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 2407.10058cs.CLcs.AI
收藏

学会拒绝:旨在缓解大语言模型中的隐私风险

Learning to Refuse: Towards Mitigating Privacy Risks in LLMs

Soochow University(苏州大学)

Zhenhua Liu, Tong Zhu, Chuanyuan Tan, Wenliang Chen

AI总结:

针对大语言模型意外记忆隐私信息的风险,该研究构建了真实个人数据遗忘数据集RETURN,并提出名称感知遗忘框架NAUF,在保护目标个人隐私的同时维持模型通用能力,遗忘表现优于现有基线。

摘要:

大语言模型(LLM)在理解和生成自然语言方面展现出卓越能力,但这类模型可能会意外记忆私人信息,带来重大隐私风险。本研究致力于解决无需完全重新训练即可让LLM保护特定个人隐私数据的难题。我们提出了RETURN(真实世界个人数据遗忘数据集),该数据集包含来自维基百科的2492名个人及对应的问答对,用于在真实场景下评估用于保护个人数据的机器遗忘(MU)方法。此外,我们还提出了面向隐私保护的名称感知遗忘框架(NAUF),该框架能让模型学会哪些个人的信息需要被保护,同时不影响其回答其他无关个人相关问题的能力。大量实验表明,NAUF取得了当前最优的平均遗忘得分,比最佳基线方法高出5.65分,在有效保护目标个人隐私数据的同时,还能维持模型的通用能力。

英文摘要:

Large language models (LLMs) exhibit remarkable capabilities in understanding and generating natural language. However, these models can inadvertently memorize private information, posing significant privacy risks. This study addresses the challenge of enabling LLMs to protect specific individuals' private data without the need for complete retraining. We propose \return, a Real-world pErsonal daTa UnleaRNing dataset, comprising 2,492 individuals from Wikipedia with associated QA pairs, to evaluate machine unlearning (MU) methods for protecting personal data in a realistic scenario. Additionally, we introduce the Name-Aware Unlearning Framework (NAUF) for Privacy Protection, which enables the model to learn which individuals' information should be protected without affecting its ability to answer questions related to other unrelated individuals. Our extensive experiments demonstrate that NAUF achieves a state-of-the-art average unlearning score, surpassing the best baseline method by 5.65 points, effectively protecting target individuals' personal data while maintaining the model's general capabilities.

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。

↑