PrivAnalogy: An Analogy Mechanism-Based Privacy Protection Framework for LLM Prompts
- 1 September 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 23 (5) , 11796-11807
Abstract
No abstract availableFunding Information
- Tianjin Science and Technology Major Special Project (25ZXSFSN00070)
- National Key Research and Development Program of China (2023YFB2703900)
- National Natural Science Foundation of China (62202244)
This publication has 31 references indexed in Scilit:
- Risk-Aware Privacy Preservation for LLM InferenceIEEE Transactions on Information Forensics and Security, 2026
- Casper: Prompt Sanitization for Protecting User Privacy in Web-Based Large Language Models2025
- CateSift: An interactive steering approach for classifying large scale textVisual Informatics, 2025
- InferDPT: Privacy-preserving Inference for Black-box Large Language ModelsIEEE Transactions on Dependable and Secure Computing, 2025
- Large Language Models (LLM) in Industry: A Survey of Applications, Challenges, and Trends2024
- Natural Language Understanding with Privacy-Preserving BERT2021
- Cheetah: Optimizing and Accelerating Homomorphic Encryption for Private Inference2021
- What Can We Learn Privately?SIAM Journal on Computing, 2011
- Mechanism Design via Differential Privacy2007
- Calibrating Noise to Sensitivity in Private Data Analysis2006