Confidence Estimation for LLMs in Multi-turn Interactions
This work addresses a critical issue for downstream applications like autonomous agents and human-in-the-loop systems by providing the first systematic study in multi-turn settings, but it is incremental as it builds on existing confidence estimation techniques.
The paper tackles the problem of confidence estimation for Large Language Models in multi-turn conversations, where existing methods struggle with calibration and monotonicity, and proposes a logit-based probe called P(Sufficient) that achieves comparatively better performance, though the task remains unsolved.
While confidence estimation is a promising direction for mitigating hallucinations in Large Language Models (LLMs), current research dominantly 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. Reliable confidence estimation in multi-turn settings is critical for many downstream applications, such as autonomous agents and human-in-the-loop systems. 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. We propose P(Sufficient), a logit-based probe that achieves comparatively better performance, although the task remains far from solved. Our work provides a foundational methodology for developing more reliable and trustworthy conversational agents.