AugLy: Data Augmentations for Robustness

TL;DR

AugLy is a versatile data augmentation library supporting multiple modalities, designed to improve robustness and generate adversarial examples, especially for social media-inspired data, and is benchmarked against existing tools.

Contribution

Introduces AugLy, a new data augmentation library with unique social media-inspired augmentations, enhancing robustness evaluation across audio, image, text, and video modalities.

Findings

AugLy outperforms existing libraries in robustness benchmarks.

It effectively generates adversarial attacks for multiple data types.

Demonstrates improved model robustness evaluation using AugLy.

Abstract

We introduce AugLy, a data augmentation library with a focus on adversarial robustness. AugLy provides a wide array of augmentations for multiple modalities (audio, image, text, & video). These augmentations were inspired by those that real users perform on social media platforms, some of which were not already supported by existing data augmentation libraries. AugLy can be used for any purpose where data augmentations are useful, but it is particularly well-suited for evaluating robustness and systematically generating adversarial attacks. In this paper we present how AugLy works, benchmark it compared against existing libraries, and use it to evaluate the robustness of various state-of-the-art models to showcase AugLy's utility. The AugLy repository can be found at https://github.com/facebookresearch/AugLy.