About Me ()

I am Yudong Zhang (张宇东), a PhD candidate at the NICS-EFC Laboratory, Department of Electronic Engineering, Tsinghua University, supervised by Professor Yu Wang (汪玉). I obtained a Bachelor’s degree from the Department of Electronic Engineering at Tsinghua University in 2020, under the supervision of Professor Jiansheng Chen (陈健生).

During my Ph.D. studies, my research focuses on enhancing the security and efficiency of vision-language models. To date, I have published 14 papers, including 6 as the first author in top academic conferences such as AAAI, ACM-MM, EMNLP and NAACL. The citation count of these papers is .

I am currently an intern in the Pre-training Group of the Large Language Model Department, Hunyuan Team at Tencent, mentored by Xingwu Sun (孙兴武) and Ruobing Xie (谢若冰). My work mainly focuses on the pre-training of large language models.

If you are interested in academic cooperation or wish to explore potential research opportunities, please feel free to contact me via email.

I will graduate in June 2026 and am actively seeking job opportunities.

📖 Education

• 2020.09 - now, PhD candidate, Department of Electronic Engineering, Tsinghua University, Beijing.
• 2016.09 - 2020.06, Undergraduate, Department of Electronic Engineering, Tsinghua University, Beijing.

🔥 News

• 2025.05: My related experiences were covered by Tsinghua University on Twitter and Facebook. For details: Twitter 1, Twitter 2, Twitter 3, Facebook.

• 2025.04: Won the Excellent Intern Award from the Machine Learning Platform Department, Tencent Technology and Engineering Group (TEG).

• 2024.12: Was named Tsinghua University 2024 Student of the Year.
• 2024.04: Joined the Pre-training Group of the Large Language Model Department, Hunyuan Team at Tencent as an intern, based in Beijing.

📝 Publications

(* indicates equal contribution, † indicates corresponding authors.)

Security of Large Models

• ACM-MM 2024 (Oral) PIP: Detecting Adversarial Examples in Large Vision-Language Models via Attention Patterns of Irrelevant Probe Questions, Yudong Zhang, Ruobing Xie†, Jiansheng Chen†, Xingwu Sun, Yu Wang† | Code | Video
• NAACL 2025 QAVA: Query-Agnostic Visual Attack to Large Vision-Language Models, Yudong Zhang, Ruobing Xie†, Jiansheng Chen†, Xingwu Sun, Zhanhui Kang, Yu Wang† | Code | Video
• ACM-MM 2025 DHCP: Detecting Hallucinations by Cross-modal Attention Pattern in Large Vision-Language Models, Yudong Zhang, Ruobing Xie†, Xingwu Sun, Yiqing Huang, Jiansheng Chen†, Zhanhui Kang, Di Wang, Yu Wang† | Code | Video
• ACM-MM 2025 (Oral) Fighting Fire with Fire (F3): A Training-free and Efficient Visual Adversarial Example Purification Method in LVLMs, Yudong Zhang, Ruobing Xie†, Yiqing Huang, Jiansheng Chen†, Xingwu Sun, Zhanhui Kang, Di Wang, Yu Wang† | Code | Video
• EMNLP 2025 (Findings) The Security Threat of Compressed Projectors in Large Vision-Language Models, Yudong Zhang, Ruobing Xie†, Xingwu Sun, Jiansheng Chen†, Zhanhui Kang, Di Wang, Yu Wang† | Code | Video

Efficiency of pre-training

• AAAI 2025 Enhancing Contrastive Learning Inspired by the Philosophy of “the Blind Men and the Elephant”, Yudong Zhang, Ruobing Xie†, Jiansheng Chen†, Xingwu Sun, Zhanhui Kang, Yu Wang† | Code | Video

• Arxiv A Survey of Linear Attention: Algorithm, Theory, Application, and Infrastructure, Yudong Zhang*, Weixuan Sun*, Xingwu Sun, Wei Ding, Ruiqi Xie, Hongyi Wang, Junjie Chen, Jiacheng Liu, Shaobo Wang, Yuwei Zhang, Yiqing Huang, Jiaming Wang, Tianchen Zhao, Weidong Han, Yanfeng Chen, Kai Zhang, Shuaipeng Li, Ruobing Xie+, Di Wang, Jiansheng Chen, Linfeng Zhang+, Chengzhong Xu, Yu Wang+ | Code

Other Publications (Co-author)

• ACM-MM 2022 3D Human Mesh Reconstruction by Learning to Sample Joint Adaptive Tokens for Transformers, Youze Xue, Jiansheng Chen, Yudong Zhang, Cheng Yu, Huimin Ma, Hongbing Ma

• CIKM 2023 Transferable Structure-based Adversarial Attack of Heterogeneous Graph Neural Network, Yu Shang, Yudong Zhang, Jiansheng Chen, Depeng Jin, Yong Li

• AAAI 2024 Step Vulnerability Guided Mean Fluctuation Adversarial Attack against Conditional Diffusion Models, Hongwei Yu, Jiansheng Chen, Xinlong Ding, Yudong Zhang, Ting Tang, Huimin Ma

• Knowledge-Based Systems (KBS) Image paragraph captioning with topic clustering and topic shift prediction, Ting Tang, Jiansheng Chen, Yiqing Huang, Huimin Ma, Yudong Zhang, Hongwei Yu

• ICCV 2025 DADet: Safeguarding Image Conditional Diffusion Models against Adversarial and Backdoor Attacks via Diffusion Anomaly Detection, Hongwei Yu, Xinlong Ding, Jiawei Li, Jinlong Wang, Yudong Zhang, Rongquan Wang, Huimin Ma, Jiansheng Chen

