This research creates multilingual datasets for mental health severity prediction across six languages, addressing a critical gap in non-English mental health applications.

• Adapts existing English mental health datasets into Greek, Turkish, French, Portuguese, German, and Finnish
• Enables comprehensive evaluation of LLMs for mental health support across multiple languages
• Investigates both challenges and potential in applying LLMs to mental health assessment
• Identifies risks and limitations to guide responsible implementation

Why it matters: This work lays the foundation for more inclusive, globally accessible mental health support systems powered by AI, while carefully considering the ethical implications and potential pitfalls of automated assessment.

Building Multilingual Datasets for Predicting Mental Health Severity through LLMs: Prospects and Challenges