While confidence estimation is a promising direction for mitigating hallucinations in Large Language Models (LLMs), current research overwhelmingly focuses on single-turn settings. The dynamics of model confidence in multi-turn conversations, where context accumulates and ambiguity is progressively resolved, remain largely unexplored. This work presents the first systematic study of confidence estimation in multi-turn interactions, establishing a formal evaluation framework grounded in two key desiderata: per-turn calibration and monotonicity of confidence as more information becomes available. To facilitate this, we introduce novel metrics, including a length-normalized Expected Calibration Error (InfoECE), and a new "Hinter-Guesser" paradigm for generating controlled evaluation datasets. Our experiments reveal that widely-used confidence techniques struggle with calibration and monotonicity in multi-turn dialogues. In contrast, a novel logit-based probe we introduce, P(Sufficient), proves comparatively more effective, robustly tracking evidence accumulation and distinguishing it from conversational filler. Our work provides a foundational methodology for developing more reliable and trustworthy conversational agents.
@article{arxiv.2601.02179,
title = {Confidence Estimation for LLMs in Multi-turn Interactions},
author = {Caiqi Zhang and Ruihan Yang and Xiaochen Zhu and Chengzu Li and Tiancheng Hu and Yijiang River Dong and Deqing Yang and Nigel Collier},
journal= {arXiv preprint arXiv:2601.02179},
year = {2026}
}