arXiv cs.MA· Ruijia Niu, Kaiwen Liu, Rose Yu, Kristina Klinkner·· 11 小时前AI 评分38
EMMA：面向多智能体工作流合成的结构化情景记忆
EMMA: Structured Episodic Memory for Multi-Agent Workflow Synthesis
AI 导读
EMMA 将多智能体系统配置（控制器路由记忆、继承通用提示词的专家智能体、分步流程技能）作为情景记忆，通过反射器把每个诊断出的失败归属到拥有该修正范围的模块并提出单次编辑，配合与骨干模型无关的守卫保证每次编辑有界、增量且范围明确。
正文
Abstract:Deploying LLM agents in specialized domains often require fine-grained task-specific procedures and knowledge a general model lacks, yet parameter training is costly and non-interpretable, and hand-engineered agent systems shift the cost to expert labor. Episodic memory updates at context level, but single agent adaptation struggles to alleviate cross-task interference, especially with complex tasks such as multi-application navigation or data pool analysis. Although multi-agent evolution alleviates this problem, existing methods lack clear definitions on update rules and rely on the backbone model to decide where and how much the system changes. This ill-scoped evolution makes learning follow the backbone's structural preferences rather than the observed failures: one backbone accumulates skills while another spawns agents. We present EMMA (structured Episodic Memory for Multi-Agent workflow synthesis), which treats the configuration of a multi-agent system (controller routing memory, specialists inheriting a generalist prompt, and per-step procedural skills) as episodic memory. EMMA makes the ownership and admissibility of updates explicit: a reflector assigns each diagnosed failure to the scope owning its correction and proposes one edit within it, and backbone-independent guards keep every accepted edit scoped, incremental and bounded. Across KramaBench, AppWorld and BBEH with five backbones and four baselines (ACE, GEPA, EvoMAS and ADAS), EMMA attains the best overall performance on every benchmark, up to 32.5% over the strongest baseline, maintains consistent improvements across backbones, and uses 17-38% fewer training executions than the most efficient baseline.
| Subjects: | Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.35966 [cs.MA] |
| (or arXiv:2610.35966v1 [cs.MA] for this version) | |
| https://doi.org/10.48550/arXiv.2610.35966 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Ruijia Niu [view email]
[v1]
Sun, 25 Oct 2026 18:00:11 UTC (4,594 KB)
来源：arXiv cs.MA · arxiv.org