LATIDIA · Ciberseguridad

KVGuard: un comprobador de seguridad de caché KV para sistemas de servicio LLM

arXiv: 2611.06138v1Tipo de anuncio: nuevo Resumen: Los sistemas de servicio LLM modernos administran la caché KV a través de PagedAttention, un asignador de GPU personalizado que divide una sola región asignada a CUDA en bloques físicos de tamaño fijo

La noticia

arXiv:2611.06138v1 Announce Type: new Abstract: Modern LLM serving systems manage the KV cache through PagedAttention, a custom GPU allocator that partitions a single CUDA-allocated region into fixed-size physical blocks and maps each request to those blocks through a CPU-built block table. This block-based design exposes a previously unstudied class of memory safety bugs in the KV cache, including double-free, use-after-free, out-of-bounds access, uninitialized read, and permission violation, which silently corrupt outputs or break request isolation without ever crashing the system. Existing memory sanitizers cannot detect them, because they reason about CUDA-level addresses rather than KV cache blocks, requests, and block tables, and their redzone- and quarantine-based designs do not fit