English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Backdoor attacks, which maliciously control a well-trained model's outputs ofthe instances with specific triggers, are recently shown to be serious threatsto the safety of reusing deep neural networks (DNNs). In this work, we proposean efficient online defense mechanism based on robustness-aware perturbations.Specifically, by analyzing the backdoor training process, we point out thatthere exists a big gap of robustness between poisoned and clean samples.Motivated by this observation, we construct a word-based robustness-awareperturbation to distinguish poisoned samples from clean samples to defendagainst the backdoor attacks on natural language processing (NLP) models.Moreover, we give a theoretical analysis about the feasibility of ourrobustness-aware perturbation-based defense method. Experimental results onsentiment analysis and toxic detection tasks show that our method achievesbetter defending performance and much lower computational costs than existingonline defense methods. Our code is available athttps://github.com/lancopku/RAP.

RAP: Robustness-Aware Perturbations for Defending against Backdoor Attacks on NLP Models