Back to Home
arXiv AI··Papers & Tech

Who May Overrule the Agent? Evidence-Gated Authority in LLM Review Institutions

中文摘要

该研究探讨了多智能体系统中AI审核员的权限界限,提出使用“证据门控”机制来优化审核员推翻代理决策的有效性。

English Summary

This study investigates AI reviewer authority, proposing an "evidence-gated" mechanism to optimize when reviewers should override upstream agent decisions in multi-agent systems.

Original Excerpt

arXiv:2611.05390v1 Announce Type: new 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; Q…