Uncertainty Propagation on LLM Agent

Publication Date: 7/29/2025

Event: The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Reference: pp. 6064–6073, 2025

Authors: Qiwei Zhao, University of North Carolina at Chapel Hill; Dong Li, University of North Carolina at Chapel Hill; Yanchi Liu, NEC Laboratories America, Inc.; Wei Cheng, NEC Laboratories America, Inc.; Yiyou Sun, University of California, Berkeley; Mika Oishi, NEC Corporation; Takao Osaki, NEC Corporation; Katsushi Matsuda, NEC Corporation; Huaxiu Yao, University of North Carolina at Chapel Hill; Chen Zhao, Baylor University; Haifeng Chen, NEC Laboratories America, Inc.; Xujiang Zhao, NEC Laboratories America, Inc.

Abstract: Large language models (LLMs) integrated into multi-step agent systems enable complex decision-making processes across various applications. However, their outputs often lack reliability, making uncertainty estimation crucial. Existing uncertainty estimation methods primarily focus on final-step outputs, which fail to account for cumulative uncertainty over the multi-step decision-making process and the dynamic interactions between agents and their environments. To address these limitations, we propose SAUP (Situation Awareness Uncertainty Propagation), a novel framework that propagates uncertainty through each step of an LLM-based agent’s reasoning process. SAUP incorporates situational awareness by assigning situational weights to each step’s uncertainty during the propagation. Our method, compatible with various one-step uncertainty estimation techniques, provides a comprehensive and accurate uncertainty measure. Extensive experiments on benchmark datasets demonstrate that SAUP significantly outperforms existing state-of-the-art methods, achieving up to 20% improvement in AUROC.

Publication Link: https://aclanthology.org/2025.acl-long.302/