PsychēChat: An Empathic Framework Focused on Emotion Shift Tracking and Safety Risk Analysis in Psychological Counseling

ACL ARR 2026 January Submission3948 Authors

Keywords: Mental Health Support, LLM Agent, Psychological Counseling, Data Synthesis

Abstract: Large language models (LLMs) have demonstrated notable advancements in psychological counseling. However, existing models generally do not explicitly model seekers' emotion shifts across counseling sessions, a core focus in classical psychological schools. Moreover, how to align counselor models' responses with these emotion shifts while proactively mitigating safety risks remains underexplored. To bridge these gaps, we propose $\textbf{PsychēChat}$, which explicitly integrates emotion shift tracking and safety risk analysis for psychological counseling. Specifically, we employ interactive role-playing to synthesize counselor--seeker dialogues, incorporating two modules: $\textbf{Emotion Management Module}$, to capture seekers' current emotions and emotion shifts; and $\textbf{Risk Control Module}$, to anticipate seekers' subsequent reactions and identify potential risks. Furthermore, we introduce two modeling paradigms. The $\textbf{Agent Mode}$ structures emotion management, risk control, and counselor responses into a collaborative multi-agent pipeline. The $\textbf{LLM Mode}$ integrates these stages into a unified chain-of-thought for end-to-end inference, balancing efficiency and performance. Extensive experiments, including interactive scoring, dialogue-level evaluation, and human assessment, demonstrate that PsychēChat outperforms existing methods for emotional insight and safety control.

Paper Type: Long

Research Area: Clinical and Biomedical Applications

Research Area Keywords: mental health, clinical dialogue systems, clinical decision support, regulatory and ethical considerations

Contribution Types: Model analysis & interpretability, NLP engineering experiment, Data resources

Languages Studied: Chinese

Submission Number: 3948