CALM-IT: Generating Realistic Long-Form Motivational Interviewing Dialogues with Dual-Actor Conversational Dynamics Tracking

TL;DR

CALM-IT is a novel framework that models dual-actor dynamics to generate realistic, goal-oriented long-form Motivational Interviewing dialogues, improving coherence, stability, and therapeutic alignment over extended interactions.

Contribution

It introduces a bidirectional state-space model for MI dialogues, explicitly capturing evolving conversational states to enhance long-term dialogue quality.

Findings

Outperforms baselines in Effectiveness and Goal Alignment

Maintains stability over longer conversations

Achieves highest client acceptance rate of 64.3%

Abstract

Large Language Models (LLMs) are increasingly used in mental health-related settings, yet they struggle to sustain realistic, goal-directed dialogue over extended interactions. While LLMs generate fluent responses, they optimize locally for the next turn rather than maintaining a coherent model of therapeutic progress, leading to brittleness and long-horizon drift. We introduce CALM-IT, a framework for generating and evaluating long-form Motivational Interviewing (MI) dialogues that explicitly models dual-actor conversational dynamics. CALM-IT represents therapist-client interaction as a bidirectional state-space process, in which both agents continuously update inferred alignment, mental states, and short-term goals to guide strategy selection and utterance generation. Across large-scale evaluations, CALM-IT consistently outperforms strong baselines in Effectiveness and Goal Alignment and remains substantially more stable as conversation length increases. Although CALM-IT initiates fewer therapist redirections, it achieves the highest client acceptance rate (64.3%), indicating more precise and therapeutically aligned intervention timing. Overall, CALM-IT provides evidence for modeling evolving conversational state being essential for generating high-quality long-form synthetic conversations.