Research agenda
Trustworthy AI systems, from recommender models to LLMs.
My work studies how machine learning systems expose, retain, and misuse training data, and how we can audit these risks in realistic model and application settings.
Research Interests
Data privacy and auditing
User data revocation, membership inference, ownership verification, and training data detection.
LLM privacy and exposure
Methods for identifying pre-training data and measuring privacy leakage in language and vision-language models.
Robust learning systems
Stress testing graph and recommender systems against adversarial and model stealing attacks.
Published Journal Article
Google ScholarForget Me If You Can: Auditing User Data Revocation in Recommendation Systems
Zhihao Zhu, Yi Yang, Yangyang Fan, Defu Lian
Information Systems Research, 2026
Network Representation Lightening from Hashing to Quantization
Defu Lian, Zhihao Zhu, Kai Zheng, Yong Ge, Xing Xie, Enhong Chen
IEEE Transactions on Knowledge and Data Engineering, 2022
Working Papers
HoneyImage: Verifiable, Harmless, and Stealthy Dataset Ownership Verification for Image Models
Zhihao Zhu, Jiale Han, Yi Yang
Management Information Systems Quarterly — Major Revision
Revealing Training Data Exposure in Vision-Language Large Models via Parameter Gradients
Zhihao Zhu, Hongyi Tang, Yi Yang, Ahmed Abbasi
Nature Communications — Major Revision
RecShield: Output-Level Attribute Unlearning in Recommender Systems
Zhihao Zhu, Yi Yang
Information Systems Research — Under Review
GraphMSA: Stress Testing Graph Classification Services Against Model Stealing Attacks
Zhihao Zhu, Yi Yang, Chenwang Wu, Defu Lian
INFORMS Journal on Computing — Under Review
Model Stealing Attacks against Recommender Systems
Zhihao Zhu, Rui Fan, Chenwang Wu, Yi Yang, Defu Lian, Enhong Chen
IEEE Transactions on Dependable and Secure Computing — Under Review
Understanding Privacy Risks of Embeddings Induced by Large Language Models
Zhihao Zhu, Ninglu Shao, Defu Lian, Chenwang Wu, Zheng Liu, Yi Yang, Enhong Chen
AI Conferences
TDDBench: A Benchmark for Training Data Detection
Zhihao Zhu, Yi Yang, Defu Lian
International Conference on Learning Representations, 2025
Identifying Pre-training Data in LLMs: A Neuron Activation-Based Detection Framework
Hongyi Tang*, Zhihao Zhu*, Yi Yang. Equal contribution.
Conference on Empirical Methods in Natural Language Processing, 2025
Membership Inference Attacks against Sequential Recommender Systems
Zhihao Zhu, Chenwang Wu, Rui Fan, Defu Lian, Enhong Chen
The Web Conference, 2023
Resisting Graph Adversarial Attack via Cooperative Homophilous Augmentation
Zhihao Zhu, Chenwang Wu, Min Zhou, Hao Liao, Defu Lian, Enhong Chen
European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2022
Awards
- National Graduate Scholarship (China), 2022
- KDD Cup 2023 "Next Product Generation" Challenge, 2nd Place
Service
Reviewer for Information Systems Research (ISR), INFORMS Journal on Computing (IJOC), NeurIPS, ICLR, and ACL Rolling Review.