About Me

Xiang Li (李想) is currently a third-year Ph.D. student at George Mason University, supervised by Prof. Qiang Zeng. He received his M.E. degree from Southeast University under the supervision of Prof. Kai Dong, and received his B.E. degree from Nanjing University of Information Science and Technology.

My primary research interest lies in the field of trustworthy artificial intelligence, with a particular focus on adversarial threats to computer vision systems.

If you are interested in discussing with me, feel free to drop me an email (xli62@gmu.edu).

Publications

• NeurIPS’26 Watermarking as a Learned Intrinsic Property of Diffusion Models
  Xiang Li, Dong-Dong Wu, Masashi Sugiyama, Lannan Luo, Qiang Zeng
  Annual Conference on Neural Information Processing Systems, 2026
  PDF Code

• ECCV’26 Making Partial-Label Datasets Easier: A Simple Yet Highly Effective Data Augmentation for Deep Partial-Label Learning
  Dong-Dong Wu, Zhaoyi Li, Xiang Li, Zhiqiang Shen
  European Conference on Computer Vision, 2026
  PDF Code

• ICCV’25 Backdoor Attacks on Neural Networks via One-Bit Flip
  Xiang Li, Lannan Luo, Qiang Zeng
  International Conference on Computer Vision, 2025
  PDF Code

• Security’25 Rowhammer-Based Trojan Injection: One Bit Flip Is Sufficient for Backdooring DNNs
  Xiang Li, Ying Meng, Junming Chen, Lannan Luo, Qiang Zeng
  USENIX Security Symposium, 2025
  PDF Code Website

Services

Program & Evaluation Committee

• [Security’26] Artifact Evaluation Committee, Usenix Security Symposium, 2026.

Conference Reviewer

• [NeurIPS’26] Annual Conference on Neural Information Processing Systems, 2026.

Journal Reviewer

• [TDSC] IEEE Transactions on Dependable and Secure Computing.
• [JSC] Journal of Computer Security.

Honors and Awards

• Student Travel Grant, USENIX Security’25.
• Broadening Participation Award, ICCV’25.
• Graduate Student Travel Fund, George Mason University.

Work Experience

• 2025.01 - Present, Graduate Research Assistant.
• 2024.08 - 2024.12, CS471, Operating Systems, Graduate Teaching Assistant.

Talks

• Rowhammer-Based Trojan Injection: One Bit Flip Is Sufficient for Backdooring DNNs. USENIX Security Symposium, Seattle, WA, USA.

Educations

• 2021.09 - 2024.06, Master, Southeast University, Nanjing.
• 2017.09 - 2021.06, Undergraduate, Nanjing University of Information Science and Technology, Nanjing.