Yuzhe Sha
1 indexed paper
Recent (6 mo)
1With code
0Influential cites
0Benchmarked
0Publications per year
126
Top categories
Crypto×1Vision×1
Frequent co-authors
Research Timeline
2026
Scaling Exposes the Trigger: Input-Level Backdoor Detection in Text-to-Image Diffusion Models via Cross-Attention Scaling
The paper introduces SET, a robust input-level backdoor detection framework that detects hidden malicious triggers in text-to-image diffusion models by analyzing systematic differences in how benign and backdoor inputs respond to controlled cross-attention scaling perturbations.
Highlighted terms show continued research focus across papers
Papers
cs.CRcs.CVRecentApr 14, 2026
Scaling Exposes the Trigger: Input-Level Backdoor Detection in Text-to-Image Diffusion Models via Cross-Attention Scaling
Zida Li, Jun Li, Yuzhe Sha, Ziqiang Li +2 more
The paper introduces SET, a robust input-level backdoor detection framework that detects hidden malicious triggers in text-to-image diffusion models by analyzing systematic differences in how benign a…
View →