Skip to main navigation Skip to search Skip to main content

PPQformer: Privacy-Preserving Quantized Transformer for Efficient and Secure Inference

  • Benchang Dong
  • , Zhili Chen*
  • *Corresponding author for this work
  • East China Normal University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In the context of the popularity of Transformer model services, the issue of privacy protection has gradually gained attention. However, the primary issues with existing private inference solutions lie in memory constraints and suboptimal computational performance. In this work, we introduce PPQformer, a framework that leverages Replicated Secret Sharing (RSS) and quantization techniques to enable efficient and secure inference of Transformer models in a Multi-Party Computation (MPC) setting. By integrating quantization into the MPC protocols, PPQformer significantly reduces memory footprint and enhances computational performance while preserving data and model privacy. Experimental results demonstrate the effectiveness of our approach, achieving competitive accuracies with reduced communication cost and inference time compared to existing methods.

Original languageEnglish
Title of host publicationProceedings of the 2025 2nd International Conference on Generative Artificial Intelligence and Information Security, GAIIS 2025
PublisherAssociation for Computing Machinery, Inc
Pages115-119
Number of pages5
ISBN (Electronic)9798400713453
DOIs
StatePublished - 3 Jun 2025
Event2025 2nd International Conference on Generative Artificial Intelligence and Information Security, GAIIS 2025 - Hangzhou, China
Duration: 21 Feb 202523 Feb 2025

Publication series

NameProceedings of the 2025 2nd International Conference on Generative Artificial Intelligence and Information Security, GAIIS 2025

Conference

Conference2025 2nd International Conference on Generative Artificial Intelligence and Information Security, GAIIS 2025
Country/TerritoryChina
CityHangzhou
Period21/02/2523/02/25

Keywords

  • Multi-Party Computation
  • Privacy-preserving model inference
  • Replicated Secret Sharing
  • Transformer models quantization

Fingerprint

Dive into the research topics of 'PPQformer: Privacy-Preserving Quantized Transformer for Efficient and Secure Inference'. Together they form a unique fingerprint.

Cite this