arXiv cs.MA· Yalun Wu, Qiwei Ma, Bingbing Xu, Dan Zhang, Long Bai, Boyang Wang, Jiatong Li, Zheqi Lv, Shaozu Yuan, Peiying Wang, Jiawei Wang, An Zhang, Tat-Seng Chua·· 12 小时前AI 评分31
EvolveMAS：面向去中心化多智能体系统的统一自进化框架
EvolveMAS: A Unified Self-Evolution Framework for Decentralized Multi-Agent Systems
AI 导读
EvolveMAS 是一个面向去中心化多智能体系统的统一自进化框架，通过为每个主智能体配备专用 MemoryAgent 实现测试时记忆治理，并在主智能体推理上下文之外维护隔离的结构化记忆。框架将适应过程分为回合内在线适应与跨回合知识进化两部分，后者蒸馏出可复用的伙伴画像与交互策略。在覆盖三类交互设置的五个基准、六种评测配置上，EvolveMAS 取得最高归一化平均分 77.2，且无需更新模型参数。
正文
Authors:Yalun Wu, Qiwei Ma, Bingbing Xu, Dan Zhang, Long Bai, Boyang Wang, Jiatong Li, Zheqi Lv, Shaozu Yuan, Peiying Wang, Jiawei Wang, An Zhang, Tat-Seng Chua
Abstract:Decentralized multi-agent systems (MAS) enable large language model agents to interact without centralized control. Agents need to adapt to other agents at test time and reuse accumulated experience to refine their interaction strategies for future interactions across cooperative, mixed-motive, and adversarial settings. However, existing approaches often develop adaptation mechanisms for particular interaction settings. This raises a central question: how can decentralized MAS support test-time adaptation and reliable experience reuse across interaction settings? We propose EvolveMAS, a unified self-evolution framework that supports experience-driven adaptation within and across episodes. EvolveMAS pairs each main agent with a dedicated MemoryAgent for test-time memory governance, which maintains isolated Structured Memory outside the main agent's reasoning context. EvolveMAS organizes adaptation into two complementary processes: Within-Episode Online Adaptation and Cross-Episode Knowledge Evolution. During Within-Episode Online Adaptation, MemoryAgent updates episodic, semantic, and working memory from interaction evidence to inform subsequent actions. During Cross-Episode Knowledge Evolution, the main agent uses memory and episode outcomes to distill reusable interaction knowledge comprising partner profiles and interaction strategies for subsequent episodes. Across five benchmarks and six evaluation configurations spanning the three interaction settings, EvolveMAS achieves the highest normalized average score of 77.2 among the compared methods. Ablations show that test-time memory governance and cross-episode knowledge reuse both improve performance. Interaction traces further illustrate how updated memory and distilled knowledge guide subsequent interactions. These results support experience-driven adaptation in decentralized MAS without updating model parameters.
| Comments: | 27 pages, 9 figures |
| Subjects: | Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.34450 [cs.MA] |
| (or arXiv:2610.34450v1 [cs.MA] for this version) | |
| https://doi.org/10.48550/arXiv.2610.34450 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Jiawei Wang [view email]
[v1]
Sun, 25 Oct 2026 17:09:49 UTC (1,614 KB)
来源：arXiv cs.MA · arxiv.org