← 研踪 PaperTrailarXiv:2611.03192 · v1
Verification Allocation in Stylized LLM Agent Pipelines under Hidden Tool Degradation: A Discrete-Time Mean-Field Analysis
首次提交 2026-11-02 22:21
版本更新 2026-11-02 22:21
雷达观测 2026-11-04 23:41
摘要
We study a stylized verification-allocation model for role-structured pipelines with binary inherited corruption and hidden tool degradation. Standard contraction arguments give serial and multi-role fixed points and a finite-population deviation bound under independence. When verification actions affect neither future errors nor observations, the belief-state choice is myopic. We characterize a two-action threshold under an explicit monotonicity condition and a restricted four-action interval and tail policy when generation rates are equal and only one check degrades. Synthetic experiments check these model-specific results and find counterexamples outside the restriction. This is conditional model analysis, not a general mean-field-game result or empirical evidence about deployed LLM systems.
作者备注：12 pages, 3 figures. Code and results: https://github.com/Shereen20112003/verification-allocation-hidden-tool-degradation
观测到的版本
相邻记录可能跳过中间版本；首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-02 22:21 · 最新观测版本
Verification Allocation in Stylized LLM Agent Pipelines under Hidden Tool Degradation: A Discrete-Time Mean-Field Analysis
We study a stylized verification-allocation model for role-structured pipelines with binary inherited corruption and hidden tool degradation. Standard contraction arguments give serial and multi-role fixed points and a finite-population deviation bound under independence. When verification actions affect neither future errors nor observations, the belief-state choice is myopic. We characterize a two-action threshold under an explicit monotonicity condition and a restricted four-action interval and tail policy when generation rates are equal and only one check degrades. Synthetic experiments check these model-specific results and find counterexamples outside the restriction. This is conditional model analysis, not a general mean-field-game result or empirical evidence about deployed LLM systems.
查看此版本解读 →查看此版本原文 ↗