arXiv：cs.AI· Jun Liu, Zhenglun Kong, Changdi Yang, Fan Yang, Tianqi Li, Peiyan Dong, Joannah Nanjekye, Hao Tang, Geng Yuan, Wei Niu, Wenbin Zhang, Pu Zhao, Xue Lin, Dong Huang, Yanzhi Wang·· 16 小时前
RCR-Router：面向多智能体 LLM 系统的角色感知上下文路由
RCR-Router: Efficient Role-Aware Context Routing for Multi-Agent LLM Systems with Structured Memory
AI 导读
研究者提出 RCR（Role-Conditioned Context Routing），一种在多智能体 LLM 系统执行时按角色和 token 预算选择性路由共享记忆条目的执行层。
正文
Authors:Jun Liu, Zhenglun Kong, Changdi Yang, Fan Yang, Tianqi Li, Peiyan Dong, Joannah Nanjekye, Hao Tang, Geng Yuan, Wei Niu, Wenbin Zhang, Pu Zhao, Xue Lin, Dong Huang, Yanzhi Wang
Abstract:Large language model (LLM) agents are increasingly organized into multi-agent systems for complex reasoning and decision-making tasks. However, scaling such systems can substantially increase inference cost, especially when multiple agents repeatedly process overlapping or role-irrelevant context. Many existing multi-agent frameworks focus on orchestration and role specialization, but still rely on full-context sharing or static context routing, leading to redundant token consumption, increased runtime, and inefficient memory utilization. We present \emph{Role-Conditioned Context Routing (RCR)}, an execution-time layer that allocates role-relevant shared-memory items to agents under token budgets. Given a structured shared memory and a set of role-specialized agents, RCR selectively routes task-relevant memory items to each agent according to its role, optional task-stage metadata, and token budget. RCR integrates role-conditioned relevance scoring, token-budgeted memory selection, and iterative memory updates to reduce redundant context exposure during multi-agent execution. Experiments on multi-hop QA, embodied reasoning (ALFWorld), and web-based decision-making (WebShop) show that RCR reduces token usage by 35\% on average and runtime by 38\% on average relative to Full-Context, while maintaining or improving task performance.
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA) |
| Cite as: | arXiv:2508.04903 [cs.CL] |
| (or arXiv:2508.04903v4 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2508.04903 arXiv-issued DOI via DataCite |
Submission history
From: Jun Liu [view email]
[v1]
Wed, 6 Aug 2025 21:59:34 UTC (720 KB)
[v2]
Sun, 10 Aug 2025 09:40:57 UTC (721 KB)
[v3]
Tue, 12 Aug 2025 16:29:05 UTC (721 KB)
[v4]
Mon, 12 Oct 2026 01:38:24 UTC (2,784 KB)
来源：arXiv：cs.AI · arxiv.org