hZACH-ViT: Curved Latent Geometry for Compact Vision Transformers in Low-Data Medical Imaging 事件

hZACH-ViT: Curved Latent Geometry for Compact Vision Transformers in Low-Data Medical Imaging arXiv:2606.00906v1 Announce Type: new Abstract: Compact Vision Transformers are attractive for medical imaging in low-data and resource-constrained settings, but most existing variants assume that Euclidean latent geometry is sufficient for organizing image representations. We introduce hZACH-ViT, a family of curved-geometry extensions of ZACH-ViT, a compact zero-token Vision Transformer that removes p