Abstract
We propose the first general-purpose gradient-based adversarial attack against transformer models. Instead of searching for a single adversarial example, we search for a distribution of adversarial examples parameterized by a continuous-valued matrix, hence enabling gradient-based optimization. We empirically demonstrate that our white-box attack attains state-of-the-art attack performance on a variety of natural language tasks, outperforming prior work in terms of adversarial success rate with matching imperceptibility as per automated and human evaluation. Furthermore, we show that a powerful black-box transfer attack, enabled by sampling from the adversarial distribution, matches or exceeds existing methods, while only requiring hard-label outputs.- Anthology ID:
- 2021.emnlp-main.464
- Volume:
- Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
- Month:
- November
- Year:
- 2021
- Address:
- Online and Punta Cana, Dominican Republic
- Editors:
- Marie-Francine Moens, Xuanjing Huang, Lucia Specia, Scott Wen-tau Yih
- Venue:
- EMNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 5747–5757
- Language:
- URL:
- https://aclanthology.org/2021.emnlp-main.464
- DOI:
- 10.18653/v1/2021.emnlp-main.464
- Cite (ACL):
- Chuan Guo, Alexandre Sablayrolles, Hervé Jégou, and Douwe Kiela. 2021. Gradient-based Adversarial Attacks against Text Transformers. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 5747–5757, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.
- Cite (Informal):
- Gradient-based Adversarial Attacks against Text Transformers (Guo et al., EMNLP 2021)
- PDF:
- https://preview.aclanthology.org/naacl-24-ws-corrections/2021.emnlp-main.464.pdf
- Code
- facebookresearch/text-adversarial-attack + additional community code
- Data
- AG News, IMDb Movie Reviews, MultiNLI