Bilevel Coordinated Reflection: A Game-Theoretic Approach to Multi-Agent LLM Systems

类型：arxiv
标识：2609.02750
链接：https://arxiv.org/abs/2609.02750
主分类：agent
形态：position
TLDR：Multi-agent LLM systems commonly use an orchestrator to decompose a task for a team of workers and then improve through textual reflection. Despite strong empirical results, these systems lack a unified account of coordination, memory improvement, and the role of external verification. We model orchestrator-worker interaction as a bilevel coordination game: under bounded coupling, the workers' local-update game is an approximate potential game whose equilibrium slack is controlled by decomposition quality. We then analyse reflection as stochastic movement over semantic memory states. For free-
待LLM分类：否
来源文件：
/inbox/tom/_candidates/2026-09-07-agent-rag-longcontext-candidates.json

2026-07-18

笔记 Cameron R. Wolfe (Deep Learning Focus) · RSS 摘要

2026-08-31

攻略 Reward Seeking：LLM 在 RL 训练中学会"讨好评分器" · 干货攻略

2026-07-29

笔记 flyP 夜间轻量精读 · 2026-06-23（cron 3d8f503a · 22:50 CST）

2026-06-23

笔记 flyP 精读与反方审稿 · Agentic RL 综述 + GLM-5.2 开放 Agent 路线 · 2026-07-15 22:50

2026-07-15

笔记 ReOPD (arXiv:2607.04763v2) 精读与批判 — flyP

2026-07-26

arXiv

2609.02750

TLDR

Multi-agent LLM systems commonly use an orchestrator to decompose a task for a team of workers and then improve through textual reflection. Despite strong empirical results, these systems lack a unified account of coordination, memory improvement, and the role of external verification. We model orchestrator-worker interaction as a bilevel coordination game: under bounded coupling, the workers' local-update game is an approximate potential game whose equilibrium slack is controlled by decomposition quality. We then analyse reflection as stochastic movement over semantic memory states. For free-

arXiv ↗ 中文全文

Bilevel Coordinated Reflection: A Game-Theoretic Approach to Multi-Agent LLM Systems

相关 · 按语义匹配