2022/04/26 by Daquan Zhou, Zhiding Yu, Zhou, Daquan +11 · 7 citations
Computer Science · #Advanced Neural Network Applications #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Visual Attention and Saliency Detection
paper · doi:10.48550/arxiv.2204.12451
openalex publication_date 2022/04/26 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Recent studies show that Vision Transformers(ViTs) exhibit strong robustness against various corruptions. Although this property is partly attributed to the self-attention mechanism, there is still a lack of systematic understanding. In this paper, we examine the role of self-attention in learning robust representations. Our study is motivated by the intriguing properties of the emerging visual grouping in Vision Transformers, which indicates that self-attention may promote robustness through improved mid-level representations. We further propose a family of fully attentional networks (FANs) that strengthen this capability by incorporating an attentional channel processing design. We validate the design comprehensively on various hierarchical backbones. Our model achieves a state-of-the-art 87.1% accuracy and 35.8% mCE on ImageNet-1k and ImageNet-C with 76.8M parameters. We also demonstrate state-of-the-art accuracy and robustness in two downstream tasks: semantic segmentation and object detection. Code is available at: https://github.com/NVlabs/FAN.