Yuntian Chen, Northwest University; Xi'an Key Laboratory of Advanced Computing and Software Security; Shaanxi International Joint Research Center for the Battery-Free Internet of Things; Tianpei Lu, The State Key Laboratory of Blockchain and Data Security, Zhejiang University; Hangzhou High-Tech Zone (Binjiang) Institute of Blockchain and Data Security; Zhanyong Tang, Northwest University; Xi'an Key Laboratory of Advanced Computing and Software Security; Shaanxi Key Laboratory of Passive Internet of Things and Neural Computing; Bingsheng Zhang, The State Key Laboratory of Blockchain and Data Security, Zhejiang University; Hangzhou High-Tech Zone (Binjiang) Institute of Blockchain and Data Security; Zhiying Shi, Northwest University; Xi'an Key Laboratory of Advanced Computing and Software Security; Shaanxi Key Laboratory of Passive Internet of Things and Neural Computing; Yuxiang Luan, Northwest University; Xi'an Key Laboratory of Advanced Computing and Software Security; Zhuzhu Wang, Northwest University; Xi'an Key Laboratory of Advanced Computing and Software Security; Shaanxi International Joint Research Center for the Battery-Free Internet of Things
The growing demand for privacy-preserving Transformer inference has led to the emergence of numerous protocols designed to protect sensitive data and model parameters. These protocols utilize diverse cryptographic tools under varying assumptions, each presenting unique characteristics and trade-offs between computation, communication, and accuracy. In this paper, we conduct a systematic and in-depth analysis of existing approaches from diverse performance perspectives, identifying their limitations and research gaps. We further evaluate the reproducibility of prior systems and re-benchmark representative solutions under standardized configurations. Our results yield a principled guideline for balancing protocol trade-offs under different deployment settings.
USENIX Security '26 Open Access Sponsored by
King Abdullah University of Science and Technology (KAUST)
Open Access Media
USENIX is committed to Open Access to the research presented at our events. Papers and proceedings are freely available to everyone once the event begins. Any video, audio, and/or slides that are posted after the event are also free and open to everyone. Support USENIX and our commitment to Open Access.
author = {Yuntian Chen and Tianpei Lu and Zhanyong Tang and Bingsheng Zhang and Zhiying Shi and Yuxiang Luan and Zhuzhu Wang},
title = {Sok: Private Transformer-based Model Inference},
booktitle = {35th USENIX Security Symposium (USENIX Security 26)},
year = {2026},
isbn = {978-1-939133-58-8},
address = {Baltimore, MD},
pages = {1127--1146},
url = {https://www.usenix.org/conference/usenixsecurity26/presentation/chen-yuntian},
publisher = {USENIX Association},
month = aug
}