Graph2Counsel: Clinically Grounded Synthetic Counseling Dialogue Generation from Client Psychological Graphs

2026-04-22

AI Summary

This study addresses the scarcity of authentic counseling dialogues due to privacy constraints and the limitations of existing synthetic data generation methods, which often neglect the structural dependencies among clients’ cognition, emotion, and behavior, resulting in psychologically inconsistent and low-quality conversations. To overcome this, the work introduces Client Psychological Graphs (CPGs) to construct a structured prompting pipeline that integrates Chain-of-Thought (CoT) reasoning with a multi-agent feedback mechanism, guiding large language models to generate clinically coherent and high-fidelity dialogues. The proposed approach ensures psychological consistency while enhancing realism and safety, significantly outperforming existing datasets in expert evaluations. Fine-tuning open-source models on 760 generated dialogues yields notable performance gains on both CounselingBench and CounselBench. The code and data are publicly released.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• 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

Abstract

Rising demand for mental health support has increased interest in using Large Language Models (LLMs) for counseling. However, adapting LLMs to this high-risk safety-critical domain is hindered by the scarcity of real-world counseling data due to privacy constraints. Synthetic datasets provide a promising alternative, but existing approaches often rely on unstructured or semi-structured text inputs and overlook structural dependencies between a client's cognitive, emotional, and behavioral states, often producing psychologically inconsistent interactions and reducing data realism and quality. We introduce Graph2Counsel, a framework for generating synthetic counseling sessions grounded in Client Psychological Graphs (CPGs) that encode relationships among clients' thoughts, emotions, and behaviors. Graph2Counsel employs a structured prompting pipeline guided by counselor strategies and CPG, and explores prompting strategies including CoT (Wei et al., 2022) and Multi-Agent Feedback (Li et al., 2025a). Graph2Counsel produces 760 sessions from 76 CPGs across diverse client profiles. In expert evaluation, our dataset outperforms prior datasets on specificity, counselor competence, authenticity, conversational flow, and safety, with substantial inter-annotator agreement (Krippendorff's $α$ = 0.70). Fine-tuning an open-source model on this dataset improves performance on CounselingBench (Nguyen et al., 2025) and CounselBench (Li et al., 2025b), showing downstream utility. We also make our code and data public.

Problem

• synthetic counseling dialogue
• psychological consistency
• client psychological graphs
• mental health support
• data scarcity

Innovation

• Client Psychological Graphs
• Synthetic Counseling Dialogue
• Structured Prompting
• Chain-of-Thought
• Multi-Agent Feedback