Pith. sign in

REVIEW

Text Adversarial Purification as Defense against Adversarial Attacks

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2203.14207 v2 pith:Q34DWK5M submitted 2022-03-27 cs.CL

Text Adversarial Purification as Defense against Adversarial Attacks

classification cs.CL
keywords adversarialpurificationattacksmodelsdefensetextualattacklanguage
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Adversarial purification is a successful defense mechanism against adversarial attacks without requiring knowledge of the form of the incoming attack. Generally, adversarial purification aims to remove the adversarial perturbations therefore can make correct predictions based on the recovered clean samples. Despite the success of adversarial purification in the computer vision field that incorporates generative models such as energy-based models and diffusion models, using purification as a defense strategy against textual adversarial attacks is rarely explored. In this work, we introduce a novel adversarial purification method that focuses on defending against textual adversarial attacks. With the help of language models, we can inject noise by masking input texts and reconstructing the masked texts based on the masked language models. In this way, we construct an adversarial purification process for textual models against the most widely used word-substitution adversarial attacks. We test our proposed adversarial purification method on several strong adversarial attack methods including Textfooler and BERT-Attack and experimental results indicate that the purification algorithm can successfully defend against strong word-substitution attacks.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.