LATIDIA · Ciberseguridad

No lea el registro: la ejecución rastrea los verificadores de contaminación en los agentes de generación de vídeo

arXiv: 2609.28564v1Tipo de anuncio: nuevo Resumen: Los sistemas de generación de video genéticos cierran un bucle entre un generador y un verificador: un LLM planifica tomas, llama a un modelo de texto a video y un juez multimodal decide si el

La noticia

arXiv:2609.28564v1 Announce Type: new Abstract: Agentic video-generation systems close a loop between a generator and a verifier: an LLM plans shots, calls a text-to-video model, and a multimodal judge decides whether the result satisfies the request. To diagnose where a long workflow fails, recent harnesses deliberately show the judge more than the video-the agent's execution trace, its plan, the narration it synthesized. We ask whether this auxiliary text moves the judge's verdict on purely \emph{visual} requirements, holding the frames fixed. On a benchmark of 109 generated two-event clips with manual labels, in which the requested event is either visibly completed or visibly missing, a trace that reports a successful tool