Teach Large Language Models to Forget Privacy

no code implementations • 30 Dec 2023 • Ran Yan, YuJun Li, Wenqian Li, Peihua Mai, Yan Pang, Yinchuan Li

Large Language Models (LLMs) have proven powerful, but the risk of privacy leakage remains a significant concern.

Privacy Preserving Zero-shot Generalization

Split-and-Denoise: Protect large language model inference with local differential privacy

no code implementations • 13 Oct 2023 • Peihua Mai, Ran Yan, Zhe Huang, Youjia Yang, Yan Pang

Large Language Models (LLMs) shows powerful capability in natural language understanding by capturing hidden semantics in vector space.

Language Modelling Large Language Model +2

PrivMVMF: Privacy-Preserving Multi-View Matrix Factorization for Recommender Systems

no code implementations • 29 Sep 2022 • Peihua Mai, Yan Pang

Then, the paper proposes a new privacy-preserving framework based on homomorphic encryption, Privacy-Preserving Multi-View Matrix Factorization (PrivMVMF), to enhance user data privacy protection in federated recommender systems.

Federated Learning Privacy Preserving +2