DF-LoGiT: Data-Free Logic-Gated Backdoor Attacks in Vision Transformers

Xiaozuo Shen, Yifei Cai, Rui Ning, Chunsheng Xin, Hongyi Wu
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:110706-110728, 2026.

Abstract

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.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-shen26a, title = {{DF}-{L}o{G}i{T}: Data-Free Logic-Gated Backdoor Attacks in Vision Transformers}, author = {Shen, Xiaozuo and Cai, Yifei and Ning, Rui and Xin, Chunsheng and Wu, Hongyi}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {110706--110728}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/shen26a/shen26a.pdf}, url = {https://proceedings.mlr.press/v306/shen26a.html}, abstract = {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.} }
Endnote
%0 Conference Paper %T DF-LoGiT: Data-Free Logic-Gated Backdoor Attacks in Vision Transformers %A Xiaozuo Shen %A Yifei Cai %A Rui Ning %A Chunsheng Xin %A Hongyi Wu %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-shen26a %I PMLR %P 110706--110728 %U https://proceedings.mlr.press/v306/shen26a.html %V 306 %X 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.
APA
Shen, X., Cai, Y., Ning, R., Xin, C. & Wu, H.. (2026). DF-LoGiT: Data-Free Logic-Gated Backdoor Attacks in Vision Transformers. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:110706-110728 Available from https://proceedings.mlr.press/v306/shen26a.html.

Related Material