Lujia Shen (Zhejiang University), Yuwen Pu (Zhejiang University), Shouling Ji (Zhejiang University), Changjiang Li (Penn State), Xuhong Zhang (Zhejiang University), Chunpeng Ge (Shandong University), Ting Wang (Penn State)

Transformer-based models, such as BERT and GPT, have been widely adopted in natural language processing (NLP) due to their exceptional performance. However, recent studies show their vulnerability to textual adversarial attacks where the model's output can be misled by intentionally manipulating the text inputs. Despite various methods that have been proposed to enhance the model's robustness and mitigate this vulnerability, many require heavy consumption resources (e.g., adversarial training) or only provide limited protection (e.g., defensive dropout).

In this paper, we propose a novel method called dynamic attention, tailored for the transformer architecture, to enhance the inherent robustness of the model itself against various adversarial attacks. Our method requires no downstream task knowledge and does not incur additional costs. The proposed dynamic attention consists of two modules: *(i) attention rectification*, which masks or weakens the attention value of the chosen tokens, and *(ii) dynamic modeling*, which dynamically builds the set of candidate tokens. Extensive experiments demonstrate that dynamic attention significantly mitigates the impact of adversarial attacks, improving up to 33% compared to previous methods against widely used adversarial attacks. The model-level design of dynamic attention enables it to be easily combined with other defense methods (e.g., adversarial training) to further enhance the model's robustness. Furthermore, we demonstrate that dynamic attention preserves the state-of-the-art robustness space of the original model compared to other dynamic modeling methods.

View More Papers

PrintListener: Uncovering the Vulnerability of Fingerprint Authentication via the...

Man Zhou (Huazhong University of Science and Technology), Shuao Su (Huazhong University of Science and Technology), Qian Wang (Wuhan University), Qi Li (Tsinghua University), Yuting Zhou (Huazhong University of Science and Technology), Xiaojing Ma (Huazhong University of Science and Technology), Zhengxiong Li (University of Colorado Denver)

Read More

Content Censorship in the InterPlanetary File System

Srivatsan Sridhar (Stanford University), Onur Ascigil (Lancaster University), Navin Keizer (University College London), François Genon (UCLouvain), Sébastien Pierre (UCLouvain), Yiannis Psaras (Protocol Labs), Etienne Riviere (UCLouvain), Michał Król (City, University of London)

Read More

Maginot Line: Assessing a New Cross-app Threat to PII-as-Factor...

Fannv He (National Computer Network Intrusion Protection Center, University of Chinese Academy of Sciences, China), Yan Jia (DISSec, College of Cyber Science, Nankai University, China), Jiayu Zhao (National Computer Network Intrusion Protection Center, University of Chinese Academy of Sciences, China), Yue Fang (National Computer Network Intrusion Protection Center, University of Chinese Academy of Sciences, China),…

Read More