Flash-KMeans revisits exact K-Means from a systems perspective and implements it as a high-performance GPU primitive. The project is also useful beyond one paper and serves as a reusable building block in efficient ML systems.
Flash-KMeans
Shuo Yang*, Haocheng Xi*, Yilong Zhao, Muyang Li, Xiaoze Fan, Jintao Zhang, Han Cai, Yujun Lin, Xiuyu Li, Kurt Keutzer, Song Han, Chenfeng Xu, Ion Stoica
|
Mar 1, 2026
