Beyond Empathy: Integrating Diagnostic and Therapeutic Reasoning with Large Language Models for Mental Health Counseling

TL;DR

PsyLLM is a novel large language model that integrates diagnostic and therapeutic reasoning for mental health counseling, trained on real-world Reddit data and evaluated with a new clinical quality benchmark.

Contribution

This work introduces PsyLLM, the first LLM to systematically combine diagnostic standards and multiple therapeutic modalities for mental health support.

Findings

PsyLLM outperforms baseline models on clinical counseling quality metrics.

Automated data synthesis from Reddit enables realistic training data.

The model incorporates DSM/ICD standards and various therapy frameworks.

Abstract

Large language models (LLMs) hold significant potential for mental health support, capable of generating empathetic responses and simulating therapeutic conversations. However, existing LLM-based approaches often lack the clinical grounding necessary for real-world psychological counseling, particularly in explicit diagnostic reasoning aligned with standards like the DSM/ICD and incorporating diverse therapeutic modalities beyond basic empathy or single strategies. To address these critical limitations, we propose PsyLLM, the first large language model designed to systematically integrate both diagnostic and therapeutic reasoning for mental health counseling. To develop PsyLLM, we design a novel automated data synthesis pipeline that processes real-world mental health posts collected from Reddit, where users frequently share psychological distress and seek community support. This pipeline processes real-world mental health posts, generates multi-turn dialogue structures, and leverages LLMs guided by international diagnostic standards (e.g., DSM/ICD) and multiple therapeutic frameworks (e.g., CBT, ACT, psychodynamic) to simulate detailed clinical reasoning processes. Rigorous multi-dimensional filtering ensures the generation of high-quality, clinically aligned dialogue data. In addition, we introduce a new benchmark and evaluation protocol, assessing counseling quality across four key dimensions. Our experiments demonstrate that PsyLLM significantly outperforms state-of-the-art baseline models on this benchmark. The model weights and dataset have been publicly released at https://github.com/Emo-gml/PsyLLM.