Secure Transformer Inference Protocol

arXiv:2312.00025v2 Announce Type: replace-cross
Abstract: 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 …

abstract adoption arxiv chatgpt concerns cs.cr cs.lg data experience hands-on experience inference insights parameters protocol security security concerns services transformer transformer models type user data while