多智能体 LLM 系统中的通信失败机制解析
Understanding Communication Failures in Multi-Agent LLM Systems
AI 导读
一项 arXiv 研究基于医疗问答、编程、金融、多跳推理等任务中 3,211 条失败的多轮轨迹，构建了多智能体 LLM 系统通信失败的分类体系。分析显示失败主要由恢复/纠错与共识/决策问题主导，而非孤立的首轮错误，且许多错误在智能体观察到同伴回复后才出现。研究者建议加强分歧处理、证据验证、少数派信号保留和最终定稿前的恢复检查。
正文 · 原文
Abstract:Multi-agent LLM systems aim to improve reasoning through debate, evidence sharing, and mutual correction, but interaction can also introduce failures that final accuracy alone cannot explain. Agents may converge too early, ignore dissent, suppress correct minority answers, or preserve a wrong framing across rounds. We present a corpus-grounded taxonomy of communication failures, built from 3,211 failed multi-round trajectories across medical QA, coding, finance, multi-hop reasoning, ethics, and embodied text tasks. Using this taxonomy, we analyze failure distributions across datasets and system designs, round-level error dynamics, recovery hardness, majority transitions, minority-signal suppression, and self-reported confidence. We find that failures are dominated by recovery/correction and consensus/decision problems rather than isolated first-round mistakes, with many emerging after agents first observe peer responses. Based on these findings, we recommend stronger disagreement handling, evidence verification, minority-signal preservation, and recovery checks before finalization.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.32060 [cs.AI] |
| (or arXiv:2610.32060v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.32060 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Priya Pitre [view email]
[v1]
Fri, 23 Oct 2026 01:55:07 UTC (202 KB)
来源：arXiv cs.AI · arxiv.org