Metraipolis › archive › dc_61bfac607c87
When Is a Memory Safe Enough to Act On? Action-Conditioned Verification for Long-Lived Agents — arXiv:2610.31446
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.31446
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-23T12:47:43.368Z
- SHA-256 of the archived text
3657d1e7b37ac907bd8dad67c7a4ad40cff0b7509ad2f7bd6c5effc62978df3b- Stable URL of this page
https://metraipolis.com/documento/dc_61bfac607c87
How to cite
When Is a Memory Safe Enough to Act On? Action-Conditioned Verification for Long-Lived Agents — arXiv:2610.31446 — arXiv (en), https://arxiv.org/abs/2610.31446 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_61bfac607c87 (Metraipolis archive dc_61bfac607c87, sha256 3657d1e7b37ac907bd8dad67c7a4ad40cff0b7509ad2f7bd6c5effc62978df3b)
BibTeX
@misc{dc61bfac607c87,
title = {When Is a Memory Safe Enough to Act On? Action-Conditioned Verification for Long-Lived Agents — arXiv:2610.31446},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.31446},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_61bfac607c87; sha256 3657d1e7b37ac907bd8dad67c7a4ad40cff0b7509ad2f7bd6c5effc62978df3b},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.31446. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Long-lived agents act on memory carried across sessions, and many of their actions are irreversible, so a stale memory can turn directly into harm. Existing work improves the memory itself --- retrieving, refreshing, or reconciling entries --- yet whether an entry must be re-checked before acting is not a property of the entry: the same remembered address can be used unchecked for a routine shipment but should be re-confirmed before a valuable one. We therefore study \emph{action-conditioned memory verification}: given a candidate action, decide which of its memory premises to re-verify and when to proceed or abstain, balancing execution risk, abstention, and verification cost. We formulate this as a sequential stopping problem, in which a verification is worth its cost only if it can change what the agent does next; under suitable verifier reliability and cost conditions, increasing action consequence can move the same memory through proceed, verify, and direct-abstention regimes. Building on this formulation, \textsc{AMV-Guard} is a practical verification gate that extracts the premises a candidate action rests on, calibrates beliefs about whether they are wrong, and selects checks by bounded value of information (VOI). On ToolSandbox, relative to a static policy given the same inputs, it lowers mean verification cost by 33.9\% while keeping 100\% safe-proceed recall and 0\% Gate harmful, and raises Proceed coverage from 38.83\% to 44.17\%; in a gate-core transfer to AppWorld, which carries over the stopping/VOI core but not the full structured pipeline, it lowers verification cost by 21.3\% at a small cost in safe-proceed recall and task success. These results suggest that verification effort should be set by what an agent is about to do, not only by how reliable a memory appears.
Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.MA, nuovi lavori del giorno).
Categorie arXiv: Multiagent Systems (cs.MA).
Pagina del lavoro: https://arxiv.org/abs/2610.31446 — 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 2026-10-23T12:47:42.913Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).