LATIDIA · Ciberseguridad

CacheTrap: presentación de un troyano de caja gris más sigiloso contra los LLM

arXiv: 2511.22681v3Tipo de anuncio: reemplazar Resumen: El rápido avance de los modelos de lenguaje grandes (LLM) ha despertado un creciente interés en comprender sus vulnerabilidades de seguridad, particularmente los ataques de troyanos que ena

La noticia

arXiv:2511.22681v3 Announce Type: replace Abstract: The rapid advancement of large language models (LLMs) has sparked growing interest in understanding their security vulnerabilities, particularly Trojan attacks that enable stealthy manipulation of model behavior. Traditional Trojan methods typically alter inputs and/or model weights, relying on white-box assumptions that require access to data or model internal parameters. In this work, we present CacheTrap, the first gray-box Trojan attack targeting the Key-Value (KV) cache of LLMs. This method induces a single-bit flip in the KV cache, serving as a transient trigger. When activated, this trigger causes the model to exhibit targeted actions without changing inputs or model weights. CacheTrap introduces an efficient search algorithm