Psy-Copilot: Visual Chain of Thought for Counseling

2025-03-05

AI Summary

The opaque reasoning processes of large language models (LLMs) in psychotherapy undermine trustworthiness and efficacy in human–AI collaboration. Method: We propose Psy-COT—the first visualization framework for chain-of-thought reasoning tailored to counseling—representing LLM inference paths explicitly via a graph-structured knowledge map. Building upon this, we design Psy-Copilot, an AI co-pilot integrating retrieval-augmented generation (RAG), dialogue state tracking, and multimodal interaction to support therapeutic strategy matching, similar-session retrieval, and response provenance tracing. A semi-structured conversation annotation schema enables interpretable, traceable, and intervenable collaborative assistance. Contribution/Results: Empirical evaluation on an open-source platform demonstrates significant improvements in clinicians’ trust in and adoption rate of AI-generated recommendations, establishing a novel “augmentation-over-replacement” paradigm for human–AI collaboration in mental health care.

Technology Category

• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction
• Machine Learning: Large Multimodal Models (LMMs)
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Large language models (LLMs) are becoming increasingly popular in the field of psychological counseling. However, when human therapists work with LLMs in therapy sessions, it is hard to understand how the model gives the answers. To address this, we have constructed Psy-COT, a graph designed to visualize the thought processes of LLMs during therapy sessions. The Psy-COT graph presents semi-structured counseling conversations alongside step-by-step annotations that capture the reasoning and insights of therapists. Moreover, we have developed Psy-Copilot, which is a conversational AI assistant designed to assist human psychological therapists in their consultations. It can offer traceable psycho-information based on retrieval, including response candidates, similar dialogue sessions, related strategies, and visual traces of results. We have also built an interactive platform for AI-assisted counseling. It has an interface that displays the relevant parts of the retrieval sub-graph. The Psy-Copilot is designed not to replace psychotherapists but to foster collaboration between AI and human therapists, thereby promoting mental health development. Our code and demo are both open-sourced and available for use.

Problem

• Visualize LLM thought processes in therapy sessions
• Assist therapists with traceable psycho-information retrieval
• Promote AI-human collaboration in psychological counseling

Innovation

• Visualizes LLM reasoning with Psy-COT graph
• Develops Psy-Copilot for AI-assisted counseling
• Creates interactive platform for therapy collaboration