# VSA: Faster Video Diffusion with Trainable Sparse Attention

arXiv: https://arxiv.org/abs/2505.13389 · HTML: https://mlsanity.com/paper/2505.13389/

## Key facts

- Cited by 104 papers in the map (data through 14 Oct 2026).
- Placed in Video generation, one of 14 directions in Diffusion models.
- Builds on 46 of its 49 references that are in the map.
- First posted on arXiv 19 May 2025.

## Placed in

- Diffusion models / [Video generation](https://mlsanity.com/diffusion/video-generation/)

## Builds on (most cited first, 30 of 46)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024)
- [Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow](https://mlsanity.com/paper/2209.03003/) (2022)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Longformer: The Long-Document Transformer](https://mlsanity.com/paper/2004.05150/) (2020)
- [Training Compute-Optimal Large Language Models](https://mlsanity.com/paper/2203.15556/) (2022)
- [Wan: Open and Advanced Large-Scale Video Generative Models](https://mlsanity.com/paper/2503.20314/) (2025)
- [CogVideoX: Text-to-Video Diffusion Models with An Expert Transformer](https://mlsanity.com/paper/2408.06072/) (2024)
- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [VBench: Comprehensive Benchmark Suite for Video Generative Models](https://mlsanity.com/paper/2311.17982/) (2023)
- [Big Bird: Transformers for Longer Sequences](https://mlsanity.com/paper/2007.14062/) (2020)
- [HunyuanVideo: A Systematic Framework For Large Video Generative Models](https://mlsanity.com/paper/2412.03603/) (2024)
- [Generating Long Sequences with Sparse Transformers](https://mlsanity.com/paper/1904.10509/) (2019)
- [DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model](https://mlsanity.com/paper/2405.04434/) (2024)
- [H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models](https://mlsanity.com/paper/2306.14048/) (2023)
- [Improved Distribution Matching Distillation for Fast Image Synthesis](https://mlsanity.com/paper/2405.14867/) (2024)
- [Open-Sora: Democratizing Efficient Video Production for All](https://mlsanity.com/paper/2412.20404/) (2024)
- [Latte: Latent Diffusion Transformer for Video Generation](https://mlsanity.com/paper/2401.03048/) (2024)
- [FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision](https://mlsanity.com/paper/2407.08608/) (2024)
- [Efficient Content-Based Sparse Attention with Routing Transformers](https://mlsanity.com/paper/2003.05997/) (2020)
- [Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention](https://mlsanity.com/paper/2502.11089/) (2025)
- [MInference 1.0: Accelerating Pre-filling for Long-Context LLMs via Dynamic Sparse Attention](https://mlsanity.com/paper/2407.02490/) (2024)
- [LAVIE: High-Quality Video Generation with Cascaded Latent Diffusion Models](https://mlsanity.com/paper/2309.15103/) (2023)
- [Effective Long-Context Scaling of Foundation Models](https://mlsanity.com/paper/2309.16039/) (2023)
- [Open-Sora Plan: Open-Source Large Video Generation Model](https://mlsanity.com/paper/2412.00131/) (2024)

