I am Kaisheng Pang (庞开升), a Red Bird MPhil student in Robotics and Autonomous Systems (ROAS) at HKUST(GZ), supervised by Professor Haoang Li. During my undergraduate studies, I was advised by Professor Runmin Cong and Associate Professor Feng Li.

My research interests lie in embodied AI 🤖 and computer vision 🌍.

🎓 Education

  • 2026-Present: HKUST(GZ), Red Bird MPhil, Robotics and Autonomous Systems (ROAS)
  • 2022-2026: Shandong University, Undergraduate, Electronic Science and Technology

📰 News

  • 2026.9: Our ActiveScale preprint is now available. [Project Page]
  • 2026.2: DiffLLFace is accepted by TIP 2026.

📚 Publications

# indicates co-first author.

TIP 2026
DiffLLFace: Learning Alternate Illumination-Diffusion Adaption for Low-Light Face Super-Resolution and Beyond

DiffLLFace: Learning Alternate Illumination-Diffusion Adaption for Low-Light Face Super-Resolution and Beyond

Runmin Cong#, Kaisheng Pang#, Feng Li, Hua Li, Huihui Bai, Sam Kwong, Wei Zhang

IEEE Transactions on Image Processing (CCF-A, JCR Q1) 2026

  • Efficient and robust low-light face super-resolution via diffusion-based adaptation.

Paper · Code

Preprint
ActiveScale: Scaling Active Perception for Robots across Model, Data, and Hardware

ActiveScale: Scaling Active Perception for Robots across Model, Data, and Hardware

Shuai Zhou#, Kaisheng Pang#, Wenxuan Song, Wenjie Zhang, Xinhu Zheng, Haoang Li

  • ActiveScale scales active perception through a pose-grounded temporal VLA, human–robot mid-training on 1,000 hours of data, and the AMP mobile-manipulation platform. It achieves state-of-the-art performance across five real-world active-perception tasks.

Paper · Code · Homepage

🎖️ Selected Honors

  • National Scholarship
  • Outstanding Graduate of Shandong University
  • Outstanding Graduation Project of Shandong University
  • First-Class Outstanding Student Award (twice)
  • Outstanding Student of Shandong University (twice)

📬 Contact