Zhu Yufan

Secure and Privacy-Preserving Machine Learning

Zhu Yufan

Computer Science PhD Student at the National University of Singapore

I develop secure systems for private large-model inference. My research sits at the intersection of machine learning systems and applied cryptography, with an emphasis on protecting both user data and model assets across untrusted infrastructure.

My current work spans homomorphic encryption, secure multiparty computation, LLM systems security, encrypted state-space models, and GPU acceleration.

Research Interests

Applied Cryptography · Privacy-Preserving Machine Learning · Data Privacy and Sanitization

Education

PhD in Computer Science

National University of Singapore

Research focus: LLM systems security and privacy-preserving machine learning. NUS Research Scholarship.

B.Comp. in Computer Science, Highest Distinction

National University of Singapore

Second Major in Statistics · GPA 4.6/5.0 · Focus areas: Artificial Intelligence and Databases.

Publications

2026

IEEE Transactions on Dependable and Secure Computing

EncFormer: Secure and Efficient Transformer Inference over Encrypted Data

Yufan Zhu, Chao Jin, Khin Mi Mi Aung, Xiaokui Xiao

Accepted for publication.

Publisher arXiv PDF
2024

IEEE Conference on Artificial Intelligence (CAI)

A Personalized Learning Tool for Physics Undergraduate Students Built on a Large Language Model for Symbolic Regression

Yufan Zhu, Zi-Yu Khoo, Jonathan Sze Choong Low, Stéphane Bressan

In Proceedings of the 2024 IEEE Conference on Artificial Intelligence.

Publisher PDF

Manuscripts

2026

NDSS 2027

FESC: Remodeling Long-Context Private Inference with Encrypted State-Space Models

Yufan Zhu, Chao Jin, Khin Mi Mi Aung, Xiaokui Xiao

Manuscript submitted to the Network and Distributed System Security Symposium.

arXiv

Professional Service

Reviewer

ACM International Conference on Information and Knowledge Management (CIKM 2026)

Reviewer

IEEE Transactions on Information Forensics and Security (TIFS)