KernelBand: Steering LLM-based Kernel Optimization via Hardware-Aware Multi-Armed Bandits
Dezhi Ran, Shuxiao Xie, Mingfang Ji, Anmin Liu, Mengzhou Wu, Yuan Cao, Yuzhe Guo, Hao Yu, Linyi Li, Yitao Hu, Wei Yang, Tao Xie
š arxiv.org/abs/2511.18868
High-performance GPU kernels are critical for efficient LLM serving, yet their optimization remains a bottleneck requiring deep system expertise. While code LLMs show promise in generating functionall...
