TextAttack: A Framework for Adversarial Attacks, Data Augmentation, and Adversarial Training in NLP

TL;DR

TextAttack is a flexible Python framework that simplifies creating, applying, and combining adversarial attacks, data augmentation, and adversarial training to improve NLP model robustness and performance across various datasets and models.

Contribution

It introduces a modular, extensible framework for NLP adversarial attacks, enabling easy construction and application of attacks, data augmentation, and training techniques.

Findings

Includes 16 implemented adversarial attacks from literature.

Supports models like BERT and all GLUE tasks.

Facilitates easy experimentation with adversarial methods.

Abstract

While there has been substantial research using adversarial attacks to analyze NLP models, each attack is implemented in its own code repository. It remains challenging to develop NLP attacks and utilize them to improve model performance. This paper introduces TextAttack, a Python framework for adversarial attacks, data augmentation, and adversarial training in NLP. TextAttack builds attacks from four components: a goal function, a set of constraints, a transformation, and a search method. TextAttack's modular design enables researchers to easily construct attacks from combinations of novel and existing components. TextAttack provides implementations of 16 adversarial attacks from the literature and supports a variety of models and datasets, including BERT and other transformers, and all GLUE tasks. TextAttack also includes data augmentation and adversarial training modules for using components of adversarial attacks to improve model accuracy and robustness. TextAttack is democratizing NLP: anyone can try data augmentation and adversarial training on any model or dataset, with just a few lines of code. Code and tutorials are available at https://github.com/QData/TextAttack.