arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2006.09615cs.CRcs.CLcs.LG
收藏

指纹语言生成的文本去歧义化

De-Anonymizing Text by Fingerprinting Language Generation

Cornell University(康奈尔大学);Cornell Tech(康奈尔理工学院);Tel Aviv University(特拉维夫大学)

Zhen Sun, Roei Schuster, Vitaly Shmatikov

摘要:

机器学习系统的组件(尚未)被视为安全热点。安全编码实践,例如确保没有执行路径依赖于机密输入,尚未被ML开发人员采用。我们开始研究ML系统的代码安全性,通过调查核采样-一种用于生成文本的流行方法,用于自动完成等应用程序-无意中泄漏用户键入的文本。我们的主要结果是,许多自然英语单词序列的核大小的系列是一个独特的指纹。然后,我们展示了攻击者如何通过合适的侧通道(例如,缓存访问时间),解释这种攻击如何帮助匿名文本去匿名化,并讨论防御措施。

英文摘要:

Components of machine learning systems are not (yet) perceived as security hotspots. Secure coding practices, such as ensuring that no execution paths depend on confidential inputs, have not yet been adopted by ML developers. We initiate the study of code security of ML systems by investigating how nucleus sampling---a popular approach for generating text, used for applications such as auto-completion---unwittingly leaks texts typed by users. Our main result is that the series of nucleus sizes for many natural English word sequences is a unique fingerprint. We then show how an attacker can infer typed text by measuring these fingerprints via a suitable side channel (e.g., cache access times), explain how this attack could help de-anonymize anonymous texts, and discuss defenses.

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。

↑