SHAPAttack: Shapley-Guided Multigranularity Adversarial Attack Against Text Transformers
IEEE Intelligent Systems · 2024-05 · 39 (3) · 45-53
DOI: 10.1109/mis.2024.3379377 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Intelligent Systems, 2024-05. Verfasst von Jiahui Shi und 2 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 20 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Despite the great success of text transformers, recent studies have revealed their vulnerability to textual adversarial attacks. Existing attack methods are limited to a single granularity and often suffer from a low attack success rate and a high query cost. To mitigate these issues, we propose a Shapley-guided multigranularity adversarial attack (SHAPAttack) that generates adversarial examples (AEs). SHAPAttack expands the perturbation space by combining granularities at both the word and phrase levels, which enhances the diversity of the generated AEs. To improve attack efficiency and reduce the query cost, SHAPAttack adopts a query-free constituent importance ranking method guided by the Shapley value to measure the importance of each constituent. We conduct extensive experiments on three benchmark datasets across three text transformers. The experimental results demonstrate that SHAPAttack outperforms strong baselines in terms of both attack success rate and model queries, indicating the effectiveness and efficiency of the proposed method.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Shi, J., Li, L., & Zeng, D. (2024). SHAPAttack: Shapley-Guided Multigranularity Adversarial Attack Against Text Transformers. IEEE Intelligent Systems, 39(3), 45-53. https://doi.org/10.1109/mis.2024.3379377
Vancouver
Shi J, Li L, Zeng D. SHAPAttack: Shapley-Guided Multigranularity Adversarial Attack Against Text Transformers. IEEE Intelligent Systems. 2024;39(3):45-53. doi:10.1109/mis.2024.3379377
BibTeX
@article{Shi2024,
author = {Shi, Jiahui and Li, Linjing and Zeng, Daniel},
title = {SHAPAttack: Shapley-Guided Multigranularity Adversarial Attack Against Text Transformers},
journal = {IEEE Intelligent Systems},
year = {2024},
volume = {39},
number = {3},
pages = {45--53},
doi = {10.1109/mis.2024.3379377}
}
EUVIMED