Not logged in
Share
SaveFrom Threads to Tiles: T2T, a Compiler for CUDA-to-NPU Translation via 2D Vectorization
Li, Shuaijiang; Zhao, Jiacheng; Liu, Ying; Zhang, Shuoming; Chen, Lei; Li, Yijin; Zhang, Yangyu; Li, Zhicheng; Zhou, Runyu; Shi, Xiyu; Xia, Chunwei; Wen, Yuan; Feng, Xiaobing; Cui, Huimin
Share
SaveParDiff: Efficiently Parallelizing Reverse-Mode Automatic Differentiation with Direct Indexing
Huang, Shuhong; Tang, Shizhi; Wen, Yuan; Cao, Huanqi; Tang, Ruibai; Chen, Yidong; Yu, Jiping; Li, Yang; Jiang, Chao; Xiao, Limin; Zhai, Jidong
Share
Save
Share
Save
Share
Save
Share
Save
Share
Save
Share
Save