Work

BAE: BERT-based Adversarial Examples for Text Classification

Year: 2020

Type: conference-paper

Abstract: Modern text classification models are susceptible to adversarial examples, perturbed versions of the original text indiscernible by humans which get misclassified by the model.Recent works in NLP use rule-based synonym replacement strategies to generate adversarial examples.These strategies can lead to outof-context and unnaturally complex token replacements, which are easily identifiable by humans.We present BAE, a black box attack for generating adversarial examples using contextual perturbations from a BERT masked language model.BAE replaces and inserts tokens in the original text by masking a portion of the text and leveraging the BERT-MLM to generate alternatives for the masked tokens.Through automatic and human evaluations, we show that BAE performs a stronger attack, in addition to generating adversarial examples with improved grammaticality and semantic coherence as compared to prior work.

Source: Conference on Empirical Methods in Natural Language Processing (EMNLP)

DOI: 10.18653/v1/2020.emnlp-main.498

Authors: Siddhant Garg, Goutham Ramakrishnan

Institutions: Amazon (United States), University of Wisconsin–Madison

Language: English

Topic: Adversarial Robustness in Machine Learning

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Funder: National Science Foundation

Award: CAREER: Algorithmic Foundations and Modern Applications for Program Synthesis