Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
May 19, 2024

Text-CRS: A Generalized Certified Robustness Framework against Textual Adversarial Attacks

View Full Paper
Ask AI
Bookmark
Share

Authors

XZXinyu ZhangHHHanbin HongYHYuan Hong

Discussion

Loading...

Member takes

Overview

Computational study demonstrates certified robustness across four word-level adversarial operations in language models, indicating improved defense against text-based attacks.

Key Points

  • Text-CRS establishes certified robustness against four word-level adversarial operations in classification models, achieving significant accuracy improvements over previous methods.
  • Theoretical modeling with randomized smoothing evaluates word alterations by mapping perturbations into combined permutation and embedding transformation spaces.
  • Highlights a generalized certification benchmark for diverse text attacks, outperforming state-of-the-art defenses against synonym substitution across multiple language models.

Cite This Study

Zhang et al. (2024) studied this question.

synapsesocial.com/papers/68e696f4b6db64358761d022https://doi.org/10.1109/sp54263.2024.00053
View Full Paper
Ask AI
Bookmark
Share