Hi, I’m Zhipeng Yang (杨致芃). I received my Bachelor’s degree in Robot Engineering from Southeast University, China. My research interests include Explainable and Trustworthy AI. Currently, I’m working on Mechanistic Interpretability for LLM, which reverse engineer the detailed computations performed by transformers. Before that, I also did some works on Action Detection and Robot Task Planning.

I am currently seeking Research Assistant or PhD positions. If you are interested, please feel free to contact me via email: yangzp135@outlook.com.

🔥 News

None.

📝 Publications

* indicates equal contribution.

• Yang, Z., Li, J., Xia, S., & Hu, X.. Internal Chain-of-Thought: Empirical Evidence for Layer‑wise Subtask Scheduling in LLMs. arXiv preprint. arXiv code
• Wang, R.*, Yang, Z.*, Zhao, Z., Tong, X., Hong, Z., & Qian, K.. LLM-based Robot Task Planning with Exceptional Handling for General Purpose Service Robots. In 2024 43rd Chinese Control Conference (CCC). (Oral) arXiv
• Yang, Z.*, Wang, R.*, Tan, Y., & Xie, L.. MALT: Multi-scale Action Learning Transformer for Online Action Detection. In 2024 International Joint Conference on Neural Networks (IJCNN). (Oral) arXiv

📖 Educations

• 2021.08 - 2025.06, Bachelor in Robot Engineering, Southeast University.

💻 Internship

• 2024.10 - 2025.5, Remote Intern, HKUST GZ, supervised by Prof. Xuming Hu.
• 2025.6 - now, Research Assistant, MBZUAI, supervised by Prof. Lijie Hu.