mit-han-lab/spatten-llm
[HPCA'21] SpAtten: Efficient Sparse Attention Architecture with Cascade Token and Head Pruning
ScalaMIT
Stargazers
- adamgallasInstitute of Automation, Chinese Academy of Sciences
- buaabaiBUAA
- chenshih1
- chhzh123Cornell University
- chuyi369
- dhjoo98OneandZero
- dongdong1203
- Dream-my-heart
- dreamflyings
- fangtaosongUniversity of Chinese Academy of Sciences, NLP LAB
- fly51flyPRIS
- genni613
- Hanrui-WangMIT
- IamXuLiang
- JJJayyyyDurham, NC
- kentang-mitCambridge, Massachusetts, United States
- learning-chip
- leliyliu
- ltz880
- lu-m13Intel Labs China
- MARD1NOSiliconFlow
- mingo99
- nevilshah235
- nuaaceieyty
- raneryGeorgia Institute of Technology
- royessTsinghua University
- SakitsShanghai Jiao Tong University
- Stronger-Huang
- umiswingNEU
- wangyuyueLos Angeles
- wwbitejotunnUESTC
- xie-1399BUAA
- xiurui-panTsinghua University
- YangWang92
- zhengyue08Minnesota, USA
- zhouyecsPeking University