Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention
Paper
•
2502.11089
•
Published
•
142
hakorro here
hakorro
·
AI & ML interests
None yet
Recent Activity
upvoted
a
paper
5 days ago
Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse
Attention
upvoted
a
paper
5 days ago
MoBA: Mixture of Block Attention for Long-Context LLMs
upvoted
a
paper
6 months ago
Portrait Video Editing Empowered by Multimodal Generative Priors
Organizations
hakorro's activity
upvoted
2
papers
5 days ago