• Arxiv Towards a Comprehensive Scaling Law of Mixture-of-Experts, Guoliang Zhao, Yuhan Fu, Shuaipeng Li, Xingwu Sun, Ruobing Xie, An Wang, Weidong Han, Zhen Yang, Weixuan Sun, Yudong Zhang, Cheng-zhong Xu, Di Wang, Jie Jiang

• Arxiv Task-Oriented Data Synthesis and Control-Rectify Sampling for Remote Sensing Semantic Segmentation, Yunkai Yang, Yudong Zhang, Kunquan Zhang, Jinxiao Zhang, Xinying Chen, Haohuan Fu, Runmin Dong

🎖 Honors and Awards

Highlighted Honors

• 2025.04 Tencent TEG Machine Learning Platform Department Excellent Intern Award
• 2024.12 Tsinghua University 2024 Student of the Year (selected no more than 10 people annually)
• 2023.12 Tsinghua University Lin Feng Counselor Award (the highest honor for “Double-Role” Political Counselors at Tsinghua University)
• 2018.12 Tsinghua University Jiang Nanxiang Scholarship (the highest scholarship for junior undergraduates at Tsinghua University)
• 2025.12 Tsinghua University Outstanding Teaching Assistant Award（selected 10 people annually）
• 2023.10 Tsinghua University Top Outstanding Student Cadre（selected 30 people annually）
• 2019.11 Tsinghua University Top 10 Volunteers (selected no more than 10 people every two years)
• 2024.12 Feng Chongxi Outstanding Scientific Research Scholarship, Department of Electronic Engineering, Tsinghua University
• 2025.04 As the first completer, I led the project to win the First-Class Award and Third-Class Award of the 35th Tsinghua University Student Laboratory Construction Contribution Award

Other Honors

• 2022.12 Tsinghua University “December 9th” Counselor Award
• 2024.12 Graduate Student “Electronic Star”, Department of Electronic Engineering, Tsinghua University (selected no more than 5 people annually)
• 2019.12 Undergraduate Student “Electronic Star”, Department of Electronic Engineering, Tsinghua University (selected no more than 5 people annually)
• 2025.10 Lu Dajin and Liu Runsheng Excellent Teaching Assistant Award (Grand Prize), Department of Electronic Engineering, Tsinghua University
• 2022.10 Lu Dajin and Liu Runsheng Excellent Teaching Assistant Award (Grand Prize), Department of Electronic Engineering, Tsinghua University
• Tsinghua University Comprehensive Excellent Scholarship (twice during undergraduate studies, five times during doctoral studies)
• Tsinghua University Technological Innovation Excellence Award, Volunteer and Public Welfare Excellence Award, Social Work Excellence Award (twice), Tsinghua Friends - Changfei Scholarship (First-Class and Second-Class)
• Excellent Student Cadre of Tsinghua University (three times)
• 2020.09 As the team leader, I led the Tsinghua University Science and Technology Service Team to win the Tan Haoqiang Computer Education Foundation Scholarship - Excellent Student Group Award

💻 Internships

• 2024.04 - now, Tencent, Large Language Model Department , Beijing.
• 2023.07 - 2024.01, Infinigence, Basic System Group , Beijing.
• 2021.07 - 2021.08, Ministry of Education of the People’s Republic of China, China Education Television (CETV) , Beijing.
• 2019.07 - 2019.08, MEGVII, MEGVII Research Institute , Beijing.

🔧 Patent

Chinese Patents (Granted)

• Kubernetes Container Access Method, Device, Computing Device, and Storage Medium (Patent No.: 2024104387144), Yu Wang, Yudong Zhang.
• Distributed Task Dynamic Service Discovery Method, Device, and Task Training System (Patent No.: 2024104387341), Yu Wang, Yudong Zhang.
• Multi-user Collaborative GPU Computing Power Utilization Method and Device (Patent No.: 2024104384428), Yu Wang, Yudong Zhang (Inventor registration error, change application in progress).
• Sample Purification Method, Device, Equipment, and Medium (Patent No.: 2024109947320), Yudong Zhang, Ruobing Xie, Xingwu Sun, Zhanhui Kang.
• Model Hallucination Detection Method, Device, Equipment, Storage Medium, and Program Product (Patent No.: 2024110474366), Yudong Zhang, Ruobing Xie, Xingwu Sun, Zhanhui Kang.
• Text Generation Method, Device, Equipment, and Readable Storage Medium (Patent No.: 2024109116223), Yudong Zhang, Ruobing Xie, Xingwu Sun, Zhanhui Kang.

Chinese Patents (Pending)

• Adversarial Sample Image Generation Method, Device, Computer Equipment, and Storage Medium (Application No.: 2024110458537), Yudong Zhang, Ruobing Xie, Xingwu Sun, Zhanhui Kang.
• An Image Processing Method and Related Equipment (Application No.: 2024110598759), Yudong Zhang, Ruobing Xie, Xingwu Sun, Zhanhui Kang.
• Image Processing Method, Device, Equipment, Readable Storage Medium, and Program Product (Application No.: 2024111072007), Yudong Zhang, Ruobing Xie, Xingwu Sun, Zhanhui Kang.

🧑‍🏫 Teaching

Teaching Assistant for “Data and Algorithms” of Department of Electronic Engineering, Tsinghua University

• from Spring Semester 2021 to Fall Semester 2025

✍️ Academic Service

Reviewer

• AAAI
• ACM-MM
• ARR（ACL/EMNLP/NAACL）
• CVPR
• ICCV
• ECCV
• ICLR