The paper converts single-label mental health datasets into multi-label ones by having LLMs annotate additional disorders, producing SPAADE-DR, but the main evaluation validates models against labels the models themselves generated.
Mental health discourse on reddit: Self-disclosure, social support, and anonymity
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.AI 1years
2024 1verdicts
REJECT 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Automated Multi-Label Annotation for Mental Health Illnesses Using Large Language Models
The paper converts single-label mental health datasets into multi-label ones by having LLMs annotate additional disorders, producing SPAADE-DR, but the main evaluation validates models against labels the models themselves generated.