Metraipolis › archive › dc_d48445648ef0

Grounding Agent Memory: Environment-Probing Curation for Enterprise Agents — arXiv:2609.11060

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.11060
License (as archived)
CC0-1.0
Archived
2026-10-16T05:23:39.862Z
SHA-256 of the archived text
8a48644f08afa193633b235b67e1b2f2caeaef63877900b5f41e27ad82fb35c0
Stable URL of this page
https://metraipolis.com/documento/dc_d48445648ef0

How to cite

Grounding Agent Memory: Environment-Probing Curation for Enterprise Agents — arXiv:2609.11060 — arXiv (en), https://arxiv.org/abs/2609.11060 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d48445648ef0 (Metraipolis archive dc_d48445648ef0, sha256 8a48644f08afa193633b235b67e1b2f2caeaef63877900b5f41e27ad82fb35c0)
BibTeX
@misc{dcd48445648ef0,
  title = {Grounding Agent Memory: Environment-Probing Curation for Enterprise Agents — arXiv:2609.11060},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.11060},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d48445648ef0; sha256 8a48644f08afa193633b235b67e1b2f2caeaef63877900b5f41e27ad82fb35c0},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.11060. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Persistent memory is entering production-oriented agent platforms to help long-horizon agents accumulate experience across sessions. Yet a post-task curator agent restricted to completed trajectories can preserve errors, overgeneralize partial evidence, or retain stale knowledge. We introduce environment-probing curation, a deployment-compatible extension that gives an existing asynchronous curator agent least-privilege, read-only world tools to check, scope, and refresh candidate memories. It requires no model retraining and leaves the task agent, retriever, memory representation, and production write authority unchanged. In a production-like GitHub Copilot (GHCP) harness built on its SDK, we compare stateless execution, full in-context learning, GHCP + Mem, and GHCP + Mem (w/ Env Probing) on CLBench database exploration and 90 adapted APEX management-consulting tasks. On CLBench, probing raises pass rate from 39% to 73% and pass-discounted reward from 8.60 to 22.60 while reducing queries from 8.8 to 4.7 per question and task-agent cost from \$3.38 to \$1.68. Across six APEX worlds, all 18 memory-versus-baseline mean reward comparisons are positive and task-agent tool calls fall by 16--75%; probing gives the best task-agent reward gain per dollar in five worlds. Probing also attains higher mean reward than GHCP + Mem on both Sonnet 4.6 and Opus 4.7 without schema drift. Environment probing therefore turns existing agent-memory curation into an environment-informed, auditable process while preserving a compact task-time interface.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-11 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Software Engineering (cs.SE).
Pagina del lavoro: https://arxiv.org/abs/2609.11060 — 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 2026-10-16T05:17:51.962Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).