让边缘智能体协作记忆，不更新模型也能提升长任务表现。
CoMIC: Collaborative Memory and Insights Circulation for Long-Horizon LLM Agents in Cloud-Edge Systems
- 边缘代理本地执行，云端异步评估并共享经验。
- 在五类长程任务中，弱边缘模型进度提升显著。
- 适合资源受限的边缘部署场景，无需微调模型。
将轻量级大语言模型（LLM）代理部署在边缘服务器可降低延迟、贴近用户，但资源受限的边缘模型在需要持久记忆、子目标追踪和反思的长周期任务中表现不佳。部署后微调成本高且难以在异构节点间扩展，而纯本地记忆会导致经验孤立和提示上下文膨胀。本文提出 extsc{CoMIC}，一种无需参数更新的云边协同记忆与经验循环框架。 extsc{CoMIC} 采用「集中反思、分布式执行」设计：边缘代理使用面向子目标的层次化记忆和选择性历史重展开本地执行；云端大模型评议员异步评估完成的任务轨迹，筛选可复用经验，并基于语义子目标标识聚合跨代理指导。在涵盖符号规划与文本交互的五类长周期任务中， extsc{CoMIC} 提升了弱边缘代理的进展率与动作置信度，实现了不依赖模型参数更新的任务成功率增益。
原文摘要 · Abstract (English)
Deploying lightweight Large Language Model (LLM) agents on edge servers can reduce latency and move agentic services closer to users, but resource-constrained edge models often struggle with long-horizon tasks that require persistent memory, subgoal tracking, and reflection. Fine-tuning edge models after deployment is costly and difficult to scale across heterogeneous nodes, while purely local memory leaves agents with isolated experience and growing prompt context. We propose \textsc{CoMIC}, a parameter-update-free cloud-edge framework for Collaborative Memory and Insights Circulation. \textsc{CoMIC} follows a \textit{Centralized Reflection, Decentralized Execution} design: edge agents execute locally using subgoal-oriented hierarchical memory and selective re-expansion of relevant histories, while a cloud-side LLM critic asynchronously evaluates completed trajectories, filters reusable experience, and aggregates cross-agent guidance keyed by semantic subgoal identifiers. Across five long-horizon agent tasks spanning symbolic planning and text interaction, \textsc{CoMIC} improves progress rate and action grounding for weak edge agents and yields task-dependent success-rate gains without updating model parameters.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。