arXiv cs.MA· Fuad Ali·· 11 小时前AI 评分31
谁有权否决智能体？LLM 评审机制中的证据门控权限
Who May Overrule the Agent? Evidence-Gated Authority in LLM Review Institutions
AI 导读
研究将多智能体系统中的评审者越权问题定义为"机构管辖权"问题：在相同提案、任务信息、模型与评审输出下，评审者何时有权让自己的建议具有约束力。在 Qwen 与 Mistral 两个本地开源权重模型、96 个空间政策任务及 96 任务组合迁移基准上，要求评审者先给出机械可验证的约束违反记录与合规修复的证据门控，大幅减少对正确提案的破坏，同时保留多数违规修复。
正文 · 原文
Abstract:Multi-agent language-model systems often add a reviewer to improve an upstream agent's decision. Yet a reviewer can both repair errors and replace decisions that were already correct. We study this as a problem of institutional jurisdiction: conditional on the same proposal, task information, model, and review output, when is the reviewer authorized to make its recommendation binding? We compare broad override authority with an evidence gate that permits replacement only after the reviewer produces a mechanically valid record of a binding-constraint violation and a compliant repair. Across two locally run open-weight models, 96 spatial-policy tasks, and a 96-task portfolio-transfer benchmark, the gate sharply reduces corruption of correct proposals while retaining most repairs of violations. The directions of this tradeoff are partly implied by the rule; its magnitude depends on reviewer transitions and upstream proposal states. A prospective validation then gives the reviewer an explicit choice to retain, replace, or escalate. This eliminates correct-proposal corruption for Qwen, but Mistral still replaces all 48 correct proposals; Qwen also declines every welfare-only repair that Mistral frequently attempts. Thus interface design changes reviewer behavior but is not a model-independent safeguard. The result is not that restricted review is universally better. It is that reviewer competence, action interface, and binding jurisdiction jointly determine both the correction domain and the corruption surface.
| Comments: | 11 pages, 3 figures, 3 tables. Code, frozen outputs, and reproduction instructions: this https URL |
| Subjects: | Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA) |
| Cite as: | arXiv:2611.05390 [cs.AI] |
| (or arXiv:2611.05390v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.05390 arXiv-issued DOI via DataCite |
Submission history
From: Fuad Ali [view email]
[v1]
Sat, 12 Sep 2026 17:58:39 UTC (125 KB)
来源：arXiv cs.MA · arxiv.org