arXiv cs.AI· Xufei Shi, Derong Xu, Wei Chen, Yanyu Yao, Zhi Zheng, Tong Xu·· 12 小时前AI 评分31
CoEvolve：多智能体系统中记忆与协作的反事实引导协同进化
CoEvolve: Counterfactual-Guided Co-Evolution of Memory and Collaboration in Multi-Agent Systems
AI 导读
CoEvolve 是一个以技能为中心的框架，用于实现多智能体系统中记忆与协作的对齐协同进化，通过对比原始与反事实轨迹识别局部进化信号，协调技能库、技能记忆与执行器记忆的跨轮更新。在 ALFWorld 和 WebShop 基准上，CoEvolve 在所评估基线中取得最佳整体性能，消融实验进一步验证了对齐协同进化与反事实引导进化的有效性。
正文
Abstract:Large language model (LLM)-based multi-agent systems have demonstrated remarkable capabilities across diverse tasks. Yet their memory and collaboration architectures often remain static, constraining continued adaptation and improvement over time. Recent work addresses this limitation by evolving memory or collaboration based on interaction experience. However, updates to these interdependent components are not necessarily aligned: an evolved workflow may require experience that memory does not provide, while evolved memory may encode guidance that the workflow cannot use. To address this problem, we propose CoEvolve, a skill-centered framework for aligned memory--collaboration evolution. CoEvolve represents reusable local collaboration patterns as skills that link evolving collaboration decisions to their supporting memories. It further compares original and counterfactual trajectories to identify local evolution signals, which coordinate updates to the skill library, skill memory, and executor memories across rounds. Experiments on the ALFWorld and WebShop benchmarks confirm that CoEvolve achieves the best overall performance among the evaluated baselines, while ablation studies further validate the effectiveness of aligned co-evolution and counterfactual-guided evolution.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.33539 [cs.AI] |
| (or arXiv:2610.33539v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.33539 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Xufei Shi [view email]
[v1]
Sat, 24 Oct 2026 08:35:39 UTC (3,366 KB)
来源：arXiv cs.AI · arxiv.org