MIThinker: A Plug-and-Play Policy-Optimized Thinker For Motivational Interviewing Counseling

2026-06-28

AI Summary

Current large language model–based motivational interviewing (MI) systems lack an internal reasoning mechanism explicitly aligned with MI counseling skills, limiting their effectiveness. This work proposes MIThinker—a lightweight reasoning framework that generates therapeutic thoughts to guide MI agents in strategic decision-making and response generation. We further introduce AugR1-MI, a novel automated data construction method that retroactively derives chains of thought from counselor responses. By integrating supervised fine-tuning with reinforcement learning in a two-stage training pipeline, the resulting MindfulMI agent achieves state-of-the-art performance in MI competency while reducing computational overhead by an order of magnitude.

Abstract

Reasoning large language models (LLMs) have recently made much progress in complex problem-solving, leveraging internal reasoning (or thought) to guide their solution generation. However, existing LLM-based counseling agents, including those using Motivational Interviewing (MI), generate responses without explicitly aligning thoughts with counseling techniques, limiting their effectiveness. We propose MIThinker, a lightweight thinking model that generates therapeutic thoughts to guide MI counseling agents in strategy selection and response generation. To overcome the lack of annotated thought data, we introduce AugR1-MI, an automated pipeline that reverse-engineers counselor's thoughts from observed responses. Through two-stage training combining supervised fine-tuning and reinforcement learning, MIThinker demonstrates improved theory-of-mind assessment and strategy alignment. Comprehensive evaluations show that MindfulMI, our agent leveraging MIThinker, achieves MI competency comparable to state-of-the-art systems with an order of magnitude less computation.

Problem

• Motivational Interviewing
• Large Language Models
• Counseling Agents
• Reasoning Alignment
• Therapeutic Thoughts

Innovation

• MIThinker
• Motivational Interviewing
• reasoning LLMs
• AugR1-MI
• plug-and-play thinker