arXiv cs.AI· Hajime Shimao, Warut Khern-am-nuai·· 10 小时前AI 评分26
确定性 LLM 审议中的低边际分支点：五智能体委员会白盒审计
Low-Margin Branch Points in Deterministic LLM Deliberation
AI 导读
一项白盒审计在五智能体委员会、贪心解码下考察确定性 LLM 审议的局部分支敏感性：低边际 token 位置比高边际对照更易触发转录分歧，匹配的降边际梯度方向也比升边际对照更常引发分支。
正文 · 原文
Abstract:Deterministic multi-agent LLM deliberation can exactly re- play a fixed transcript, yet exact replay alone does not test local resistance to controlled internal interventions. Because each response becomes context for later agents, a changed greedy token can redirect a transcript-mediated execution. We present a white-box audit of this local failure mode in a five-agent committee under greedy decoding. We prospectively select low- and high-margin token positions, require exact no-op re- play, inject norm-relative residual displacements, and measure first transcript divergence within ten generated tokens. Low- margin positions are more branch-sensitive than high-margin controls, while matched margin-reducing gradient directions branch substantially more often than margin-increasing con- trols. We then continue selected realized branches through the committee to distinguish textual persistence from preference and final-decision changes. The contribution is a controlled audit for local branch susceptibility in this specified protocol, not an estimate of deployment frequency, decision correct- ness, or a phenomenon unique to multi-agent systems.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.00546 [cs.AI] |
| (or arXiv:2611.00546v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.00546 arXiv-issued DOI via DataCite |
Submission history
From: Hajime Shimao [view email]
[v1]
Sat, 31 Oct 2026 02:01:39 UTC (108 KB)
[v2]
Mon, 9 Nov 2026 22:12:06 UTC (108 KB)
来源：arXiv cs.AI · arxiv.org