Competition Domains Matter: Selective Head Globalization for Training-Free Sparse Attention in Video Diffusion Transformers
cs.CV
Submitted: 2026-10-10
Updated: 2026-10-10
Terminology
Sources
- SeerAttention: Learning Intrinsic Sparse Attention in Your LLMs
- HunyuanVideo: A Systematic Framework For Large Video Generative Models
- Compact Attention: Exploiting Structured Spatio-Temporal Sparsity for Fast Video Generation
- Radial Attention: $O(n\log n)$ Sparse Attention with Energy Decay for Long Video Generation
- DraftAttention: Fast Video Diffusion via Low-Resolution Attention Guidance
- LiteAttention: A Temporal Sparse Attention for Diffusion Transformers
- Wan: Open and Advanced Large-Scale Video Generative Models
- USV: Unified Sparsification for Accelerating Video Diffusion Models
- DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads
- Training-Free Efficient Video Generation via Dynamic Token Carving
- PAROAttention: Pattern-Aware ReOrdering for Efficient Sparse and Quantized Attention in Visual Generation Models
- HEART: Exploiting Head Heterogeneity in Sparse Attention for Video Diffusion
- Full Attention Strikes Back: Transferring Full Attention into Sparse within Hundred Training Steps
- Tactic: Adaptive Sparse Attention with Clustering and Distribution Fitting for Long-Context LLMs
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models