Sparsity Should Be ALLOTted by Output: Training-Free Allocation of Block-Sparse Attention for Video Diffusion Transformers
Paper • 2610.29441 • Published • 23 •
3
👋 Open to Work
Tianze Xia
Mukayi01
·
AI & ML interests
Computer Vision, MLLM, Diffusion Model
Recent Activity
commentedon a paper 3 days ago
Sparsity Should Be ALLOTted by Output: Training-Free Allocation of Block-Sparse Attention for Video Diffusion TransformersOrganizations
commented a
paper 3 days ago