LATIDIA · Ciberseguridad

SoK: PIAtlas: Descomposición del espacio ilimitado de ataque de inyección rápida

arXiv: 2610.24999v1Tipo de anuncio: nuevo Resumen: La inyección rápida es el riesgo de seguridad dominante para las aplicaciones y agentes de modelo de lenguaje grande (LLM), sin embargo, el campo aún carece de un mapa coherente de cómo se ejecutan estos ataques

La noticia

arXiv:2610.24999v1 Announce Type: new Abstract: Prompt injection is the dominant security risk for large-language-model (LLM) applications and agents, yet the field still lacks a coherent map of how these attacks are constructed. More than twenty public datasets describe prompt injection using incompatible vocabularies, often organized around delivery vector, attacker goal, or generation method. In this Systematization of Knowledge, we present PIAtlas, a taxonomy organized instead around how the injected instruction is constructed to subvert the underlying model. Its six classes, one per subversion mechanism, form the roots of a hierarchy of techniques and sub-techniques that researchers can use to describe attacks consistently and design more systematic evaluations. We derive the