CV

Kaichen He (何凯辰)

2024-Present: Ph.D., School of Intelligence Science and Technology, Peking University, Beijing. Advisor: Prof. Yiwu Zhong (since March 2026); previously advised by Prof. Yitao Liang. Research on multi-modal reinforcement learning in Minecraft.
2020-2024: B.E., Institute for Interdisciplinary Information Sciences (IIIS), Tsinghua University, Beijing.
2017-2020: Yali High School, Hunan. Gold Medalist, the 34th Chinese Mathematical Olympiad (CMO).

2023.09-2024.07: Research intern, Prof. Yitao Liang’s Group.
2023.01-2023.07: Summer research, University of Washington, with Prof. Sheng Wang.
2022.07-2023.01: Research intern, Prof. Zhiyuan Liu’s Group, on natural language processing.

Accepted

Kaichen He, Zihao Wang, Muyao Li, Anji Liu, and Yitao Liang. “CrossAgent: Bridging Cross-level Actions into One Agentic Model via Reinforcement Learning.” IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026. Highlight.
Zihao Wang, Muyao Li, Kaichen He, Xiangyu Wang, Zhancun Mu, Minghao Liu, Anji Liu, and Yitao Liang. “OpenHA: A Series of Open-Source Hierarchical Agentic Models in Minecraft.” International Conference on Machine Learning (ICML), 2026.
Zihao Wang, Muyao Li, Kaichen He, Haowei Lin, Xiaojian Ma, Anji Liu, and Yitao Liang. “DeepHA: Scaling Action Chains Elicits Deep Hierarchical Agents.” International Conference on Machine Learning (ICML), 2026.
Xinyue Zheng, Haowei Lin, Kaichen He, Zihao Wang, Qiang Fu, Haobo Fu, Zilong Zheng, and Yitao Liang. “MCU: An Evaluation Framework for Open-Ended Game Agents.” International Conference on Machine Learning (ICML), 2025. Spotlight.
Muyao Li, Zihao Wang, Kaichen He, Xiaojian Ma, and Yitao Liang. “JARVIS-VLA: Post-Training Large-Scale Vision Language Models to Play Visual Games with Keyboards and Mouse.” Findings of the Association for Computational Linguistics (ACL Findings), 2025.
Weize Chen, Xu Han, Yankai Lin, Kaichen He, Ruobing Xie, Jie Zhou, Zhiyuan Liu, and Maosong Sun. “Hyperbolic Pre-Trained Language Model.” IEEE/ACM Transactions on Audio, Speech, and Language Processing (TASLP), 2024.

Frameworks: PyTorch, PyTorch Lightning, verl, trl, Transformers, OpenRLHF.
Systems: Linux, SLURM cluster management, distributed training with DeepSpeed/FSDP.
Environments: Minecraft (MineDojo/Malmo), ALFWorld, Language-Table.
Languages: Python, C++, LaTeX, shell scripting.