ADDMU: Detection of Far-Boundary Adversarial Examples with Data and Model Uncertainty Estimation

Fan Yin, Yao Li, Cho-Jui Hsieh, Kai-Wei Chang


Abstract
Adversarial Examples Detection (AED) is a crucial defense technique against adversarial attacks and has drawn increasing attention from the Natural Language Processing (NLP) community. Despite the surge of new AED methods, our studies show that existing methods heavily rely on a shortcut to achieve good performance. In other words, current search-based adversarial attacks in NLP stop once model predictions change, and thus most adversarial examples generated by those attacks are located near model decision boundaries. To surpass this shortcut and fairly evaluate AED methods, we propose to test AED methods with Far Boundary (FB) adversarial examples. Existing methods show worse than random guess performance under this scenario. To overcome this limitation, we propose a new technique, ADDMU, adversary detection with data and model uncertainty, which combines two types of uncertainty estimation for both regular and FB adversarial example detection. Our new method outperforms previous methods by 3.6 and 6.0 AUC points under each scenario. Finally, our analysis shows that the two types of uncertainty provided by ADDMU can be leveraged to characterize adversarialexamples and identify the ones that contribute most to model’s robustness in adversarial training.
Anthology ID:
2022.emnlp-main.440
Volume:
Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates
Editors:
Yoav Goldberg, Zornitsa Kozareva, Yue Zhang
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
6567–6584
Language:
URL:
https://aclanthology.org/2022.emnlp-main.440
DOI:
10.18653/v1/2022.emnlp-main.440
Bibkey:
Cite (ACL):
Fan Yin, Yao Li, Cho-Jui Hsieh, and Kai-Wei Chang. 2022. ADDMU: Detection of Far-Boundary Adversarial Examples with Data and Model Uncertainty Estimation. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 6567–6584, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
Cite (Informal):
ADDMU: Detection of Far-Boundary Adversarial Examples with Data and Model Uncertainty Estimation (Yin et al., EMNLP 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.emnlp-main.440.pdf