LATIDIA · Ciberseguridad

Origen es todo lo que necesita: Transformadores conscientes de la procedencia para la separación de límites de confianza estructural

arXiv:2609.21088v1 Tipo de anuncio: nuevo Resumen: La inyección rápida indirecta (IPI) sigue siendo un desafío central de seguridad para los sistemas de modelo de lenguaje grande (LLM) porque los transformadores estándar carecen de notificación arquitectónica

La noticia

arXiv:2609.21088v1 Announce Type: new Abstract: Indirect prompt injection (IPI) remains a central safety and security challenge for large language model (LLM) systems because standard transformers lack architectural notion of source authority. Retrieved documents, user inputs, and system instructions are all processed through the same undifferentiated attention mechanism, forcing the model to infer from wording alone what should be obeyed and what should be treated as data. We propose Provenance-Aware Transformers, a provenance-aware defense that makes application-supplied source labels actionable inside the model. Each input token is assigned a ring ID encoding its origin, and the model is augmented with origin embeddings, a learnable origin attention bias, and a learnable origin