attack arXiv Feb 3, 2026 · 8w ago
Xiaozuo Shen, Yifei Cai, Rui Ning et al. · University of Arizona · Iowa State University +1 more
Injects backdoors into ViT checkpoints via weight editing with logic-gated attention triggers, requiring no training data
Model Poisoning vision
The widespread adoption of Vision Transformers (ViTs) elevates supply-chain risk on third-party model hubs, where an adversary can implant backdoors into released checkpoints. Existing ViT backdoor attacks largely rely on poisoned-data training, while prior data-free attempts typically require synthetic-data fine-tuning or extra model components. This paper introduces Data-Free Logic-Gated Backdoor Attacks (DF-LoGiT), a truly data-free backdoor attack on ViTs via direct weight editing. DF-LoGiT exploits ViT's native multi-head architecture to realize a logic-gated compositional trigger, enabling a stealthy and effective backdoor. We validate its effectiveness through theoretical analysis and extensive experiments, showing that DF-LoGiT achieves near-100% attack success with negligible degradation in benign accuracy and remains robust against representative classical and ViT-specific defenses.
transformer University of Arizona · Iowa State University · Old Dominion University
ML Security Papers