Mingyu Zhao

Mingyu Zhao

Ph.D. Student
Rutgers University
zhao.mingyu@rutgers.edu
📄 CV
🔗 GitHub
🔗 Google Scholar
🔗 LinkedIn

Bio

I am a Ph.D. Student in Computer Science at Rutgers University, New Brunswick, advised by Professor Dimitris N. Metaxas. My research interests mainly include Computer Vision (e.g., recognition, classification, segmentation, and 3D reconstruction), Multimodal Large Language Models, Efficient AI, and Multi-step Reasoning with LLMs. I am also interested in diffusion language models. I earned my B.S. degree in Artificial Intelligence from the University of Science and Technology of China (USTC).

Publications

First-author Publications
  1. Y Zhou*, M Zhao*, Z Wang, D Gu, B Guo, R Ye, L Han, C Jin, DN Metaxas
    M3-Bench: Multi-Modal, Multi-Hop, Multi-Threaded Tool-Using MLLM Agent Benchmark
    submitted to CVPR 2026
    paper · code
  2. Z Li*, M Zhao*, X Dong, H Ling, B Huang
    CAPAA: Classifier-Agnostic Projector-Based Adversarial Attack
    IEEE ICME 2025
    paper · code
  3. M Zhao, Z Yang, Y Zhou, Z Xia, C Jin, X He, C Neidle, DN Metaxas
    MHB: Multimodal Handshape-aware Boundary Detection for Continuous Sign Language Recognition
    submitted to WACV 2026
    paper
Other Publications
  1. C Jin, Y Li, M Zhao, S Zhao, Z Wang, X He, L Han, T Che, DN Metaxas
    Lor-VP: Low-rank visual prompting for efficient vision model adaptation
    ICLR 2025
    paper · code
  2. S Zhang, X He, D Liu, Z Xia, M Zhao, C Tan, V Li, B Liu, DN Metaxas, ...
    Large Sign Language Models: Toward 3D American Sign Language Translation
    WACV 2026
    paper