PrivacyMind: Large Language Models Can Be Contextual Privacy Protection Learners
Nov 12, 2024·
,,,,,,,,,,·
1 min read
Yijia Xiao
Yiqiao Jin
Yushi Bai
Yue Wu
Xianjun Yang
Xiao Luo
Wenchao Yu
Xujiang Zhao
Yanchi Liu
Haifeng Chen
Wei Wang
Wei Cheng
Abstract
Deploying LLMs on private text data raises serious contextual privacy concerns. We introduce PrivacyMind, which teaches LLMs to be contextual privacy protection learners — recognizing sensitive content in context and adapting their outputs accordingly. PrivacyMind combines targeted training signals with contextual prompts to preserve utility while reducing leakage of sensitive content. EMNLP 2024 acceptance rate: 20.8%.
Type
Publication
Conference on Empirical Methods in Natural Language Processing (EMNLP) 2024, Main Track
Abstract
Deploying LLMs on private text data raises serious contextual privacy concerns. We introduce PrivacyMind, which teaches LLMs to be contextual privacy protection learners — recognizing sensitive content in context and adapting their outputs accordingly.
