H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models
Authors:
- Zhenyu ZhangZZZhenyu Zhang
- Ying ShengYSYing Sheng
- Tianyi ZhouTZTianyi Zhou
- Tianlong ChenTCTianlong Chen
- Lianmin ZhengLZLianmin Zheng
- Ruisi CaiRCRuisi Cai
- Zhao SongZSZhao Song
- Yuandong TianYTYuandong Tian
- Christopher RéCRChristopher Ré
- Clark BarrettCBClark Barrett
- Zhangyang Atlas WangZWZhangyang Atlas Wang
- Beidi ChenBCBeidi Chen
- 1 January 2023
- conference paper
- Published by Curran Associates, Inc.
- p. 34661-34710