Newest papers

ClawMMU: Protection-Coherent Virtual Memory for Stateful LLM Agents

Jinhao Hu, Bardia Mohammadi, Mofasshara Binte Rafique, Lars Henning Klein, Akhil Arora, Ashvin Goel, Laurent Bindschaedler · 2026 · arXiv

DOI: 10.48550/arXiv.2610.28093 · open-access PDF

Portuguese

Os agentes da IA mantêm memória de longo prazo, grande parte da qual vem de páginas da web e resultados de ferramentas. Os agentes aproveitam, o software que executa o modelo, não registram quem pode agir nesta memória, de modo que uma instrução escondida em uma página recolhida pode ser tratada como instrução própria do usuário. Proponemos ClawMMU, uma camada de proteção que rotula cada pedaço de memória com seu proprietário, âmbito, origens e autoridade. ClawMMU verifica essas etiquetas sempre que o conteúdo entra no contexto do modelo, é escrito para armazenamento, retorna do armazenamento ou conduz uma chamada de ferramenta.

Machine translation (Meta M2M100); the English original follows.

Abstract

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 authori

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Is this paper real? This page was built from OpenAlex and arXiv records, not generated. Software can verify any reference in one call: GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.28093, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txt

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange