arXiv cs.MA· Jiayuan Liu, Shiyi Du, Weihua Du, Xinhao Cheng, Mingyu Guo, Vincent Conitzer·· 11 小时前AI 评分34
共识陷阱：用交错协作生成抵御被污染多数派
The Consensus Trap: Collaborative Interleaved Generation to Defend against a Corrupted Majority
AI 导读
研究提出“共识陷阱”现象：多数 LLM 智能体若被引导倾向同一错误答案，多数投票反而会强化错误。为此提出 chunk-interleaved Round-Robin（RR）协作，让智能体轮流续写同一条未完成的推理轨迹，使掌握正确推理的少数派能在答案完成前插入修正；RRMaj 再对多条轨迹做最终投票。
正文 · 原文
Abstract:Majority voting offers a simple way to combine answers from multiple large language model (LLM) agents. When a majority follows instructions favoring the same incorrect answer, however, agreement reinforces the error. We call this failure the "consensus trap": a minority may identify the error, yet its reasoning cannot change a vote over completed answers. We introduce chunk-interleaved Round-Robin (RR) collaboration, in which agents take turns extending one unfinished reasoning trajectory. A minority of agents with correct reasoning can insert corrections that subsequent speakers read before the answer is complete. RRMaj combines several such trajectories by a final vote. We characterize a majority barrier for outcome-level voting over final answer labels and derive a local sufficient condition for expected progress when the minority's cumulative corrections outweigh opposing corrupted updates. Across model families and verifiable reasoning benchmarks, RRMaj substantially improves accuracy under majority corruption while retaining comparable average performance under minority corruption in the main advisory-instruction setting. RR also strengthens full-response debate. Scoring the shared reasoning between turns shows that, on average, continuations from uncorrupted agents increase support for the correct answer over the incorrect target. Parallel local serving adds modest latency in the measured deployment. These findings show how language shared during reasoning, and the timing of that exchange, can help a minority correct a group's answer.
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA) |
| Cite as: | arXiv:2604.17139 [cs.CL] |
| (or arXiv:2604.17139v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2604.17139 arXiv-issued DOI via DataCite |
Submission history
From: Jiayuan Liu [view email]
[v1]
Sat, 18 Apr 2026 20:31:37 UTC (138 KB)
[v2]
Fri, 6 Nov 2026 11:13:49 UTC (150 KB)
来源：arXiv cs.MA · arxiv.org