AugLy: Data Augmentations for Robustness
This provides a tool for researchers and practitioners to enhance model robustness, particularly in social media contexts, but it is incremental as it builds on existing augmentation concepts.
The authors introduced AugLy, a data augmentation library designed to improve adversarial robustness across multiple modalities, and demonstrated its utility by benchmarking it against existing libraries and evaluating the robustness of state-of-the-art models.
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.