Zihan Yang

Curious Mind | Gentle but Resolute

School of Electronics Engineering and Computer Science, Peking University

About Me

I am Zihan Yang, an undergraduate student in the Intelligent Science and Technology program (Zhi Class) at Peking University, expected to graduate in 2028. I currently serve as the Vice President of the Academic Committee of Intelligence Science and Technology (ZhiXueHui).

My research interests lie in Agentic AI, especially the learning, reasoning, planning, and self-improvement of intelligent agents. I am particularly interested in how agents can acquire capabilities through interaction, reinforcement learning, and post-training, and how they can turn high-level goals into reliable actions in complex environments. I am currently advised by Prof. Ling Yang at Discovery Intelligence Group (DIG).

My previous work focuses on physical agents—agents that perceive, reason, learn, and act in the physical world. This includes Code as Policy, Vision-Language Navigation (VLN), Vision-Language-Action Models (VLA), and World Action Models (WAM). More recently, I have been transitioning toward agentic research through projects such as Playful Agentic Robot Learning, with a growing focus on how agents can learn, reason, and improve autonomously across diverse tasks and environments.

I enjoy playing the guitar, shuttlecock kicking, playing table tennis, and chatting with people.

Research Interests

• Recursive Self-Improvement (RSI)
• LLM / Agent Post-Training and Reinforcement Learning
• Physical Agents and Embodied Intelligence
• Vision-Language-Action Models / World Models

Selected Publications

See What I Need, Dream What I Want: A Navigation Foundation Model on a Self-Constructed Representation Space

Jiawei Chen*, Yinxi Wang*†, Zihan Yang*, Meijun Gao*, Yaowei Liu*, Shuming Zhang*, Zecheng Li, Xiangju Mi, Yuxuan Liu, Runzhong Huang, Tianlin Li, Xun Zhang, Zhengyang Wang, Xu Miao✉, Tong Yang✉.

arXiv 2026 Paper

LaST_0.5: A Generalist World Action Model with Interaction-Centric Latent Reasoning

Xuheng Zhang†, Yiming Zhang†, Jialin Gao†, Hao Chen†, Jiaming Liu†, Ji Li, Runzhong Huang, Yinxi Wang, Chenyang Gu, Zihan Yang, Yaowei Liu, Zhonghao Yan, Ziyu Guo, Renrui Zhang, Peng Jia, Pheng-Ann Heng, Shanghang Zhang‡.

arXiv 2026

Playful Agentic Robot Learning

Junyi Zhang*,†, Jiaxin Ge*,†, Hanjun Yoo†, Letian Fu‡, Zihan Yang‡, Yaowei Liu‡, Raj Saravanan‡, Shaofeng Yin, Justin Yu, Dantong Niu, Zirui Wang, Roei Herzig, Ken Goldberg, Yutong Bai, David M. Chan, Ion Stoica, Angjoo Kanazawa, Jiahui Lei§, Haiwen Feng§, Trevor Darrell.

CoRL 2026 Spotlight Paper

VTInstructor: Visual Trajectory Prompting for Navigation Instruction Generation in Continuous Environments

Haolin Yang†, Yuxing Long†, Zihan Yang†, Hao Dong‡.

ACM MM 2026 Paper

NavSpace: How Navigation Agents Follow Spatial Intelligence Instructions

Haolin Yang†, Yuxing Long†, Zhuoyuan Yu, Zihan Yang, Minghan Wang, Jiapeng Xu, Yihan Wang, Ziyan Yu, Wenzhe Cai, Lei Kang, Hao Dong‡.

ICRA 2026 Paper

CorrectNav: Self-Correction Flywheel Empowers Vision-Language-Action Navigation Model

Zhuoyuan Yu†, Yuxing Long†, Zihan Yang, Chengyan Zeng, Hongwei Fan, Jiyao Zhang, Hao Dong‡.

AAAI 2026 Paper

Selected Awards & Funding

Education

• 2024.9 - Present Peking University
• 2021.9 - 2024.9 ShenZhen Middle School
• 2018.9 - 2021.9 Shenzhen Middle School

Contact

• Email: yangzihan@stu.pku.edu.cn
• Institution: Peking University
• Location: Beijing, China