@misc{osama2023stream,
  title={Stream-K: Work-centric Parallel Decomposition for Dense Matrix-Matrix Multiplication on the GPU},
  author={Muhammad Osama and Duane Merrill and Cris Cecka and Michael Garland and John D. Owens},
  year={2023},
  eprint={2301.03598},
  archivePrefix={arXiv},
  primaryClass={cs.DS},
  url={https://arxiv.org/abs/2301.03598}
}