LATIDIA · Ciberseguridad

Protección de los agentes de LLM contra amenazas a largo plazo mediante la memoria en la sombra.

arXiv: 2605.03228v2Tipo de anuncio: reemplazar Resumen: A medida que los agentes impulsados por el modelo de lenguaje grande (LLM) se despliegan cada vez más para realizar tareas complejas en el mundo real, se enfrentan a una clase creciente de ataques que explotan

La noticia

arXiv:2605.03228v2 Announce Type: replace Abstract: As large language model (LLM)-powered agents are increasingly deployed to perform complex, real-world tasks, they face a growing class of attacks that exploit extended user-agent-environment interactions to pursue malicious objectives improbable in single-turn settings. Such long-horizon threats pose significant risks to the safe deployment of LLM agents in critical domains. In this paper, we present ShadowMem, a novel defensive framework designed to counter a wide range of long-horizon threats. Inspired by the "shadow stack" abstraction in systems security, ShadowMem maintains a dedicated, safety-focused agentic memory that distills and retains safety-critical context across the agent's full execution trajectory, leveraging this shadow memory to proactively assess the