CALM-IT：用对话动态提升合成动机式访谈对话质量

View PDF HTML (experimental)

Abstract:Conversational dynamics describe how rapport, emotion, and goals change over a dialogue. These changes are important in therapeutic conversations, where they can influence when and how therapists intervene. However, large language model (LLM)-based synthetic therapeutic dialogue generators rarely track how these properties evolve together, which may limit the quality of longer conversations. We introduce CALM-IT, a dual-agent framework that tracks client states and therapist-side inferences to guide strategy selection and utterance generation. We evaluate synthetic Motivational Interviewing (MI) dialogues using automated Motivational Interviewing Treatment Integrity (MITI) ratings, conversational metrics, and comparisons across dialogue lengths, with ratings from four licensed psychologists providing support for key findings. Across 8,232 dialogues covering 686 grounded scenarios, four frameworks, and three target lengths, CALM-IT achieves the highest mean scores on all four MITI 4.2 global ratings and shows the smallest aggregate change in quality between short- and long-form conditions. CALM-IT also achieves the highest simulated client redirection acceptance rate (64.3\%), with its strongest redirections followed by increased change talk and reduced sustain talk. Ablation experiments show that removing conversational dynamics components reduces dialogue quality. Together, these results suggest that explicitly modeling conversational dynamics can improve the quality and consistency of long-form synthetic MI dialogues.

Comments:	55 pages
Subjects:	Computation and Language (cs.CL)
Cite as:	arXiv:2601.10085 [cs.CL]
	(or arXiv:2601.10085v3 [cs.CL] for this version)
	https://doi.org/10.48550/arXiv.2601.10085 arXiv-issued DOI via DataCite

Submission history

From: Viet Cuong Nguyen [view email]
[v1] Thu, 15 Jan 2026 05:21:54 UTC (1,923 KB)
[v2] Wed, 27 May 2026 05:06:18 UTC (3,155 KB)
[v3] Tue, 13 Oct 2026 20:27:41 UTC (3,157 KB)