CAPE: Context-Aware Prompt Perturbation Mechanism with Differential Privacy

no code implementations • 9 May 2025 • Haoqi Wu, Wei Dai, Li Wang, Qiang Yan

Large Language Models (LLMs) have gained significant popularity due to their remarkable capabilities in text understanding and generation.

Nimbus: Secure and Efficient Two-Party Inference for Transformers

1 code implementation • 24 Nov 2024 • Zhengyi Li, Kang Yang, Jin Tan, Wen-jie Lu, Haoqi Wu, Xiao Wang, Yu Yu, Derun Zhao, Yancheng Zheng, Minyi Guo, Jingwen Leng

For the linear layer, we propose a new 2PC paradigm along with an encoding approach to securely compute matrix multiplications based on an outer-product insight, which achieves $2. 9\times \sim 12. 5\times$ performance improvements compared to the state-of-the-art (SOTA) protocol.

Ditto: Quantization-aware Secure Inference of Transformers upon MPC

1 code implementation • 9 May 2024 • Haoqi Wu, Wenjing Fang, Yancheng Zheng, Junming Ma, Jin Tan, Yinggui Wang, Lei Wang

Then, we propose novel MPC primitives to support the type conversions that are essential in quantization and implement the quantization-aware MPC execution of secure quantized inference.

Quantization

SoK: Training Machine Learning Models over Multiple Sources with Privacy Preservation

no code implementations • 6 Dec 2020 • Lushan Song, Guopeng Lin, Jiaxuan Wang, Haoqi Wu, Wenqiang Ruan, Weili Han

At first, we define the problem of Training machine learning Models over Multiple data sources with Privacy Preservation (TMMPP for short).

BIG-bench Machine Learning Federated Learning +1