SAGE: A Strategy-Aware Graph-Enhanced Generation Framework For Online Counseling

2026-04-29

AI Summary

This work addresses the limited capacity of current general-purpose large language models to perform structured clinical reasoning essential for psychotherapy, particularly in effectively integrating psychological theory, real-time emotional cues, and intervention strategies. To bridge this gap, the authors propose a psychology-driven, graph-augmented generative framework that constructs a heterogeneous knowledge graph unifying dialogue dynamics and psychological theories. The framework employs a Next Strategy Classifier to select optimal therapeutic interventions and incorporates a graph-aware attention mechanism to guide the large language model in generating clinically appropriate responses. This approach uniquely combines theory-informed strategy selection with graph-enhanced prompting, significantly outperforming baseline methods in both automatic metrics and expert evaluations—especially in intervention accuracy and response quality—demonstrating its potential as an intelligent decision-support tool for high-stakes psychological counseling.

Abstract

Effective mental health counseling is a complex, theory-driven process requiring the simultaneous integration of psychological frameworks, real-time distress signals, and strategic intervention planning. This level of clinical reasoning is critical for safety and therapeutic effectiveness but is often missing in general-purpose Large Language Models (LLMs). We introduce SAGE (Strategy-Aware Graph-Enhanced), a novel framework designed to bridge the gap between structured clinical knowledge and generative AI. SAGE constructs a heterogeneous graph that unifies conversational dynamics with a psychologically grounded layer, explicitly anchoring interactions in a theory-driven lexicon. Our architecture first employs a Next Strategy Classifier to identify the optimal therapeutic intervention. Subsequently, a Graph-Aware Attention mechanism projects graph-derived structural signals into soft prompts, conditioning the LLM to generate responses that maintain clinical depth. Validated through both automated metrics and expert human evaluation, SAGE outperforms baselines in strategy prediction and recommended response quality. By providing actionable intervention recommendations, SAGE serves as a cutting-edge decision-support tool designed to augment human expertise in high-stakes crisis counseling.

Problem

• online counseling
• clinical reasoning
• therapeutic strategy
• large language models
• mental health

Innovation

• Strategy-Aware Generation
• Graph-Enhanced LLM
• Clinical Decision Support
• Heterogeneous Knowledge Graph
• Therapeutic Intervention Planning