MentalGPT: Harnessing AI for Compassionate Mental Health Support
Abstract: In this paper, we introduce MentalGPT, fine-tuned large language models (LLMs) designed to offer accessible, empathetic, and effective mental health support by efficient instruction fine-tuning. We demonstrate our fine-tuning process and the model's real-world performance, showcasing MentalGPT's potential to enhance mental health support services. Our research contributes valuable datasets for further research, introduces comprehensive metrics for evaluating mental health language models, and demonstrates that MentalGPT outperforms existing LLMs of the same size in the field. Extensive testing confirms our framework significantly enhances foundational LLMs, establishing MentalGPT as a promising tool for expanding accessible mental health support and reducing stigma around seeking help.
Paper Type: Long
Research Area: NLP Applications
Research Area Keywords: Dialogue and Interactive Systems, NLP Applications, Question Answering
Contribution Types: NLP engineering experiment, Approaches to low-resource settings, Data resources
Languages Studied: English
Submission Number: 649
Loading