I am a Ph.D. student in Computer and Information Technology at Purdue University, advised by Prof. Wenhai Sun.
My research focuses on AI security and privacy, with current interests in robust watermarking and provenance for generative language, speech, and audio. I study how security and privacy mechanisms behave under adversarial manipulation, spanning differential privacy, data poisoning, backdoor attacks, and privacy-preserving machine learning. My current work develops robust watermarking for LLM-generated text, with a focus on improving detection after rewriting while preserving generation quality.
🔥 News
- Sep 2026, Our paper “Your Privacy My Cloak: Backdoor Attacks on Differentially Private Federated Learning” was accepted to IEEE S&P 2027! Grateful to my coauthors for their dedication and support!
- Aug 2026, I joined Xmotors AI in Santa Clara as a Machine Learning Engineer Intern, working on audio watermarking and provenance for generative speech and audio.
- Oct 2025, Honored to be selected by Purdue University as an RSAC Security Scholar. Grateful for the opportunity and excited to join RSAC 2026 in San Francisco!
- Apr 2025, Our paper “Mitigating Data Poisoning Attacks to Local Differential Privacy” was accepted to ACM CCS 2025! Thanks to all of my collaborators.
📝 Publications
Xiaolin Li, Ning Wang, Ninghui Li, Wenhai Sun. Your Privacy My Cloak: Backdoor Attacks on Differentially Private Federated Learning. IEEE Symposium on Security and Privacy (S&P), 2027. Accepted.
[PDF] [arXiv]
Studies how differential privacy can conceal malicious updates in federated learning, exposing the limits of existing defenses against backdoor attacks.
Xiaolin Li, Ninghui Li, Boyang Wang, Wenhai Sun. Mitigating Data Poisoning Attacks to Local Differential Privacy. ACM Conference on Computer and Communications Security (CCS), 2025.
[PDF]
Develops malicious-report detection and attack-resilient post-processing to mitigate poisoning and recover utility in private frequency estimation.
Xiaolin Li, Qikui Xu, Zhenyu Xu, Hongyan Zhang, Li Xu. Graph Reconfigurable Pooling for Graph Representation Learning. IEEE Transactions on Emerging Topics in Computing. 2023. [PDF]
Xiaolin Li, Li Xu, Hongyan Zhang, Qikui Xu. Differential privacy preservation for graph auto-encoders: A novel anonymous graph publishing model. Neurocomputing. 2023. [PDF]
-
Zhenyu Xu, Yifan Li,
Xiaolin Li, Xinxin Zhang, Li Xu. Influence Maximization in Partially Observable Mobile Social Networks. MobiMedia. 2023. [PDF] -
Hongyan Zhang,
Xiaolin Li, Jiayu Xu, Li Xu, Graph matching based privacy-preserving scheme in social networks. SocialSec. 2021.
[PDF]
🔬 Research Experience
Machine Learning Engineer Intern · Xmotors AI
Santa Clara, CA · Aug. 2026–Present
- Research audio watermarking and provenance for generative speech and audio, including detection that remains reliable after audio re-encoding.
- Develop evaluation pipelines on open speech and music models to assess robustness to compression, time shifts, and speed changes, alongside detection accuracy and audio quality.
Research Assistant · Purdue University
West Lafayette, IN · Sep. 2023–Present
- Robust LLM watermarking: Develop watermarking methods for AI-generated text, studying the trade-offs among robustness to rewriting, detection reliability, and generation quality.
- Study security and privacy in machine learning, including poisoning defenses for local differential privacy and backdoor attacks on differentially private federated learning.
Research Assistant · Fujian Normal University
Fuzhou, China · Sep. 2020–Jun. 2023
- Studied differential privacy and graph learning for privacy-preserving graph publication and representation learning.
🏅 Honors and Awards
- 2025.10, Selected as an RSAC 2026 Security Scholar, Purdue University.
- 2023.09, Presidential Doctoral Excellence Award, Purdue University.
🎓 Education
- 2023.09–Present, Ph.D. in Computer and Information Technology, Purdue University, West Lafayette, IN, USA.
- 2020.09–2023.06, M.S. in Cybersecurity, Fujian Normal University, Fuzhou, China.
- 2016.09–2020.06, B.E. in Network Engineering, Shandong Agricultural University, Taian, China.
💬 Professional Service
- Reviewer
- IEEE INFOCOM 2026.
- IEEE TIFS 2025; IEEE TDSC 2025/2026; IEEE TKDE 2026; ACM TKDD 2025/2026; Journal of the Chinese Institute of Engineers 2025; The Journal of Supercomputing 2024.