Paper 2023/1763
Secure Transformer Inference
, University of Science and Technology of ChinaAbstract
Security of model parameters and user data is critical for Transformer-based services, such as ChatGPT. While recent strides in secure two-party protocols have successfully addressed security concerns in serving Transformer models, their adoption is practically infeasible due to the prohibitive cryptographic overheads involved. Drawing insights from our hands-on experience in developing two real-world Transformer-based services, we identify the inherent efficiency bottleneck in the two-party assumption. To overcome this limitation, we propose a novel three-party threat model. Within this framework, we design a semi-symmetric permutation-based protection scheme and present STIP, the first secure Transformer inference protocol without any inference accuracy loss. Experiments on representative Transformer models in real systems show that STIP has practical security and outperforms state-of-the-art secure two-party protocols in efficiency by millions of times.
Metadata
- Available format(s)
-
PDF
- Category
- Applications
- Publication info
- Preprint.
- Keywords
- large language modeltransformerinferencesecure protocol
- Contact author(s)
-
ym0813 @ mail ustc edu cn
zhanglan @ ustc edu cn
xiangyangli @ ustc edu cn - History
- 2024-05-08: revised
- 2023-11-15: received
- See all versions
- Short URL
- https://ia.cr/2023/1763
- License
-
CC BY
BibTeX
@misc{cryptoeprint:2023/1763,
author = {Mu Yuan and Lan Zhang and Xiang-Yang Li},
title = {Secure Transformer Inference},
howpublished = {Cryptology ePrint Archive, Paper 2023/1763},
year = {2023},
note = {\url{https://eprint.iacr.org/2023/1763}},
url = {https://eprint.iacr.org/2023/1763}
}
