将热核先验引入流形变分框架，为贝叶斯深度学提供新的几何先验设计思路。
arXiv:2606.18658v1 Announce Type: new Abstract: Learning unsupervised representations of medical imaging cohorts can reveal clinically meaningful prot…
提出利用投影正则化实现近似等变性的新方法，为几何深度学习提供更灵活的理论框架
arXiv:2601.05028v2 Announce Type: replace Abstract: Equivariance is a powerful inductive bias in neural networks, improving generalisation and physica…
被ICML 2026接收，提出在全秩相关矩阵上构建黎曼网络的新几何深度学习方法。
arXiv:2605.19073v1 Announce Type: new Abstract: Representations on the Symmetric Positive Definite (SPD) manifold have garnered significant attention …
点云学习新方法，用微分形式编码高阶切信息，突破传统坐标与距离限制。
arXiv:2605.15524v1 Announce Type: cross Abstract: Point cloud learning often rests on the premise that observed samples are noisy traces of an underly…