Yuntao Du (杜云滔)

I am a PhD candidate in computer science at Purdue University, advised by Prof. Ninghui Li. Here is my CV (updated May 20, 2026).

I study security and privacy risks in machine learning and LLM agents, from both adversarial and defensive perspectives. My current research focuses on:

• AI Security and Privacy. (1) Uncovering emerging privacy threats posed by LLM misuse (ACL’26, Data Eng. Bulletin’25); (2) Building practical data-use detection methods to support the responsible use of training data in LLMs.
• Data Privacy in Machine Learning. (1) Designing principled membership inference attacks to assess information leakage in ML models (NDSS’26, USENIX Security’26); (2) Auditing privacy risks of LLMs (USENIX Security’26 [1], [2]; ICLR’26);
• Differentially Private Data Synthesis. Developing practical data synthesis algorithms with provable privacy guarantees for various types of sensitive data (CCS’25, VLDB’23).

My research has been recognized and supported by the Ross Fellowship (2023-2027), Presidential Doctoral Excellence Award (2023-2027), and Herbold Scholarship (2023-2024).