arXiv cs.MA· Hanwen Li, Jinhao Duan, Xiaoshuang Shi, Yue Zhang, Tianlong Chen, Kaidi Xu, Chenxi Yuan·· 12 小时前AI 评分31
UMAS：面向多智能体 LLM 系统的系统级不确定性量化框架
UMAS: System-Level Uncertainty Quantification for Multi-Agent LLM Systems
AI 导读
UMAS 是一个无需参数、不增加推理成本的多智能体 LLM 系统级不确定性量化框架，利用各次生成的内在置信度估计智能体间影响的可信度，并据此调节节点级不确定性更新、聚合为系统级估计。
正文
Abstract:Multi-Agent Systems (MAS) are a promising paradigm for improving the reasoning capabilities of large language models, yet they remain prone to hallucinations and erroneous outputs. Uncertainty Quantification (UQ) is crucial for assessing system reliability, but existing methods are largely designed for single-agent settings; how to exploit MAS structure for UQ while keeping inference cost-neutral remains underexplored. We propose UMAS, a parameter-free, cost-neutral, system-level UQ framework that estimates the credibility of inter-agent influence from each generation's intrinsic confidence. UMAS uses these credibility signals to modulate how uncertainty is updated at each node, and then aggregates node-level uncertainty into a system-level estimate. Extensive evaluations on Debate and DyLAN across diverse benchmarks show that UMAS consistently outperforms state-of-the-art baselines by an average of 10.54 AUROC points. Beyond uncertainty estimation, UMAS enables hallucination detection and uncertainty-aware answer selection, improving MAS accuracy by up to 12.6 points on specific tasks while enhancing reliability. Code is available at this https URL.
| Comments: | Accepted at NeurIPS 2026 |
| Subjects: | Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.33013 [cs.MA] |
| (or arXiv:2610.33013v1 [cs.MA] for this version) | |
| https://doi.org/10.48550/arXiv.2610.33013 arXiv-issued DOI via DataCite |
Submission history
From: Hanwen Li [view email]
[v1]
Thu, 8 Oct 2026 04:05:08 UTC (7,029 KB)
来源：arXiv cs.MA · arxiv.org