Yang Cao

Dynamic Kernel Graph Sparsifiers

Training Overparametrized Neural Networks in Sublinear Time

A Dynamic Low-Rank Fast Gaussian Transform

Decoupling Sign and Magnitude in Optimization with Adaptive Momentum Scaling

RoPE Attention Can Be Trained in Almost Linear Time

Exploring the Frontiers of Softmax: Provable Optimization, Applications in Diffusion Model, and Beyond

Towards High-Order Mean Flow Generative Models: Feasibility, Expressivity, and Provably Efficient Criteria

Training Tensor Attention Efficiently: From Cubic to Almost Linear Time

HOFAR: High-Order Augmentation of Flow Autoregressive Transformers

Video Latent Flow Matching: Optimal Polynomial Projections for Video Interpolation and Extrapolation

Orthonormal Regularization in Low-Rank Adaptation

Differentially Private Lewis Weight Computation

Fundamental Limits of Crystalline Equivariant Graph Neural Networks: A Circuit Complexity Perspective

Training Multi-Layer Transformers in Almost Linear Time

Visual Autoregressive Transformers Must Use $\Omega(n^2 d)$ Memory

Modern Hopfield Networks Cannot Solve $\mathsf{NC}^1$-Hard Problems

Dynamic Weighted Projection Maintenance with $\ell^p$-Lewis Weight

Fast Block Attention Computation via Dynamic Algorithm

Differentially Private Random Spanning Tree Generation

Efficient $k$-Sparse Band–Limited Interpolation with Improved Approximation Ratio

Faster Algorithms for Structured John Ellipsoid Computation

SORSA: Singular Values and Orthonormal Regularized Singular Vectors Adaptation of Large Language Models