Hao Fang (方豪)

Hao Fang

2-year Master degree candidate

Address: SIGS, Tsinghua University, Shenzhen, Guangdong Province, China

Research Interest: Trustworthy AI, AIGC Security, Computer Vision

Lab: ITML, SIGS, Tsinghua University

E-mail: fang-h23@mails.tsinghua.edu.cn

Biography

I am a second-year master's student in the ITML lab in the SIGS, Tsinghua University, supervised by Prof.Bin-Chen and Prof.Shu-Tao Xia. I received my B.S. degree in Computer Science and Technology from Harbin Institute of Technology, Shenzhen in 2023. My research interests include Trustworthy AI, Generative Models, and Foundation Models.
And I enjoy playing basketball and doing some workouts in my spare time!

Received Honors

Chinese National Scholarship for Undergraduate Students Top 1.5%

GongJin scholarship Top 1.85%

Excellent Student Model 0.3%

Outstanding Student Award for Scientific Research Commendation of Guangdong Provincial Key Laboratory

Publication

Hao Fang, Bin Chen#, Xuan Wang, Zhi Wang, Shu-Tao Xia, GIFD: A Generative Gradient Inversion Method with Feature Domain Optimization, International Conference on Computer Vision 2023 (ICCV-23), accepted. (CCF A, TsinghuaCS: A). paper, code.

Hao Fang, Jiawei Kong, Bin Chen#, Tao Dai, Hao Wu, Shu-Tao Xia, CLIP-Guided Networks for Transferable Targeted Attacks, European Conference on Computer Vision 2024 (ECCV-24), accepted. (CCF B, TsinghuaCS: A). paper, code.

Yixiang Qiu, Hao Fang (co-first author), Hongyao Yu, Bin Chen#, Meikang Qiu, Shu-Tao Xia, A Closer Look at GAN Priors: Exploiting Intermediate Features for Enhanced Model Inversion Attacks, European Conference on Computer Vision 2024 (ECCV-24 Oral), accepted. (CCF B, TsinghuaCS: A). paper, code.

Xinhao Zhong, Hao Fang (co-first author), Bin Chen#, Xulin Gu, Meikang Qiu, Shuhan Qi, Shu-Tao Xia, Hierarchical Features Matter: A Deep Exploration of Progressive Parameterization Method for Dataset Distillation, Conference on Computer Vision and Pattern Recognition 2025 (CVPR-2025), accepted. (CCF A, TsinghuaCS: A). paper, code.

Tianqu Zhuang, Hongyao Yu, Yixiang Qiu, Hao Fang (co-first author), Bin Chen, Shu-Tao Xia, Stealthy Shield Defense: A Conditional Mutual Information-Based Approach against Black-Box Model Inversion Attacks, The Thirteenth International Conference on Learning Representations (ICLR-2025), accepted. (TsinghuaCS: A). paper, code.

Xinhao Zhong, Bin Chen, Hao Fang , Xulin Gu, Shu-Tao Xia, EN-HUI YANG, Going Beyond Feature Similarity: Effective Dataset distillation based on Class-aware Conditional Mutual Information, The Thirteenth International Conference on Learning Representations (ICLR-2025), accepted. (TsinghuaCS: A). paper, code.

Service

Reviewer for IJCAI 2024, ICLR 2025, CVPR 2025, ICCV 2025

Reviewer for IEEE Transactions on Medical Imaging