PARK: Accurate Block Retrieval for Sparse Attention in Video Diffusion Transformers figure
AlphaXiv 中文概览(可滚动查看)