EUVIMEDEUVIMED entdecken
Crossref

A More Context-Aware Approach for Textual Adversarial Attacks Using Probability Difference-Guided Beam Search

Huijun Liu, Bin Ji, Jie Yu, Shasha Li, Jun Ma, Zibo Yi, Mengxue Du, Miaomiao Li, Jie Liu, Zeyao Mo

IEEE Transactions on Knowledge and Data Engineering · 2024-06 · 36 (6) · 2316-2328

DOI: 10.1109/tkde.2023.3325315 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Knowledge and Data Engineering, 2024-06. Verfasst von Huijun Liu und 9 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 40 Literaturangaben. Gefördert von: Natural Science Foundation of Hunan Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Textual adversarial attacks expose the vulnerabilities of text classifiers and can be used to improve their robustness. Previous context-aware attack models suffer from several limitations. They generally rely on out-of-date substitutes, solely consider the gold label probability, and use the greedy search when generating adversarial examples, often limiting the attack efficiency. To tackle these issues, we propose MC-PDBS, a More Context-aware textual adversarial attack model using Probability Difference-guided Beam Search. MC-PDBS generates substitutes using the newest perturbed text sequences in each attack iteration, enabling the generation of more context-aware adversarial examples. The probability difference is an overall consideration of the probabilities of all class labels, which is more effective than the gold label probability in guiding the selection of attack paths. In addition, the beam search enables MC-PDBS to search attack paths from multiple search channels, thereby avoiding the limited search space problem. Extensive experiments and human evaluation demonstrate that MC-PDBS outperforms previous best models in a series of evaluation metrics, particularly bringing up to a +19.5% attack success rate. Extensive analyses further confirm the effectiveness of MC-PDBS.

Förderung

  • Natural Science Foundation of Hunan Province

Zitieren

APA

Liu, H., Ji, B., Yu, J., Li, S., Ma, J., Yi, Z., Du, M., Li, M., Liu, J., & Mo, Z. (2024). A More Context-Aware Approach for Textual Adversarial Attacks Using Probability Difference-Guided Beam Search. IEEE Transactions on Knowledge and Data Engineering, 36(6), 2316-2328. https://doi.org/10.1109/tkde.2023.3325315

Vancouver

Liu H, Ji B, Yu J, Li S, Ma J, Yi Z, et al. A More Context-Aware Approach for Textual Adversarial Attacks Using Probability Difference-Guided Beam Search. IEEE Transactions on Knowledge and Data Engineering. 2024;36(6):2316-2328. doi:10.1109/tkde.2023.3325315

BibTeX
@article{Liu2024,
  author = {Liu, Huijun and Ji, Bin and Yu, Jie and Li, Shasha and Ma, Jun and Yi, Zibo and Du, Mengxue and Li, Miaomiao and Liu, Jie and Mo, Zeyao},
  title = {A More Context-Aware Approach for Textual Adversarial Attacks Using Probability Difference-Guided Beam Search},
  journal = {IEEE Transactions on Knowledge and Data Engineering},
  year = {2024},
  volume = {36},
  number = {6},
  pages = {2316--2328},
  doi = {10.1109/tkde.2023.3325315}
}
Kostenlos starten →