method · first seen 20 Oct 2026
sparse attention for video transformers
Techniques that restrict attention in video diffusion or video transformers to reduce computation over spatiotemporal tokens.
Mentions per week
PapersCodeDiscussion
Earliest sources
- KPage: Cross-Block Page Compaction for Hardware-Efficient Sparse AttentionPapers · 20 Oct 2026 · Junwei Feng, Qiaorui Chen, Chenyu Liang, Zhen Guo, Jingyi Yang, Xiaoling Zhou, J…
Latest sources
- KPage: Cross-Block Page Compaction for Hardware-Efficient Sparse AttentionPapers · 20 Oct 2026 · Junwei Feng, Qiaorui Chen, Chenyu Liang, Zhen Guo, Jingyi Yang, Xiaoling Zhou, J…