Scaling Exposes the Trigger: Input-Level Backdoor Detection in Text-to-Image Diffusion Models via Cross-Attention Scaling

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Li, Zida, Li, Jun, Sha, Yuzhe, Li, Ziqiang, Xiong, Lizhi, Fu, Zhangjie
Format:	Preprint
Published:	2026
Subjects:	Cryptography and Security Computer Vision and Pattern Recognition
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Internet

https://arxiv.org/abs/2604.12446