Improving Data Reuse across Blocks for Efficient Block-Sparse Transformers on GPUs
Authors: L. Hu, X. Lu, X.-H. Sun, P. Jiang
Date: October, 2026
Venue: The 35th International Conference on Parallel Architectures and Compilation Techniques (PACT 2026)
Type: Conference
Tags
Block-Sparse TransformersGPU AccelerationData ReuseSparse Computing