用随机先验编码低秩梯度，实现大模型微调的端到端隐私保护
DR-Encoder: Encode Low-rank Gradients with Random Prior for Large Language Models Differentially Privately
- 通过两阶段随机性设计，用高斯先验训练梯度自编码器
- 在多个基础模型上保持精度，同时实现高隐私保障
- 适合关注联邦学习中隐私保护的大模型研究者
大型语言模型（LLM）在语言理解、翻译、逻辑推理乃至偏微分方程求解等领域展现出卓越性能，其核心架构为Transformer。如何在LLM微调过程中实现端到端隐私保障成为关键问题。本文分析了联邦微调过程中的三类潜在信息泄露，并提出基于双阶段随机性的端到端隐私保护方案。第一阶段利用本地客户端生成的梯度统计信息，训练带有高斯随机先验的梯度自编码器；第二阶段通过添加适当高斯噪声，对整体LLM进行差分隐私微调。实验表明，该方法在多个基础模型和两个主流评估基准上均实现了高效且高精度的性能表现。此外，本文还基于高斯差分隐私（GDP）与瑞尼差分隐私（RDP）进行了全面的隐私分析。
原文摘要 · Abstract (English)
The emergence of the Large Language Model (LLM) has shown their superiority in a wide range of disciplines, including language understanding and translation, relational logic reasoning, and even partial differential equations solving. The transformer is the pervasive backbone architecture for the foundation model construction. It is vital to research how to adjust the Transformer architecture to achieve an end-to-end privacy guarantee in LLM fine-tuning. In this paper, we investigate three potential information leakage during a federated fine-tuning procedure for LLM (FedLLM). Based on the potential information leakage, we provide an end-to-end privacy guarantee solution for FedLLM by inserting two-stage randomness. The first stage is to train a gradient auto-encoder with a Gaussian random prior based on the statistical information of the gradients generated by local clients. The second stage is to fine-tune the overall LLM with a differential privacy guarantee by adopting appropriate Gaussian noises. We show the efficiency and accuracy gains of our proposed method with several foundation models and two popular evaluation benchmarks. Furthermore, we present a comprehensive privacy analysis with Gaussian Differential Privacy (GDP) and Renyi Differential Privacy (RDP).
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。