利用记忆特性反制隐私泄露，让大模型更安全
Private Memorization Editing: Turning Memorization into a Defense to Strengthen Data Privacy in Large Language Models
- 将模型记忆能力转为防御手段，主动清除敏感信息
- 在不破坏语言能力的前提下，显著降低隐私数据泄露
- 适合关注数据隐私的AI系统开发者和合规团队
大型语言模型会记忆训练数据，可能包含不应存储的个人身份信息（PII），存在泄露风险。本文提出私有记忆编辑（PME）方法，将记忆这一潜在弱点转化为隐私防护策略。通过检测并编辑模型对训练数据的记忆，有效缓解PII记忆问题。实验表明，该方法在多种配置下显著减少隐私数据泄露，部分情况下使隐私攻击准确率降至零，同时保持模型原有语言能力不受影响。
原文摘要 · Abstract (English)
Large Language Models (LLMs) memorize, and thus, among huge amounts of uncontrolled data, may memorize Personally Identifiable Information (PII), which should not be stored and, consequently, not leaked. In this paper, we introduce Private Memorization Editing (PME), an approach for preventing private data leakage that turns an apparent limitation, that is, the LLMs' memorization ability, into a powerful privacy defense strategy. While attacks against LLMs have been performed exploiting previous knowledge regarding their training data, our approach aims to exploit the same kind of knowledge in order to make a model more robust. We detect a memorized PII and then mitigate the memorization of PII by editing a model knowledge of its training data. We verify that our procedure does not affect the underlying language model while making it more robust against privacy Training Data Extraction attacks. We demonstrate that PME can effectively reduce the number of leaked PII in a number of configurations, in some cases even reducing the accuracy of the privacy attacks to zero.
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。