PSY-STEP: Structuring Therapeutic Targets and Action Sequences for Proactive Counseling Dialogue Systems

2026-04-06

AI Summary

This work addresses the challenge that existing counseling dialogue systems struggle to effectively identify and intervene on automatic negative thoughts within cognitive behavioral therapy (CBT). To bridge this gap, the authors propose STEPPER, a novel framework that explicitly integrates structured representations of automatic thoughts into dialogue system design. STEPPER leverages a newly constructed STEP dataset to model the relationship between automatic thoughts and dynamic therapeutic action sequences, enabling the agent to proactively guide users in articulating their automatic thoughts and applying CBT-based interventions. Furthermore, the system incorporates preference learning grounded in simulated dialogues to refine its decision-making and empathetic responses. Experimental results demonstrate that STEPPER significantly outperforms strong baselines in clinical appropriateness, coherence, personalization, and perceived therapist competence, without inducing emotional distress in users.

Abstract

Cognitive Behavioral Therapy (CBT) aims to identify and restructure automatic negative thoughts pertaining to involuntary interpretations of events, yet existing counseling agents struggle to identify and address them in dialogue settings. To bridge this gap, we introduce STEP, a dataset that models CBT counseling by explicitly reflecting automatic thoughts alongside dynamic, action-level counseling sequences. Using this dataset, we train STEPPER, a counseling agent that proactively elicits automatic thoughts and executes cognitively grounded interventions. To further enhance both decision accuracy and empathic responsiveness, we refine STEPPER through preference learning based on simulated, synthesized counseling sessions. Extensive CBT-aligned evaluations show that STEPPER delivers more clinically grounded, coherent, and personalized counseling compared to other strong baseline models, and achieves higher counselor competence without inducing emotional disruption.

Problem

• Cognitive Behavioral Therapy
• automatic thoughts
• counseling dialogue systems
• therapeutic intervention
• proactive elicitation

Innovation

• automatic thoughts
• cognitive behavioral therapy
• proactive dialogue system
• preference learning
• therapeutic action sequences