# THAT: Two Head Adversarial Training for Improving Robustness at Scale Source: [THAT: Two Head Adversarial Training for Improving Robustness at Scale]() Collection index: [llms.txt]() Data updated: 2025-10-25T23:03:46.301Z Authors: Zuxuan Wu, Tom Goldstein, Larry S. Davis, Ser-Nam Lim Published: 2021-03-25T05:32:38.000Z ## Abstract Many variants of adversarial training have been proposed, with most research focusing on problems with relatively few classes. In this paper, we propose Two Head Adversarial Training (THAT), a two-stream adversarial learning network that is designed to handle the large-scale many-class ImageNet dataset. The proposed method trains a network with two heads and two loss functions; one to minimize feature-space domain shift between natural and adversarial images, and one to promote high classification accuracy. This combination delivers a hardened network that achieves state of the art robust accuracy while maintaining high natural accuracy on ImageNet. Through extensive experiments, we demonstrate that the proposed framework outperforms alternative methods under both standard and "free" adversarial training settings. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. ## Original sources and implementation [Original paper]() [PDF]()