Hongyao Yu is a Master student in Computer Science at Tsinghua University, under the mentorship of Prof. Shu-Tao Xia. He works closely with Prof. Bin Chen. Before that, He completed undergraduate studies in Computer Science and Technology at Harbin Institute of Technology, Shenzhen. He has published articles on training data detection and trustworthy AI. His research interests generally include trustworthy AI, machine learning and data privacy. He has also served as a reviewer for top-tier conferences, such as ICLR and ACM CSUR.

My research interests focus on:

• Diffusion Large Language Models
• Multimodal Large Language Models
• Training Data Detection
• Trustworthy AI

📧 Concat

email: chrisqcwx@gmail.com

📝 Publications

Notes:（*）indicates the equal contributions.

SIGKDD, 2026

Rank Matters: Understanding and Defending Model Inversion Attacks via Low-Rank Feature Filtering

Hongyao Yu, Yixiang Qiu, Hao Fang, Tianqu Zhuang, Bin Chen, Sijin Yu, Bin Wang, Shu-Tao Xia and Ke Xu

Paper Code

SIGKDD, 2026

GaussTrap: Stealthy Poisoning Attacks on 3D Gaussian Splatting for Targeted Scene Confusion

Jiaxin Hong*, Sixu Chen*, Shuoyang Sun*, Hongyao Yu*, Hao Fang, Yuqi Tan, Bin Chen, Shuhan Qi and Shu-Tao Xia

Paper

ACL, 2026

Retrievals Can Be Detrimental: Unveiling the Backdoor Vulnerability of Retrieval-Augmented Diffusion Models

Hao Fang*, Xiaohang Sui*, Hongyao Yu*, Kuofeng Gao, Jiawei Kong, Sijin Yu, Bin Chen and Shu-Tao Xia

Paper Code

ACL, 2026

Chenxi Qing, Jinxi Wu, Zheng Liu, Yixiang Qiu, Hongyao Yu, Bin Chen, Hao Wu and Shu-Tao Xia

Paper

ACM MM, 2025

ICAS: Detecting Training Data from Autoregressive Image Generative Models

Hongyao Yu, Yixiang Qiu, Yiheng Yang, Hao Fang, Tianqu Zhuang, Jiaxin Hong, Bin Chen, Hao Wu, and Shu-Tao Xia

Paper Code

ICLR, 2025

Stealthy Shield Defense: A Conditional Mutual Information-Based Approach against Black-Box Model Inversion Attacks

Tianqu Zhuang*, Hongyao Yu*, Yixiang Qiu*, Hao Fang*, Bin Chen, Shu-Tao Xia

Paper Code

ECCV, 2024 (oral)

A Closer Look at GAN Priors: Exploiting Intermediate Features for Enhanced Model Inversion Attacks

Yixiang Qiu*, Hao Fang*, Hongyao Yu*, Bin Chen, MeiKang Qiu, Shu-Tao Xia

Paper Code

📖 Educations

• 2025.09 - 2028.07 (expected), MSc in Computer Science, Tsinghua University
• 2021.09 - 2025.06, Bachelor in Computer Science and Technology, Harbin Institute of Technology, Shenzhen

🏆 Awards

• Chinese National Scholarship for Undergraduate Students × 2 (2022, 2024)
• Top Ten Academic Excellence Stars of Harbin Institute of Technology (2024)
• First class scholarship × 3 (2022, 2023, 2024)
• National Second Prize in the China Undergraduate Mathematical Contest in Modeling (2023)
• National Third Prize in the National Student Computer System Capability Challenge (NSCSCC) (2023)
• Provincial First Prize in the China Undergraduate Mathematical Contest in Modeling (2022)

📝 Services

• Reviewer for ICLR 2025, 2026 and ACM CSUR.