# Psy-LLM: Scaling up Global Mental Health Psychological Services with AI-based Large Language Models

**Authors:** Tin Lai, Yukun Shi, Zicong Du, Jiajie Wu, Ken Fu, Yichao Dou, Ziqi Wang  
**arXiv:** [2307.11991](https://arxiv.org/abs/2307.11991)  
**Date:** 2023-07-22  
**Summary page:** https://www.summarizepaper.com/en/paper/2307.11991v2/

## Key points

- The demand for psychological counseling has increased significantly in recent years, especially with the global COVID-19 pandemic.
- Online psychological counseling has become the primary mode of delivering services.
- The authors propose a framework called Psy-LLM which leverages Large Language Models (LLMs) powered by artificial intelligence (AI) to provide question-answering assistance in psychological consultation settings.
- The Psy-LLM framework combines pre-trained LLMs with real-world professional Q&A from psychologists and extensively crawled psychological articles.
- It serves as a front-end tool for healthcare professionals enabling them to offer immediate responses and mindfulness activities to alleviate patient stress.
- It functions as a screening tool to identify urgent cases that require further assistance.
- Intrinsic metrics such as perplexity were used to evaluate the effectiveness of the Psy-LLM framework, along with extrinsic evaluation metrics involving human participants assessing response helpfulness, fluency, relevance, and logic.
- The results demonstrate that the Psy-LLM framework generates coherent and relevant answers to psychological questions.
- Using large language models can enhance mental health support through AI technologies, but there are potential benefits and limitations discussed in this article.
- The Psy-LLM framework aims to meet the increasing demand for mental health professionals while providing timely and accurate support to individuals seeking psychological counseling.

## Summary

The demand for psychological counseling has increased significantly in recent years, especially with the global COVID-19 pandemic which has amplified the need for timely and professional mental health support. To address this demand, online psychological counseling has become the primary mode of delivering services. In response to this trend, the authors propose a framework called Psy-LLM which leverages Large Language Models (LLMs) powered by artificial intelligence (AI) to provide question-answering assistance in psychological consultation settings. The Psy-LLM framework combines pre-trained LLMs with real-world professional Q&A from psychologists and extensively crawled psychological articles. It serves as a front-end tool for healthcare professionals enabling them to offer immediate responses and mindfulness activities to alleviate patient stress. Moreover, it functions as a screening tool to identify urgent cases that require further assistance. To evaluate the effectiveness of the Psy-LLM framework intrinsic metrics such as perplexity were used along with extrinsic evaluation metrics that involved human participants assessing response helpfulness, fluency, relevance and logic. The results demonstrate that the Psy-LLM framework generates coherent and relevant answers to psychological questions. This article discusses both the potential benefits and limitations of using large language models to enhance mental health support through AI technologies. By scaling up global mental health psychological services the Psy-LLM framework aims to meet the increasing demand for mental health professionals while providing timely and accurate support to individuals seeking psychological counseling.