LATIDIA · Ciberseguridad

Seguridad en agentes autoevolutivos: una encuesta

arXiv:2610.00093v1 Anuncio Tipo: nuevo Resumen: Los modelos de lenguaje grandes (LLM) exhiben fuertes capacidades generales, sin embargo, sus parámetros suelen permanecer fijos después de la implementación, lo que limita el aprendizaje a partir de nuevas interacciones. En

La noticia

arXiv:2610.00093v1 Announce Type: new Abstract: Large language models (LLMs) exhibit strong general capabilities, yet their parameters typically remain fixed after deployment, limiting learning from new interactions. In open-ended environments, this motivates self-evolving agents that continually update reusable state-including model parameters, memories, tool definitions, skills, and workflows-from data, feedback, and accumulated experience. This shift changes the safety problem: once experience becomes reusable state, past events become future causes, and information harmless in one context may later influence decisions with greater persistence, authority, or scope. Self-evolving agent safety therefore asks not only whether a response is aligned or an action authorized, but whether safety properties survive the accumulation, generalization, and cross-context reuse