Harry Cheng

About

I am a Research Fellow at the National University of Singapore, working with Professor Mohan Kankanhalli. My research centers on AI Safety — building detectors, defenses, and reliable multi-modal systems for a world of generative media.

I received my Ph.D. in Computer Science and Technology from the School of Computer Science and Technology, Shandong University (Qingdao), in June 2025. Before that, I earned my M.Sc. from the School of Computer Science and Technology, Shandong University (2021), and my B.Eng. from the School of Software, Shandong University (2018). I also go by my Chinese name, Xuan-Ang Cheng (程轩昂).

Deepfake Detection Security & Reliability of Diffusion Models Multi-modal Computing Fairness in AI

News

  • 2025 · 11Won 2nd place (2 / 163) in The Competition of Fairness in AI Face Detection @ NeurIPS 25.🥈 2nd
  • 2025 · 10Paper accepted by ACM ToMM.
  • 2025 · 09Paper accepted by NeurIPS 2025.
  • 2025 · 07Paper accepted by ACM MM 2025.
  • 2025 · 06Two papers accepted by ICCV 2025.
  • 2025 · 03Paper accepted by IJCAI 2025.
  • 2024 · 07Two papers accepted by ACM MM 2024.
  • 2023 · 09Paper accepted by ACM ToMM.
  • 2023 · 07Paper accepted by ACM MM 2023.
  • 2023 · 03Paper accepted by ACM ToMM.

Publications

First-author papers, each with a teaser figure. Bold–underlined marks my name; every entry links to its PDF and code.

2026
Teaser Detecting Deepfakes via Hamiltonian Dynamics — teaser figure
Detecting Deepfakes via Hamiltonian Dynamics
Harry Cheng, Ming-Hui Liu, Tianyi Wang, Weili Guan, Liqiang Nie, Mohan Kankanhalli
arXiv 2026Preprint
PDF
2025
Teaser Fair Deepfake Detectors Can Generalize — teaser figure
Fair Deepfake Detectors Can Generalize
Harry Cheng, Ming-Hui Liu, Yangyang Guo, Tianyi Wang, Liqiang Nie, Mohan Kankanhalli
NeurIPS 2025CCF-A 🥈 2nd — Fairness in AI Face Detection @ NeurIPS 25
PDF Code Competition Code
Teaser Social Debiasing for Fair Multi-modal LLMs — teaser figure
Social Debiasing for Fair Multi-modal LLMs
Harry Cheng, Yangyang Guo, Qingpei Guo, Ming Yang, Tian Gan, Weili Guan, Liqiang Nie
ICCV 2025CCF-A
Project
Teaser Towards Generalizable Deepfake Detection by Primary Region Regularization — teaser figure
Towards Generalizable Deepfake Detection by Primary Region Regularization
Harry Cheng, Yangyang Guo, Tianyi Wang, Liqiang Nie, Mohan Kankanhalli
ACM ToMMCCF-B
PDF Code
2024
Teaser Diffusion Facial Forgery Detection — teaser figure
Diffusion Facial Forgery Detection
Harry Cheng, Yangyang Guo, Tianyi Wang, Liqiang Nie, Mohan Kankanhalli
ACM MM 2024CCF-A
PDF Code
2023
Teaser Sample Less, Learn More — teaser figure
Sample Less, Learn More: Efficient Action Recognition via Frame Feature Restoration
Harry Cheng, Yangyang Guo, Liqiang Nie, Zhiyong Cheng, Mohan Kankanhalli
ACM MM 2023CCF-A
PDF Code
Teaser Voice-Face Homogeneity Tells Deepfake — teaser figure
Voice-Face Homogeneity Tells Deepfake
Harry Cheng, Yangyang Guo, Tianyi Wang, Qi Li, Xiaojun Chang, Liqiang Nie
ACM ToMMCCF-B
PDF Code
2022
Teaser Multi-Relation Extraction via a Global–Local Graph Convolutional Network — teaser figure
Multi-Relation Extraction via a Global–Local Graph Convolutional Network
Harry Cheng, Lizi Liao, Linmei Hu, Liqiang Nie
IEEE TBDCCF-C
PDF Code
2021
Teaser Audio-driven Talking Video Frame Restoration — teaser figure
Audio-driven Talking Video Frame Restoration
Harry Cheng, Yangyang Guo, Jianhua Yin, Haonan Chen, Jiafang Wang, Liqiang Nie
IEEE TMMCCF-A
PDF
This page highlights my first-author work. For co-authored and collaborative papers, please see my complete list on Google Scholar.