Towards Evaluating the Robustness of Chinese BERT Classifiers

by   Boxin Wang, et al.

Recent advances in large-scale language representation models such as BERT have improved the state-of-the-art performances in many NLP tasks. Meanwhile, character-level Chinese NLP models, including BERT for Chinese, have also demonstrated that they can outperform the existing models. In this paper, we show that, however, such BERT-based models are vulnerable under character-level adversarial attacks. We propose a novel Chinese char-level attack method against BERT-based classifiers. Essentially, we generate "small" perturbation on the character level in the embedding space and guide the character substitution procedure. Extensive experiments show that the classification accuracy on a Chinese news dataset drops from 91.8 than 2 characters on average based on the proposed attack. Human evaluations also confirm that our generated Chinese adversarial examples barely affect human performance on these NLP tasks.


page 1

page 2

page 3

page 4


Glyph-aware Embedding of Chinese Characters

Given the advantage and recent success of English character-level and su...

Expanding Scope: Adapting English Adversarial Attacks to Chinese

Recent studies have revealed that NLP predictive models are vulnerable t...

FireBERT: Hardening BERT-based classifiers against adversarial attack

We present FireBERT, a set of three proof-of-concept NLP classifiers har...

EFSG: Evolutionary Fooling Sentences Generator

Large pre-trained language representation models (LMs) have recently col...

SemAttack: Natural Textual Attacks via Different Semantic Spaces

Recent studies show that pre-trained language models (LMs) are vulnerabl...

Enhancing Model Robustness By Incorporating Adversarial Knowledge Into Semantic Representation

Despite that deep neural networks (DNNs) have achieved enormous success ...

Please sign up or login with your details

Forgot password? Click here to reset