arXiv cs.AI· Priya Pitre, Jun Hou, Yi Fang, Naren Ramakrishnan, Xuan Wang·· 12 小时前AI 评分45
多智能体 LLM 系统为何沟通失败：基于 3,211 条失败轨迹的分类分析
Understanding Communication Failures in Multi-Agent LLM Systems
AI 导读
研究基于医疗问答、编程、金融、多跳推理、伦理和具身文本任务中的 3,211 条失败多轮轨迹，构建了多智能体 LLM 系统沟通失败分类体系。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，许多失败在智能体首次看到同伴回复后才出现。作者据此建议加强分歧处理、证据核验、少数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