Newest papers

ClawMMU: Protection-Coherent Virtual Memory for Stateful LLM Agents

Jinhao Hu, Bardia Mohammadi, Mofasshara Binte Rafique, Lars Klein, Akhil Arora, Ashvin Goel, Laurent Bindschaedler · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.28093 · cited by 0 works (OpenAlex) · open-access PDF

Korean

AI 에이전트는 웹 페이지 및 도구 결과에서 오고있는 장기 메모리를 보관합니다. 에이전트는 모델을 실행하는 소프트웨어를 활용하여이 메모리에 행동 할 수있는 사람을 기록하지 않으므로 수집 된 페이지에 숨겨진 지침이 사용자의 자신의 지침으로 취급 될 수 있습니다. ClawMMU는 소유자, 범위, 기원 및 권한으로 메모리의 모든 조각을 라벨링하는 보호 레이어입니다. ClawMMU는 컨텐츠가 모델의 맥락에 들어갈 때마다이 라벨을 확인합니다. 저장소에 쓰여지거나 저장소에서 반환되거나 도구 호출을 실행합니다. 중앙 규칙은 정보가 권한에 올라갈 수 있지만 권한은 결코 자신을 들어 올리지 않습니다.

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 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.

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

Topics: Computer science, Stateful firewall, Software, Layer (electronics), Web page, Programming language

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