arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2305.04701cs.LGcs.CR
收藏

差分隐私注意力计算

Differentially Private Attention Computation

The University of Washington(华盛顿大学);The Simons Institute for the Theory of Computing at the University of California, Berkeley(加州大学伯克利分校西蒙斯理论计算研究所);University of Pennsylvania(宾夕法尼亚大学)

Yeqi Gao, Zhao Song, Xin Yang, Yufa Zhou

AI总结:

针对大型语言模型推理中敏感信息泄露的隐私问题,本文提出一种新颖高效的注意力矩阵近似计算算法,基于快速注意力计算和差分隐私矩阵发布技术,在保证计算效率的同时提供差分隐私保证。

摘要:

大型语言模型(LLMs),尤其是基于Transformer架构的模型,对日常生活的诸多方面产生了深远影响,如自然语言处理、内容生成和研究方法等。然而,关于大型语言模型推理结果的一个关键问题是安全与隐私。鉴于大型语言模型在许多场景下生成的结果可能泄露敏感机密或版权信息,因此计算具有可证明隐私保证的注意力矩阵至关重要,因为“注意力就是你所需要的一切”。在这项工作中,我们提出了一种新颖且高效的算法来近似计算注意力矩阵,同时提供差分隐私(DP)保证。为此,我们基于快速注意力计算和差分隐私矩阵发布的最新进展。

英文摘要:

Large language models (LLMs), especially those based on the Transformer architecture, have had a profound impact on various aspects of daily life, such as natural language processing, content generation, research methodologies, and more. Nevertheless, a crucial concern regarding the inference results of large language models is the issue of security and privacy. Given that large language models can generate results that may leak sensitive confidential or copyright information in many scenarios, it is crucial to compute the attention matrix with provable privacy guarantees, as attention is all you need. In this work, we propose a novel and efficient algorithm for approximating the attention matrix while providing differential privacy (DP) guarantees. To achieve this, we build on recent advancements in fast attention computation and differentially private matrix publishing.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