cs.CL提交于 2026-10-15已译

共同进化，各得其所：社会脚手架如何塑造自进化多智能体系统

Evolve Together, Fall into Place: How Social Scaffolding Shapes Self-Evolving Multi-Agent Systems

Seojin Hwang · Youngmin Choi · Hwanhee Lee

中文摘要

近期研究允许大语言模型智能体进化其自身框架乃至团队结构。借鉴发展心理学中的脚手架概念，研究探讨自进化多智能体系统（MAS）的社会脚手架（即团队组成与相互依赖结构，如协作、从众或竞争）如何影响团队进化。提出 AGORA 框架，在三种进化机制中独立变化这两个维度。在网页搜索、代码库理解和文档问答任务中发现：(1) 单独进化很少有帮助，而合适的脚手架比单独进化的强智能体高出最多 25 分；(2) 其机制是将权力转移给最有能力的成员，其他成员退后；(3) 该成员在团队中成长后，表现优于单独使用同一模型。因此，设计自进化 MAS 需设计其社会脚手架，这不仅塑造智能体如何学习，还决定谁成为领导者。

关键要点

01提出 AGORA 框架，独立变化团队组成与相互依赖结构以研究社会脚手架对 MAS 进化的影响。
02在网页搜索、代码理解和文档问答中，合适脚手架比单独进化的强智能体性能高出最多 25 分。
03进化机制将权力转移给最有能力成员，其他成员退后，该成员在团队中成长后表现优于单独模型。
04设计自进化 MAS 需关注社会脚手架，它决定智能体学习方式和领导权归属。

解读

尚无解读。

原始英文摘要

arXiv:2610.19799v1 Announce Type: new Abstract: Recent work lets LLM agents evolve their own harness, up to the team itself. Yet in human groups, the same abilities achieve different outcomes under different social conditions. Borrowing the notion of scaffolding from developmental psychology, we investigate whether the social scaffolding of a self-evolving multi-agent system (MAS), namely its team composition and structure of interdependence (e.g., collaboration, conformity, or competition), shapes how the team evolves. We present AGORA (Agents Growing through Organization and Relational Arrangements), a framework that varies the two axes independently across three evolution mechanisms. Across web search, codebase understanding, and document QA, we find that (1) evolution alone rarely helps, while the right scaffolding surpasses a strong agent evolved alone by up to 25 points; (2) it does so by moving authority to the most capable member while the others step back; and (3) this member, having grown within the team, outperforms the same model alone. Designing a self-evolving MAS thus requires designing its social scaffolding, which shapes not only how its agents learn, but who gets to lead.