SpargeAttention2: Trainable Sparse Attention via Hybrid Top-k+Top-p Masking and Distillation Fine-Tuning Paper • 2602.13515 • Published Feb 13 • 44
SLA2: Sparse-Linear Attention with Learnable Routing and QAT Paper • 2602.12675 • Published Feb 13 • 58
TurboDiffusion: Accelerating Video Diffusion Models by 100-200 Times Paper • 2512.16093 • Published Dec 18, 2025 • 97
SageAttention2++: A More Efficient Implementation of SageAttention2 Paper • 2505.21136 • Published May 27, 2025 • 45
SageAttention3: Microscaling FP4 Attention for Inference and An Exploration of 8-Bit Training Paper • 2505.11594 • Published May 16, 2025 • 75
SpargeAttn: Accurate Sparse Attention Accelerating Any Model Inference Paper • 2502.18137 • Published Feb 25, 2025 • 60
SageAttention: Accurate 8-Bit Attention for Plug-and-play Inference Acceleration Paper • 2410.02367 • Published Oct 3, 2024 • 50
