346446,0.0004164716,0.001590219,0.00003592475],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9991358,0.00003853547,0.0003713163,0.00004737539,0.000003553798,0.00004631384,0.00003249685,0.000002076473,0.0003225621],"genre_scores_gemma":[0.9988081,0.00005207881,0.0006856855,0.00004090056,0.00000302566,0.00009959051,0.00008516461,0.000003644784,0.0002218278],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01133423,"threshold_uncertainty_score":0.05994183,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1035796775665596,"score_gpt":0.5112289900193032,"score_spread":0.4076493124527437,"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."}}