VORTA: Efficient Video Diffusion via Routing Sparse Attention
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912643762094080 |
|---|---|
| author | Sun, Wenhao Tu, Rong-Cheng Ding, Yifu Jin, Zhao Liao, Jingyi Liu, Shunyu Tao, Dacheng |
| author_facet | Sun, Wenhao Tu, Rong-Cheng Ding, Yifu Jin, Zhao Liao, Jingyi Liu, Shunyu Tao, Dacheng |
| contents | Video diffusion transformers have achieved remarkable progress in high-quality video generation, but remain computationally expensive due to the quadratic complexity of attention over high-dimensional video sequences. Recent acceleration methods enhance the efficiency by exploiting the local sparsity of attention scores; yet they often struggle with accelerating the long-range computation. To address this problem, we propose VORTA, an acceleration framework with two novel components: 1) a sparse attention mechanism that efficiently captures long-range dependencies, and 2) a routing strategy that adaptively replaces full 3D attention with specialized sparse attention variants. VORTA achieves an end-to-end speedup $1.76\times$ without loss of quality on VBench. Furthermore, it can seamlessly integrate with various other acceleration methods, such as model caching and step distillation, reaching up to speedup $14.41\times$ with negligible performance degradation. VORTA demonstrates its efficiency and enhances the practicality of video diffusion transformers in real-world settings. Codes and weights are available at https://github.com/wenhao728/VORTA. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_18809 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | VORTA: Efficient Video Diffusion via Routing Sparse Attention Sun, Wenhao Tu, Rong-Cheng Ding, Yifu Jin, Zhao Liao, Jingyi Liu, Shunyu Tao, Dacheng Computer Vision and Pattern Recognition Video diffusion transformers have achieved remarkable progress in high-quality video generation, but remain computationally expensive due to the quadratic complexity of attention over high-dimensional video sequences. Recent acceleration methods enhance the efficiency by exploiting the local sparsity of attention scores; yet they often struggle with accelerating the long-range computation. To address this problem, we propose VORTA, an acceleration framework with two novel components: 1) a sparse attention mechanism that efficiently captures long-range dependencies, and 2) a routing strategy that adaptively replaces full 3D attention with specialized sparse attention variants. VORTA achieves an end-to-end speedup $1.76\times$ without loss of quality on VBench. Furthermore, it can seamlessly integrate with various other acceleration methods, such as model caching and step distillation, reaching up to speedup $14.41\times$ with negligible performance degradation. VORTA demonstrates its efficiency and enhances the practicality of video diffusion transformers in real-world settings. Codes and weights are available at https://github.com/wenhao728/VORTA. |
| title | VORTA: Efficient Video Diffusion via Routing Sparse Attention |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2505.18809 |