{"slug": "nuvit-vision-transformers-for-fourier-space-data", "title": "NUViT: Vision Transformers for Fourier Space Data", "summary": "Researchers posted arXiv:2610.32041v1, introducing Non-Uniform ViT (NUViT), a general-purpose vision transformer that redesigns image tokenization, data representation and positional encoding to respect the frequency and phase structure of Fourier-space data. NUViT outperformed standard architectures on Fourier-domain image classification, 2D crystal lattice identification, MRI annotation and cryo-EM protein heterogeneity analysis, according to the paper's abstract. The work targets modalities such as medical imaging and microscopy, where standard ViT square-patch tokenization on a Cartesian grid poorly fits Fourier data.", "body_md": "arXiv:2610.32041v1 Announce Type: new \nAbstract: Vision Transformers (ViTs) have demonstrated strong performance in many computer vision tasks but do not naturally extend beyond the natural image domain. One important example is Fourier-space data, which is used to represent a rich and diverse set of modalities, including medical imaging and microscopy. Standard ViT tokenization partitions an image into square patches on a Cartesian grid, which poorly fit the frequency and phase structure of Fourier-space data. We introduce Non-Uniform ViT, a simple, general-purpose vision transformer that redesigns image tokenization, data representation, and positional encoding to respect this structure. Experiments on Fourier-domain image classification, 2D crystal lattice identification, MRI annotation, and cryo-EM protein heterogeneity analysis show that NUViT consistently captures the structure of Fourier data better than standard architectures.", "url": "https://wpnews.pro/news/nuvit-vision-transformers-for-fourier-space-data", "canonical_source": "https://arxiv.org/abs/2610.32041", "published_at": "2026-10-26 04:00:00+00:00", "updated_at": "2026-10-26 04:12:36.200465+00:00", "lang": "en", "topics": ["computer-vision", "machine-learning", "ai-research", "artificial-intelligence"], "entities": ["NUViT", "Non-Uniform ViT", "arXiv"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/nuvit-vision-transformers-for-fourier-space-data", "markdown": "https://wpnews.pro/news/nuvit-vision-transformers-for-fourier-space-data.md", "text": "https://wpnews.pro/news/nuvit-vision-transformers-for-fourier-space-data.txt", "jsonld": "https://wpnews.pro/news/nuvit-vision-transformers-for-fourier-space-data.jsonld"}}