Metraipolis › archive › dc_d0f598f7bf2b

EMMA: Structured Episodic Memory for Multi-Agent Workflow Synthesis — arXiv:2610.35966

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.35966
License (as archived)
CC0-1.0
Archived
2026-10-28T03:01:03.611Z
SHA-256 of the archived text
032f065befa62d4dfaf659178da2cabcae0ba89cab2162969e7510f895a84283
Stable URL of this page
https://metraipolis.com/documento/dc_d0f598f7bf2b

How to cite

EMMA: Structured Episodic Memory for Multi-Agent Workflow Synthesis — arXiv:2610.35966 — arXiv (en), https://arxiv.org/abs/2610.35966 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d0f598f7bf2b (Metraipolis archive dc_d0f598f7bf2b, sha256 032f065befa62d4dfaf659178da2cabcae0ba89cab2162969e7510f895a84283)
BibTeX
@misc{dcd0f598f7bf2b,
  title = {EMMA: Structured Episodic Memory for Multi-Agent Workflow Synthesis — arXiv:2610.35966},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.35966},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d0f598f7bf2b; sha256 032f065befa62d4dfaf659178da2cabcae0ba89cab2162969e7510f895a84283},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.35966. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

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.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.MA, nuovi lavori del giorno).
Categorie arXiv: Multiagent Systems (cs.MA) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.35966 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-28T03:01:03.518Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).