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

EM Eye: Characterizing Electromagnetic Side-channel Eavesdropping on Embedded Cameras

Yan Long (University of Michigan), Qinhong Jiang (Zhejiang University), Chen Yan (Zhejiang University), Tobias Alam (University of Michigan), Xiaoyu Ji (Zhejiang University), Wenyuan Xu (Zhejiang University), Kevin Fu (Northeastern University)

Read More

Not your Type! Detecting Storage Collision Vulnerabilities in Ethereum...

Nicola Ruaro (University of California, Santa Barbara), Fabio Gritti (University of California, Santa Barbara), Robert McLaughlin (University of California, Santa Barbara), Ilya Grishchenko (University of California, Santa Barbara), Christopher Kruegel (University of California, Santa Barbara), Giovanni Vigna (University of California, Santa Barbara)

Read More

GNNIC: Finding Long-Lost Sibling Functions with Abstract Similarity

Qiushi Wu (University of Minnesota), Zhongshu Gu (IBM Research), Hani Jamjoom (IBM Research), Kangjie Lu (University of Minnesota)

Read More

Pisces: Private and Compliable Cryptocurrency Exchange

Ya-Nan Li (The University of Sydney), Tian Qiu (The University of Sydney), Qiang Tang (The University of Sydney)

Read More