NVIDIA & SakanaAILabs Collaborate on ICML2026 Paper: Sparse Transformer Kernels Optimized for NVIDIA GPUs
NVIDIA and SakanaAILabs co-author an ICML2026 paper introducing TwELL sparse packing and fused CUDA kernels, achieving >20% inference/training speedup at scale.
入选理由:TwELL稀疏打包格式可实现99%以上神经元稀疏度,对下游性能影响小于1%

