Yuhao Shen

M.Phil Student in Computer Science
School of Data Science
The Chinese University of Hong Kong, Shenzhen

Hello! I’m Yuhao. My primary research interest is AI for Healthcare. As a member of the HEAL Group, I’m fortunate to work under the supervision of Prof. Juexiao Zhou. My current work centers on the following three areas:

Medical Vision Language Models (VLMs) & Agents
LLM Reasoning & Reinforcement learning
Medical Imaging & Computer Vision

I'm always open to academic discussions, potential collaborations and interdisciplinary projects. Feel free to reach out at: yuhaoshen [at] link [dot] cuhk [dot] edu [dot] cn.

🔥 News

  • 2025.06: I graduated from Hangzhou Dianzi University as an outstanding graduate.
  • 2025.05: I officially joined the HEAL Group as an MPhil student.

📚 Publication

# Equal Contribution

CVPR 2026 · Under Review
SkinGPT-R1

SkinGPT-R1: Adapter-Only Dual Distillation for Efficient Dermatology Reasoning
Yuhao Shen#, Jiahe Qian#, Zhangtianyi Chen, Yuanhao He, Juexiao Zhou

  • Introduce SkinGPT-R1, a dermatology-focused VLM that performs explicit, step-by-step, verifiable diagnostic chain-of-thought reasoning.
Scientific Data · Under Review
SkinCaRe Dataset

SkinCaRe: A Multimodal Dermatology Dataset Annotated with Medical Caption and Chain-of-Thought Reasoning [Dataset]
Yuhao Shen#, Liyuan Sun#, Yan Xu#, Wenbin Liu#, Shuping Zhang#, Shawn Afvari, Zhongyi Han, Jiaoyan Song, Yongzhi Ji, Tao Lu, Xiaonan He, Xin Gao, Juexiao Zhou

  • Release SkinCaRe, unifying SkinCAP (medical captions) and SkinCoT (clinician-verified chains-of-thought) for transparent dermatologic reasoning.
ICLR 2026 · Under Review
DermBench & DermEval

Towards Trustworthy Dermatology MLLMs: A Benchmark and Multimodal Evaluator for Diagnostic Narratives
Yuhao Shen#, Jiahe Qian#, Shuping Zhang#, Zhangtianyi Chen, Tao Lu, Juexiao Zhou*

  • Propose DermBench (six clinical dimensions) and DermEval (reference-free evaluator) for image–text dermatology reasoning aligned with physician scoring.
CVPR 2026 · Under Review
CoTBox-TTT

CoTBox-TTT: Grounding Medical VQA with Visual Chain-of-Thought Boxes During Test-time Training
Jiahe Qian#, Yuhao Shen#, Zhangtianyi Chen, Juexiao Zhou, Peisong Wang

  • Evidence-first test-time training with all backbones frozen; update a small set of continuous soft prompts guided by visual chain-of-thought boxes.

📖 Education

  • 2025.09 - Present: M.Phil Student in Computer Science, School of Data Science, The Chinese University of Hong Kong, Shenzhen.
  • 2021.09 - 2025.06: B.Eng. in Computer Science and Technology, School of Computer Science, Hangzhou Dianzi University, Hangzhou.
  • 2018.09 - 2021.06: Jinhua No.1 High School, Zhejiang.

💻 Internship

  • 2025.03 - 2025.05: Reasearch Assistant, The Chinese University of Hong Kong, Shenzhen.
  • 2024.12 - 2025.02: Algorithm Engineer, Hangzhou Huaxin Mechanical and Electrical Engineering Co., Ltd, Hangzhou.
  • 2023.07 - 2023.09: Reasearch Assistant, Hangzhou Institute of Technology of Xidian University, Hangzhou.