{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004956043,0.0003033618,0.0005149095,0.0002079769,0.001222959,0.0001182105,0.0002518123,0.00007902085,0.0002204108],"category_scores_gemma":[0.000005351651,0.0002214718,0.0001407886,0.0003037168,0.0002159671,0.0004314828,0.0001027652,0.0003291413,0.000004510969],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001103393,"about_ca_system_score_gemma":0.0004612397,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001285253,"about_ca_topic_score_gemma":0.0002163347,"domain_scores_codex":[0.9956268,0.001024596,0.001294434,0.0006186769,0.0007319778,0.0007035249],"domain_scores_gemma":[0.9984825,0.0002498086,0.0006049723,0.0004198189,0.00006871916,0.0001742285],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"qualitative","study_design_scores_codex":[0.002039919,0.01002832,0.1130003,0.007035566,0.001044753,0.00002057342,0.6703345,0.000151886,0.004325936,0.1553601,0.003312308,0.03334586],"study_design_scores_gemma":[0.008947414,0.01119213,0.1951654,0.003935686,0.00007134627,0.0002363281,0.7591122,0.01168199,0.00209233,0.006506612,0.0002712429,0.0007873206],"study_design_candidate":"qualitative","study_design_consensus":"qualitative","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9872317,0.00255155,0.002745175,0.0004993586,0.0009018661,0.00488082,0.0005740222,0.00009764366,0.0005178062],"genre_scores_gemma":[0.9965681,0.000008969537,0.001527828,0.0006229487,0.0002418626,0.0003402227,0.0002302451,0.00007390483,0.0003859372],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1488535,"threshold_uncertainty_score":0.9406137,"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."}}