avatar

Zhaoyu Chen

Ph.D. Candidate
Academy for Engineering & Technology,
Fudan University
A4008, Interdisciplinary Building No. 2, Jiangwan Campus
zhaoyuchen20 (at) fudan.edu.cn
omenzychen (at) gmail.com


About Me

As a Researcher at TikTok, I focus on the research and application on AI Security and Privacy. Before joining TikTok, I completed my Ph.D. at Fudan University in 2025 under the supervision of Dr. Wenqiang Zhang as part of the Robotics-Oriented Intelligence Lab (ROILab). I hold a B.E. degree, which I received from Shandong University in 2020

I have published more than 20 papers in top international journals and conferences (CVPR, NeurIPS, ICCV, etc.), including 8 papers as first author or co-first author. In addition, I am the team leader of the AI ​​security group in ROILab, supervising 4 PhDs and 2 Masters. Welcome to further cooperation, if you are interested in my work or have ideas, please feel free to contact me: omenzychen (at) gmail.com

Research Interest:

News

Publications [Google Scholar][Semantic Scholar]

(* indicates equal contribution; # indicates corresponding authorship.)

  1. ESWA 2024
    Boosting the Transferability of Adversarial Attacks with Global Momentum Initialization
    Jiafeng Wang*, Zhaoyu Chen*, Kaixun Jiang, Dingkang Yang, Lingyi Hong, Pinxue Guo, Haijing Guo, Wenqiang Zhang#
    Expert Systems with Applications
    PDF Code BibTex

  2. AAAI 2024
    Out of Thin Air - Exploring Data-Free Adversarial Robustness Distillation
    Yuzheng Wang*, Zhaoyu Chen*, Dingkang Yang, Pinxue Guo, Kaixun Jiang, Wenqiang Zhang, Lizhe Qi#
    AAAI 2024
    PDF BibTex

  3. NeurIPS 2023
    Content-based Unrestricted Adversarial Attack
    Zhaoyu Chen, Bo Li#, Shuang Wu, Kaixun Jiang, Shouhong Ding, Wenqiang Zhang#
    NeurIPS 2023
    PDF Code BibTex

  4. ICCV 2023
    Efficient Decision-based Black-box Patch Attacks on Video Recognition
    Kaixun Jiang*, Zhaoyu Chen*, Hao Huang, Jiafeng Wang, Dingkang Yang, Bo Li, Yan Wang, Wenqiang Zhang#
    ICCV 2023
    PDF Code BibTex

  5. TIFS 2023
    Query-Efficient Decision-based Black-Box Patch Attack
    Zhaoyu Chen, Bo Li#, Shuang Wu, Shouhong Ding, Wenqiang Zhang#
    IEEE Transactions on Information Forensics and Security
    PDF BibTex

  6. ACMMM 2023 Oral
    Towards Decision-based Sparse Attacks on Video Recognition
    Kaixun Jiang*, Zhaoyu Chen*, Xinyu Zhou, Jingyu Zhang, Lingyi Hong, JiaFeng Wang, Bo Li, Yan Wang#, Wenqiang Zhang#
    ACMMM 2023 Oral
    PDF BibTex

  7. ACMMM 2023
    Exploring the Adversarial Robustness of Video Object Segmentation via One-shot Adversarial Attacks
    Kaixun Jiang, Lingyi Hong, Zhaoyu Chen#, Pinxue Guo, Zeng Tao, Yan Wang, Wenqiang Zhang#
    ACMMM 2023
    PDF BibTex

  8. ECCV 2022
    Shape Matters - Deformable Patch Attack
    Zhaoyu Chen, Bo Li#, Shuang Wu, Jianghe Xu, Shouhong Ding, Wenqiang Zhang
    ECCV 2022
    PDF Code BibTex

  9. CVPR 2022
    Towards Practical Certifiable Patch Defense with Vision Transformer
    Zhaoyu Chen, Bo Li#, Jianghe Xu, Shuang Wu, Shouhong Ding, Wenqiang Zhang#
    CVPR 2022
    PDF BibTex

Experiences

  • 2025.05-present, TikTok, Research Scientist
  • 2024.06-2025.02, VIVO Imaging Algorithm Research Department, Intern Algorithm Engineer
  • 2021.04-2024.04, Youtu Lab at Tencent, Research Intern

Talks

  1. Poster
    NeurIPS 2023 - Content-based Unrestricted Adversarial Attack
    Zhaoyu Chen
    The main content is how to obtain natural and real unrestricted adversarial examples with strong black-box attack performance.
    PDF Code Video

  2. Lecture
    Adversarial patch attacks and defenses on image classification
    Zhaoyu Chen
    Share the current status of attack and defense and technical research on adversarial patches.
    Video

Services

Journal Reviewers:

  • IEEE: TPAMI, TNNLS, TCSVT, TCDS, TMLR
  • Elsevier: Neural Networks, Knowledge-based System, Machine Learning, Pattern Recognition, International Journal of Human-Computer Studies

Conference Reviewers:

  • Computer Vision: CVPR, ICCV, ECCV, ACCV, WACV
  • Machine Learning and Deep Learning: NeurIPS, ICLR, ICML
  • Artificial Intelligence and Multimodal: AAAI, IJCAI, ACMMM, ICASSP, ECAI

Contact

Address: A4008, Interdisciplinary Building No. 2, Jiangwan Campus, Fudan University
Email: zhaoyuchen20 (at) fudan.edu.cn