Zhenhua Liu

Probing Language Models for Pre-training Data Detection

2024 · arXiv (Cornell University)

Large Language Models (LLMs) have shown their impressive capabilities, while also raising concerns about the data contamination problems due to privacy issues and leakage of benchmark datasets in the pre-training phase. Therefore, it is vital …

Efficient Threshold and Unbalanced Quorum PSI Protocols for Secure Data Sharing in Cloud-Assisted Internet of Things

2025 · IEEE Internet of Things Journal

Threshold private set intersection (t-PSI) protocol enables the secure determination of whether the intersection between two sets meets a specified threshold, facilitating privacy-preserving data collaboration in cloud computing for internet of things (IoT). However, existing …