TheraMind: A Strategic and Adaptive Agent for Longitudinal Psychological Counseling

2025-10-29

AI Summary

Contemporary large language models for psychological counseling suffer from weak affective understanding, poor strategy adaptability, insufficient multi-turn intervention capability, and absence of cross-session long-term memory—limiting their clinical continuity. This paper introduces TheraMind, a novel framework featuring a dual-loop architecture: an inner loop enables real-time affective perception and intra-session strategy execution, while an outer loop leverages long-term memory and therapeutic feedback to perform cross-session treatment planning and dynamic optimization. TheraMind integrates affective state modeling, cross-session memory mechanisms, interpretable strategy selection, and a closed-loop evaluation module. Evaluated in a high-fidelity clinical simulation environment, TheraMind significantly outperforms existing approaches on key metrics—including multi-turn coherence, intervention flexibility, and therapeutic coordination—thereby establishing the first AI-based psychological intervention agent exhibiting strategic reasoning, adaptive behavior, and longitudinal continuity.

Technology Category

• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Multiagent Systems: Adversarial Agents
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• User Modeling, Personalization and Recommendation: Psychology-informed user models and recommender systems
• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Large language models (LLMs) in psychological counseling have attracted increasing attention. However, existing approaches often lack emotional understanding, adaptive strategies, and the use of therapeutic methods across multiple sessions with long-term memory, leaving them far from real clinical practice. To address these critical gaps, we introduce TheraMind, a strategic and adaptive agent for longitudinal psychological counseling. The cornerstone of TheraMind is a novel dual-loop architecture that decouples the complex counseling process into an Intra-Session Loop for tactical dialogue management and a Cross-Session Loop for strategic therapeutic planning. The Intra-Session Loop perceives the patient's emotional state to dynamically select response strategies while leveraging cross-session memory to ensure continuity. Crucially, the Cross-Session Loop empowers the agent with long-term adaptability by evaluating the efficacy of the applied therapy after each session and adjusting the method for subsequent interactions. We validate our approach in a high-fidelity simulation environment grounded in real clinical cases. Extensive evaluations show that TheraMind outperforms other methods, especially on multi-session metrics like Coherence, Flexibility, and Therapeutic Attunement, validating the effectiveness of its dual-loop design in emulating strategic, adaptive, and longitudinal therapeutic behavior. The code is publicly available at https://0mwwm0.github.io/TheraMind/.

Problem

• Addresses emotional understanding gaps in AI counseling
• Develops adaptive therapeutic strategies across multiple sessions
• Enables long-term therapy continuity through dual-loop architecture

Innovation

• Dual-loop architecture decouples counseling process
• Intra-Session Loop manages dialogue and emotional states
• Cross-Session Loop enables long-term therapeutic adaptation