no code implementations • 8 Apr 2024 • Yihe Fan, Yuxin Cao, Ziyu Zhao, Ziyao Liu, Shaofeng Li
Multimodal Large Language Models (MLLMs) demonstrate remarkable capabilities that increasingly influence various aspects of our daily lives, constantly defining the new boundary of Artificial General Intelligence (AGI).
no code implementations • 3 Feb 2024 • Lu Chen, Shaofeng Li, Benhao Huang, Fan Yang, Zheng Li, Jie Li, Yuan Luo
In the context of deep neural networks, we expose the existence of a harmless perturbation space, where perturbations leave the network output entirely unaltered.
no code implementations • CVPR 2022 • Zirui Peng, Shaofeng Li, Guoxing Chen, Cheng Zhang, Haojin Zhu, Minhui Xue
In this paper, we propose a novel and practical mechanism which enables the service provider to verify whether a suspect model is stolen from the victim model via model extraction attacks.
1 code implementation • 19 Nov 2021 • Ruoxi Sun, Minhui Xue, Gareth Tyson, Tian Dong, Shaofeng Li, Shuo Wang, Haojin Zhu, Seyit Camtepe, Surya Nepal
We find that (i) commercial antivirus engines are vulnerable to AMM-guided test cases; (ii) the ability of a manipulated malware generated using one detector to evade detection by another detector (i. e., transferability) depends on the overlap of features with large AMM values between the different detectors; and (iii) AMM values effectively measure the fragility of features (i. e., capability of feature-space manipulation to flip the prediction results) and explain the robustness of malware detectors facing evasion attacks.
1 code implementation • 1 May 2021 • Shaofeng Li, Hui Liu, Tian Dong, Benjamin Zi Hao Zhao, Minhui Xue, Haojin Zhu, Jialiang Lu
We are able to demonstrate the adversary's high success rate of attacks, while maintaining functionality for regular users, with triggers inconspicuous by the human administrators.
no code implementations • 16 Jul 2020 • Shaofeng Li, Shiqing Ma, Minhui Xue, Benjamin Zi Hao Zhao
The trigger can take a plethora of forms, including a special object present in the image (e. g., a yellow pad), a shape filled with custom textures (e. g., logos with particular colors) or even image-wide stylizations with special filters (e. g., images altered by Nashville or Gotham filters).
1 code implementation • 6 Sep 2019 • Shaofeng Li, Minhui Xue, Benjamin Zi Hao Zhao, Haojin Zhu, Xinpeng Zhang
We show that the proposed invisible backdoors can be fairly effective across various DNN models as well as four datasets MNIST, CIFAR-10, CIFAR-100, and GTSRB, by measuring their attack success rates for the adversary, functionality for the normal users, and invisibility scores for the administrators.