About Me

I am currently a Master of Science candidate at The Hong Kong University of Science and Technology (Guangzhou), working under the supervision of Prof. Ying-Cong Chen. I completed my undergraduate studies in Management Information Systems at Beijing Jiaotong University, where I was mentored by Prof. Jing Luan and graduated in 2024.

My research interests lie in building scalable, physically grounded, and self-improving embodied AI systems that generate, execute, verify, and refine robot programs toward generalizable manipulation in complex real-world environments:

  • 🧬 Synthetic Simulation & Scalable Data for Embodied AI
  • 🔧 Tool-Integrated Perception & Geometric Reasoning
  • 🕹️ Interactive Feedback & Long-horizon Robot Manipulation

Research Experience

RoboStressBench: four physically grounded visual stress dimensions (Material, Viewpoint, Lighting, Geometry) in embodied scenes

RoboStressBench: Benchmarking VLM Robustness to Physical Visual Stress in Embodied Scenes

Leyi Wu*, Yifan Zhao*, Jinjie Zhang*, Suzeyu Chen*, Wosong Chen, Zhifei Chen, Tianshuo Xu, Qingchun He, Hongxin Hu, Haojian Huang, Yangkai Wei, Wenqian Li, Yinchuan Li, Ying-Cong Chen

* Equal contribution

arXiv:2606.00828 · 2026

arXiv Code (GitHub) Data (Hugging Face)
SPOT-Occ framework: Latent Feature Extraction and Sparse Prototype-guided Transformer Head

SPOT-Occ: Sparse Prototype-guided Transformer for Camera-based 3D Occupancy Prediction

Suzeyu Chen, Leheng Li, Ying-Cong Chen

arXiv:2602.04240 · 2026

arXiv Code (GitHub)

Professional Experience

Knowin.ai Oct 2025 - Present

Robotics Algorithm Engineering Intern, Foundation Model / Multimodal LLM Team

  • Focused on synthetic data generation and evaluation automation for embodied AI and real-world robotic deployment.
  • Developed pipelines for generating, filtering, and validating task-oriented synthetic data.
  • Supported automated evaluation of multimodal large models under robotics-relevant scenarios.
  • Participated in on-device deployment, testing, and debugging of large models on real robot platforms.