Mamba-3惊现ICLR 2026投稿：三重升级打满“推理优先”范式
"数值稳定性" 相关文章
In awe at the scale of these tensors – a gentle introduction to Unit-Scaled Maximal Update Parametrization
A Variational Physics-Informed Neural Network Framework Using Petrov-Galerkin Method for Solving Singularly Perturbed Boundary Value Problems
Uncertain but Useful: Leveraging CNN Variability into Data Augmentation
Attribution-based parameter decomposition for linear maps
