Summary

Ph.D. student at Peking University, working on multi-modal reinforcement learning, embodied AI, and vision-language agents.

Education

• School of Intelligence Science and Technology

  Present
  Peking University

  Courses: Advisor: Prof. Yiwu Zhong (since March 2026); previously advised by Prof. Yitao Liang, Research on multi-modal reinforcement learning in Minecraft
• Institute for Interdisciplinary Information Sciences (IIIS)

  2024
  Tsinghua University
• Hunan

  2020
  Yali High School

  Courses: Gold Medalist, the 34th Chinese Mathematical Olympiad (CMO)

Work Experience

• Research Intern

  2023-09 - 2024-07
• Summer Researcher with Prof. Sheng Wang

  2023-01 - 2023-07
• Research Intern

  2022-07 - 2023-01
  Research on natural language processing.

Skills

Publications

• CrossAgent: Bridging Cross-level Actions into One Agentic Model via Reinforcement Learning

  2026
  IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

  CVPR 2026 Highlight. Authors: Kaichen He, Zihao Wang, Muyao Li, Anji Liu, and Yitao Liang.

  View Publication
• MCU: An Evaluation Framework for Open-Ended Game Agents

  2025
  International Conference on Machine Learning (ICML)

  ICML 2025 Spotlight. Authors: Xinyue Zheng, Haowei Lin, Kaichen He, Zihao Wang, Qiang Fu, Haobo Fu, Zilong Zheng, and Yitao Liang.

  View Publication
• JARVIS-VLA: Post-Training Large-Scale Vision Language Models to Play Visual Games with Keyboards and Mouse

  2025
  Findings of the Association for Computational Linguistics (ACL Findings)

  Authors: Muyao Li, Zihao Wang, Kaichen He, Xiaojian Ma, and Yitao Liang.

  View Publication
• Hyperbolic Pre-Trained Language Model

  2024
  IEEE/ACM Transactions on Audio, Speech, and Language Processing (TASLP)

  Authors: Weize Chen, Xu Han, Yankai Lin, Kaichen He, Ruobing Xie, Jie Zhou, Zhiyuan Liu, and Maosong Sun.

  View Publication
• OpenHA: A Series of Open-Source Hierarchical Agentic Models in Minecraft

  2026
  International Conference on Machine Learning (ICML)

  Authors: Zihao Wang, Muyao Li, Kaichen He, Xiangyu Wang, Zhancun Mu, Minghao Liu, Anji Liu, and Yitao Liang.

  View Publication
• DeepHA: Scaling Action Chains Elicits Deep Hierarchical Agents

  2026
  International Conference on Machine Learning (ICML)

  Authors: Zihao Wang, Muyao Li, Kaichen He, Haowei Lin, Xiaojian Ma, Anji Liu, and Yitao Liang.

  View Publication

Interests

• Research

  Multi-modal reinforcement learning, Embodied AI, Vision-language-action models, Minecraft agents