LATIDIA · Ciberseguridad

Análisis de agentes locales de LLM de peso abierto en tareas de reproducción de vulnerabilidades

arXiv: 2611.06257v1Tipo de anuncio: nuevo Resumen: los modelos de lenguaje de peso abierto que se ajustan a una sola GPU de consumo se están volviendo capaces de realizar una gama cada vez mayor de tareas, y son importantes porque pueden ejecutarse localmente sin enviar

La noticia

arXiv:2611.06257v1 Announce Type: new Abstract: Open-weight language models that fit on a single consumer GPU are becoming capable on a growing range of tasks, and they matter because they can run locally without sending the code to a hosted service. Vulnerability reproduction, turning a vulnerability description into an input that crashes the vulnerable build, is a natural target for them, since it is an automated analysis task closer to test generation than to code comprehension and it runs against a build that must be executed rather than read. In order to assess whether current models can perform it, we evaluate twelve configurations of local models, spanning three generations, two architectures