A Reinforcement Learning-Based Facilitator for Simulated Group Motivational Interviewing

Researchers from Sorbonne Université and CNRS have developed the first reinforcement learning-based dialogue manager for simulated group motivational interviewing, using a discrete Soft-Actor-Critic policy to select therapist acts and LLM-generated utterances. The system, presented at SIGDIAL 2026, outperformed four LLM-based baselines by generating more directive acts and adapting to participant profiles, particularly in groups with an open-to-change participant.

A Reinforcement Learning-Based Facilitator for Simulated Group Motivational Interviewing https://aclanthology.org/2026.sigdial-1.61.pdf Alafate Abulimiti /people/alafate-abulimiti/unverified/ , Vladislav Maraev /people/vladislav-maraev/unverified/ , Agnès Helme-Guizon /people/agnes-helme-guizon/unverified/ , Catherine Pelachaud /people/catherine-pelachaud/ Abstract Motivational Interviewing MI is a widely validated approach to behavior change, but existing virtual MI agents operate only in one-on-one settings, ignoring the cost-effectiveness and peer-support dynamics of group MI. We present a simulation environment and reinforcement learning RL based dialogue manager for group MI, in which a discrete Soft-Actor-Critic SAC policy selects therapist dialogue acts and a large language model generates utterances, with two LLM-prompted patient agents as interlocutors. Our model supports adaptation to different participant profiles. We compared our dialogue manager with four LLM-based ones at the dialogue acts level. We observed that RL yields a significantly different therapist policy, which showed the tendency to generate more directive acts and adapt to varying group compositions. Participant profile adaptation was the strongest in groups containing an open-to-change participant.- Anthology ID: - 2026.sigdial-1.61 - Volume: Proceedings of the 27th Annual Meeting of the Special Interest Group on Discourse and Dialogue /volumes/2026.sigdial-1/ - Month: - August - Year: - 2026 - Address: - Atlanta, Georgia, USA - Editors: Jinho D. Choi /people/jinho-d-choi/ , Yun-Nung Chen /people/yun-nung-chen/ , Kotaro Funakoshi /people/kotaro-funakoshi/ , Ali Emami /people/ali-emami/ - Venue: SIGDIAL /venues/sigdial/ - SIG: SIGDIAL /sigs/sigdial/ - Publisher: - Association for Computational Linguistics - Note: - Pages: - 871–890 - Language: - URL: https://aclanthology.org/2026.sigdial-1.61/ https://aclanthology.org/2026.sigdial-1.61/ - DOI: - Cite ACL : - Alafate Abulimiti, Vladislav Maraev, Agnès Helme-Guizon, and Catherine Pelachaud. 2026. A Reinforcement Learning-Based Facilitator for Simulated Group Motivational Interviewing https://aclanthology.org/2026.sigdial-1.61/ . In Proceedings of the 27th Annual Meeting of the Special Interest Group on Discourse and Dialogue , pages 871–890, Atlanta, Georgia, USA. Association for Computational Linguistics. - Cite Informal : A Reinforcement Learning-Based Facilitator for Simulated Group Motivational Interviewing https://aclanthology.org/2026.sigdial-1.61/ Abulimiti et al., SIGDIAL 2026 - PDF: https://aclanthology.org/2026.sigdial-1.61.pdf https://aclanthology.org/2026.sigdial-1.61.pdf