arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.07004cs.CRcs.AI
收藏

Casper:用于保护Web基于大语言模型用户隐私的提示净化

Casper: Prompt Sanitization for Protecting User Privacy in Web-Based Large Language Models

New Jersey Institute of Technology(新泽西理工学院)

Chun Jie Chong, Chenxi Hou, Zhihao Yao, Seyed Mohammadjavad Seyed Talebi

AI总结:

Casper通过浏览器扩展在用户设备上运行,利用规则过滤、机器学习和本地LLM技术,有效过滤用户输入中的敏感信息,保护隐私。

摘要:

基于大语言模型(LLM)的服务已被广泛采用,并已成为我们互联网体验的重要组成部分。第三方插件通过启用对现实世界数据和服务的访问来增强LLM的功能。然而,这些服务及其第三方插件的隐私影响尚未得到充分理解。敏感的提示数据由基于云的LLM提供商和第三方插件存储、处理和共享。在本文中,我们提出了Casper,一种提示净化技术,旨在通过在发送到LLM服务之前检测并删除用户输入中的敏感信息来保护用户隐私。Casper完全在用户的设备上运行,作为浏览器扩展,并不需要对在线LLM服务进行任何更改。Casper的核心是一个三层净化机制,包括基于规则的过滤器、基于机器学习(ML)的命名实体识别器和基于浏览器的本地LLM主题识别器。我们对4000个合成提示的数据集进行了评估,证明Casper可以有效地过滤出个人可识别信息(PII)和隐私敏感主题,准确率分别为98.5%和89.9%。

英文摘要:

Web-based Large Language Model (LLM) services have been widely adopted and have become an integral part of our Internet experience. Third-party plugins enhance the functionalities of LLM by enabling access to real-world data and services. However, the privacy consequences associated with these services and their third-party plugins are not well understood. Sensitive prompt data are stored, processed, and shared by cloud-based LLM providers and third-party plugins. In this paper, we propose Casper, a prompt sanitization technique that aims to protect user privacy by detecting and removing sensitive information from user inputs before sending them to LLM services. Casper runs entirely on the user's device as a browser extension and does not require any changes to the online LLM services. At the core of Casper is a three-layered sanitization mechanism consisting of a rule-based filter, a Machine Learning (ML)-based named entity recognizer, and a browser-based local LLM topic identifier. We evaluate Casper on a dataset of 4000 synthesized prompts and show that it can effectively filter out Personal Identifiable Information (PII) and privacy-sensitive topics with high accuracy, at 98.5% and 89.9%, respectively.

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。

↑