Jiashu Zhang
Titles as the sources gave them. In 1 entry, with 0 checked quotations.
In the Base
Every entry they appear in.
AUTEUR: Acoustic Plan Learning for Red-Teaming Text-to-Audio Models
A preprint presents AUTEUR, an automated red-teaming method for text-to-audio models that learns structured acoustic plans to produce test prompts passing text screening but yielding harmful audio. The authors report attack success of 79.4% on TangoFlux and 95.6% on Kling.