Metraipolis › archive › dc_e7845f77cbb9

Just-in-Time Memory: Learning to Curate Task-Adaptive Memory for LLM Agents — arXiv:2609.27334

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.27334
License (as archived)
CC0-1.0
Archived
2026-10-08T09:11:22.256Z
SHA-256 of the archived text
125bc2a49ce395b59080d70eb61e679b098081a5e14ca0dce11cc688012ac213
Stable URL of this page
https://metraipolis.com/documento/dc_e7845f77cbb9

How to cite

Just-in-Time Memory: Learning to Curate Task-Adaptive Memory for LLM Agents — arXiv:2609.27334 — arXiv (en), https://arxiv.org/abs/2609.27334 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e7845f77cbb9 (Metraipolis archive dc_e7845f77cbb9, sha256 125bc2a49ce395b59080d70eb61e679b098081a5e14ca0dce11cc688012ac213)
BibTeX
@misc{dce7845f77cbb9,
  title = {Just-in-Time Memory: Learning to Curate Task-Adaptive Memory for LLM Agents — arXiv:2609.27334},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.27334},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e7845f77cbb9; sha256 125bc2a49ce395b59080d70eb61e679b098081a5e14ca0dce11cc688012ac213},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.27334. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Agentic memory systems reuse past experience to improve future performance, yet most existing designs curate memory at write time: once a task is completed, its trajectory is distilled into a fixed artifact, such as a reflection, workflow, skill, or reasoning strategy, that is later retrieved by similarity. This forces the system to decide what is worth remembering before the future query is known, irreversibly discarding information and producing a query-independent summary that must serve many possible downstream tasks. Learning such a write-time curator is also difficult because the value of a storage decision may only become apparent when a relevant query arrives, potentially many tasks later, creating a long-horizon credit-assignment problem. We instead retain raw trajectories and defer curation until read time, when the current task is known. Given the retrieved traces and the new task, a memory curator synthesizes a compact, task-adaptive payload tailored to the immediate need. Because this payload is consumed on the same task, the curator can be trained directly from immediate task success, avoiding delayed utility signals and the need to artificially group related tasks. Across ALFWorld, WebShop, and $\tau^2$-bench, our Just-in-Time Memory (JitMem) consistently outperforms no-memory agents as well as heuristic and learned write-time memory methods, improving over the strongest baseline by 16.2, 16.3, and 3.9 absolute success-rate points, respectively. Notably, even an untrained curator is already competitive with or surpasses these baselines, showing that task-adaptive read-time curation itself is a major source of the gain; training the curator further compounds the improvement.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-24 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.27334 — 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 2026-10-08T08:13:26.506Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-24?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).