{"id":"W4395689022","doi":"10.2196/56569","title":"The Role of Humanization and Robustness of Large Language Models in Conversational Artificial Intelligence for Individuals With Depression: A Critical Analysis","year":2024,"lang":"en","type":"article","venue":"JMIR Mental Health","topic":"Mental Health via Writing","field":"Psychology","cited_by":44,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Psychology; Robustness (evolution); Artificial intelligence; Computer science; Psychotherapist; Cognitive psychology; Biology","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":[0.0393341,0.0004299958,0.0003323409,0.001668756,0.001271744,0.005900269,0.001684807,0.001572169,0.002115976],"category_scores_gemma":[0.1602854,0.0003916509,0.0004132062,0.0005928995,0.008604216,0.0072311,0.003683519,0.002541902,0.0003080489],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002528077,"about_ca_system_score_gemma":0.001796062,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002110702,"about_ca_topic_score_gemma":0.001962478,"domain_scores_codex":[0.9763969,0.01850079,0.001068907,0.001289962,0.002247145,0.000496466],"domain_scores_gemma":[0.7607431,0.2115557,0.007599189,0.009631915,0.008907262,0.001562782],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"qualitative","study_design_scores_codex":[0.001081722,0.0003403831,0.09982725,0.003740679,0.000376243,0.001486085,0.09000211,0.01751093,0.007010293,0.2388148,0.01502227,0.5247872],"study_design_scores_gemma":[0.0001043092,0.00119905,0.09267204,0.007016285,0.0005961474,0.004602847,0.0887242,0.1767736,0.02306695,0.4299034,0.1748714,0.0004697618],"study_design_candidate":"qualitative","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.4666976,0.04629094,0.2013038,0.2170518,0.001269687,0.0006627327,0.0004314946,0.0007321441,0.06555995],"genre_scores_gemma":[0.9752879,0.002621151,0.01907432,0.001876546,0.0002916365,0.0001213642,0.00004731464,0.00006371205,0.0006162056],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.0393341,"threshold_uncertainty_score":0.2080211,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03596799441104494,"score_gpt":0.4235083774749664,"score_spread":0.3875403830639215,"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."}}