Metraipolis › archive › dc_5ac627ebb409
ClawMMU: Protection-Coherent Virtual Memory for Stateful LLM Agents — arXiv:2610.28093
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.28093
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-21T07:08:15.223Z
- SHA-256 of the archived text
b0f769c64b1931eb033a4527bbb4668fe3c481058fb1874a4b773e142bf6f11f- Stable URL of this page
https://metraipolis.com/documento/dc_5ac627ebb409
How to cite
ClawMMU: Protection-Coherent Virtual Memory for Stateful LLM Agents — arXiv:2610.28093 — arXiv (en), https://arxiv.org/abs/2610.28093 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5ac627ebb409 (Metraipolis archive dc_5ac627ebb409, sha256 b0f769c64b1931eb033a4527bbb4668fe3c481058fb1874a4b773e142bf6f11f)
BibTeX
@misc{dc5ac627ebb409,
title = {ClawMMU: Protection-Coherent Virtual Memory for Stateful LLM Agents — arXiv:2610.28093},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.28093},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5ac627ebb409; sha256 b0f769c64b1931eb033a4527bbb4668fe3c481058fb1874a4b773e142bf6f11f},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.28093. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
AI agents keep long-term memory, much of which comes from web pages and tool results. Agent harnesses, the software that runs the model, do not record who may act on this memory, so an instruction hidden in a fetched page can be treated as the user's own instruction. We propose ClawMMU, a protection layer that labels every piece of memory with its owner, scope, origins, and authority. ClawMMU checks these labels whenever content enters the model's context, is written to storage, returns from storage, or drives a tool call. Its central rule is that information may rise in authority, but authority never lifts itself. Because the checks depend on where content came from, not on what it says, memory attacks become classical protection faults that the harness stops even when the model is fooled. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI); Operating Systems (cs.OS). Pagina del lavoro: https://arxiv.org/abs/2610.28093 — 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 2026-10-21T07:05:43.622Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).