## Built on by (most cited first, 30 of 104)

- [Training Agents Inside of Scalable World Models](https://mlsanity.com/paper/2509.24527/) (2025)
- [Sparse VideoGen2: Accelerate Video Generation with Sparse Attention via Semantic-Aware Permutation](https://mlsanity.com/paper/2505.18875/) (2025)
- [Mixture of Contexts for Long Video Generation](https://mlsanity.com/paper/2508.21058/) (2025)
- [SANA-Video: Efficient Video Generation with Block Linear Diffusion Transformer](https://mlsanity.com/paper/2509.24695/) (2025)
- [LongCat-Video Technical Report](https://mlsanity.com/paper/2510.22200/) (2025)
- [SLA: Beyond Sparsity in Diffusion Transformers via Fine-Tunable Sparse-Linear Attention](https://mlsanity.com/paper/2509.24006/) (2025)
- [End-to-End Training for Autoregressive Video Diffusion via Self-Resampling](https://mlsanity.com/paper/2512.15702/) (2025)
- [FlashVSR: Towards Real-Time Diffusion-Based Streaming Video Super-Resolution](https://mlsanity.com/paper/2510.12747/) (2025)
- [PAN: A World Model for General, Actionable, and Long-Horizon World Simulation](https://mlsanity.com/paper/2511.09057/) (2025)
- [Critique of World Model](https://mlsanity.com/paper/2507.05169/) (2025)
- [Block-Sparse Global Attention for Efficient Multi-View Geometry Transformers](https://mlsanity.com/paper/2509.07120/) (2025)
- [SLA2: Sparse-Linear Attention with Learnable Routing and QAT](https://mlsanity.com/paper/2602.12675/) (2026)
- [Dual-Expert Consistency Model for Efficient and High-Quality Video Generation](https://mlsanity.com/paper/2506.03123/) (2025)
- [Bidirectional Sparse Attention for Faster Video Diffusion Training](https://mlsanity.com/paper/2509.01085/) (2025)
- [OBS-Diff: Accurate Pruning For Diffusion Models in One-Shot](https://mlsanity.com/paper/2510.06751/) (2025)
- [MoGA: Mixture-of-Groups Attention for End-to-End Long Video Generation](https://mlsanity.com/paper/2510.18692/) (2025)
- [SpargeAttention2: Trainable Sparse Attention via Hybrid Top-k+Top-p Masking and Distillation Fine-Tuning](https://mlsanity.com/paper/2602.13515/) (2026)
- [PAROAttention: Pattern-Aware ReOrdering for Efficient Sparse and Quantized Attention in Visual Generation Models](https://mlsanity.com/paper/2506.16054/) (2025)
- [PISA: Piecewise Sparse Attention Is Wiser for Efficient Diffusion Transformers](https://mlsanity.com/paper/2602.01077/) (2026)
- [A Survey on Cache Methods in Diffusion Models: Toward Efficient Multi-Modal Generation](https://mlsanity.com/paper/2510.19755/) (2025)
- [Attention Surgery: An Efficient Recipe to Linearize Your Video Diffusion Transformer](https://mlsanity.com/paper/2509.24899/) (2025)
- [BLADE: Block-Sparse Attention Meets Step Distillation for Efficient Video Generation](https://mlsanity.com/paper/2508.10774/) (2025)
- [Light Forcing: Accelerating Autoregressive Video Diffusion via Sparse Attention](https://mlsanity.com/paper/2602.04789/) (2026)
- [Geometry-Aware Rotary Position Embedding for Consistent Video World Model](https://mlsanity.com/paper/2602.07854/) (2026)
- [SkyReels-V4: Multi-modal Video-Audio Generation, Inpainting and Editing model](https://mlsanity.com/paper/2602.21818/) (2026)
- [Mode Seeking meets Mean Seeking for Fast Long Video Generation](https://mlsanity.com/paper/2602.24289/) (2026)
- [VideoNSA: Native Sparse Attention Scales Video Understanding](https://mlsanity.com/paper/2510.02295/) (2025)
- [LinVideo: A Post-Training Framework towards O(n) Attention in Efficient Video Generation](https://mlsanity.com/paper/2510.08318/) (2025)
- [Sparse Forcing: Native Trainable Sparse Attention for Real-time Autoregressive Diffusion Video Generation](https://mlsanity.com/paper/2604.21221/) (2026)
- [Speed3R: Sparse Feed-forward 3D Reconstruction Models](https://mlsanity.com/paper/2603.08055/) (2026)

## Abstract (arXiv)

Scaling video diffusion transformers (DiTs) is limited by their quadratic 3D attention, even though most of the attention mass concentrates on a small subset of positions. We turn this observation into VSA, a trainable, hardware-efficient sparse attention that replaces full attention at both training and inference. In VSA, a lightweight coarse stage pools tokens into tiles and identifies high-weight critical tokens; a fine stage computes token-level attention only inside those tiles subjecting to block computing layout to ensure hard efficiency. This leads to a single differentiable kernel that trains end-to-end, requires no post-hoc profiling, and sustains 85% of FlashAttention3 MFU. We perform a large sweep of ablation studies and scaling-law experiments by pretraining DiTs from 60M to 1.4B parameters. VSA reaches a Pareto point that cuts training FLOPS by 2.53× with no drop in diffusion loss. Retrofitting the open-source Wan-2.1 model speeds up attention time by 6× and lowers end-to-end generation time from 31s to 18s with comparable quality. These results establish trainable sparse attention as a practical alternative to full attention and a key enabler for further scaling of video diffusion models. Code will be available at https://github.com/hao-ai-lab/FastVideo.