← Search

Zhi Pang

2 accepted papers

2026

PCFormer: Accelerating Privacy-preserving Transformer Inference by Partition and Combination

AAAI 2026technical

In recent years, transformer-based models have achieved remarkable success in sensitive domains, including healthcare, finance and personalized services, but their deployment raises significant privacy concerns. Existing secure inference studies have introduced cryptographic techniques such as Homom

Cited by 0SourcePDFScholar
2024

Chronic Poisoning: Backdoor Attack against Split Learning

AAAI 2024technical

Split learning is a computing resource-friendly distributed learning framework that protects client training data by splitting the model between the client and server. Previous work has proved that split learning faces a severe risk of privacy leakage, as a malicious server can recover the client's…