Security and Safety2025，Vol.4Issue(4)：P.145-168,24.DOI:10.1051/sands/2025014
PUMA: Secure inference of LLaMA-7B in five minutes
Ye Dong 1Wen-Jie Lu 2Yancheng Zheng 2Haoqi Wu 2Derun Zhao 2Jin Tan 2Zhicong Huang 2Cheng Hong 2Tao Wei 2Wen-Guang Chen 2Jianying Zhou3
作者信息
- 1. Ant Group,Beijing 100081,China National University of Singapore,Singapore 119260,Singapore
- 2. Ant Group,Beijing 100081,China
- 3. Singapore University of Technology and Design,Singapore 487372,Singapore
- 折叠
摘要
关键词
Privacy/Security/Secure Three-Party Computation/Privacy-Preserving Machine Learning/Large Language Models分类
信息技术与安全科学引用本文复制引用
Ye Dong,Wen-Jie Lu,Yancheng Zheng,Haoqi Wu,Derun Zhao,Jin Tan,Zhicong Huang,Cheng Hong,Tao Wei,Wen-Guang Chen,Jianying Zhou..PUMA: Secure inference of LLaMA-7B in five minutes[J].Security and Safety,2025,4(4):P.145-168,24.
