Full 3D attention is costly in video Diffusion Transformers. Block-sparse attention reduces this cost by computing only selected attention blocks. However, existing methods either overlook the potenti
1 comments
Sparsity Should Be ALLOTted by Output: Training-Free Allocation of Block-Sparse Attention for Video Diffusion Transformers was first seen on Hugging Face · Daily papers on Oct 22, 2026. Since then it appeared in 12 hourly checks across 2 days, peaking at #11 on Oct 22, 05:18 UTC. It was last on the list on Oct 23, 01:18 UTC.
Position on Hugging Face · Daily papers at every check (lower is better).
Best rank per day (UTC).
Platform metric as reported by Hugging Face.
Show that this trended, linking back to this page.
<a href="https://trendoday.com/hugging-face/papers/2610.29441-sparsity-should-be-allotted-by-output-training-free"><img src="https://trendoday.com/badge/hugging-face/papers/2610.29441-sparsity-should-be-allotted-by-output-training-free.svg" alt="Sparsity Should Be ALLOTted by Output: Training-Free Allocation of Block-Sparse Attention for Video Diffusion Transformers on TrendBoard" height="44"></a>
[](https://trendoday.com/hugging-face/papers/2610.29441-sparsity-should-be-allotted-by-output-training-free)