Hi, my name is Haoze Wu (吴昊泽 in Chinese). I am a first-year MPhil student at HKUST NLP Group. I am fortunate to be advised by Prof. Junxian He. Before that, I received my B.Eng. degree in Computer Science from CKC Honors College, Zhejiang University, in 2025. I previously spent some time at Shanghai AI Lab and am currently contributing to the Agentic RL team at MiniMax AI.
🧠 Research Interests
My research focuses on scaling agentic Reinforcement Learning, with an emphasis on two closely related directions:
- Environments and supervision — building diverse, realistic, long-horizon environments and developing reliable, dense supervision for agent learning.
- Compute allocation and generalization — understanding how to allocate training compute effectively and when learned capabilities generalize to new tasks, environments, and agent harnesses.
📝 Publications (* denotes co-first authors; † denotes corresponding authors)
Beyond Outcome Rewards: Step-Level Self-Distilled Policy Optimization for Deep Search Agents
Haoze Wu†, Chuqiao Kuang, Tianyi Zhuang, Xiaoguang Li
The Tool Decathlon: Benchmarking Language Agents for Diverse, Realistic, and Long-Horizon Task Execution
Junlong Li*, Wenshuo Zhao*, Jian Zhao*, Weihao Zeng*, Haoze Wu*, Xiaochen Wang, Rui Ge, Yuxuan Cao, Yuzhen Huang, Wei Liu, Junteng Liu, Zhaochen Su, Yiyang Guo, Fan Zhou, Lueyang Zhang, Juan Michelini, Xingyao Wang, Xiang Yue, Shuyan Zhou, Graham Neubig, Junxian He†
Mirage or Method? How Model-Task Alignment Induces Divergent RL Conclusions
Haoze Wu*, Cheng Wang*, Wenshuo Zhao, Junxian He†
ReCode: Updating Code API Knowledge with Reinforcement Learning
Haoze Wu, Yunzhi Yao, Wenhao Yu, Ningyu Zhang†
GW-MoE: Resolving Uncertainty in MoE Router with Global Workspace Theory
Haoze Wu*, Zihan Qiu*, Zili Wang, Hang Zhao†, Jie Fu†
🎓 Education
- 2026.08 - 2028.06 (expected), MPhil@HKUST, CSE
- 2021.09 - 2025.06, B.Eng.@ZJU, Computer Science
💻 Internships
- 2026.06 - Present, MiniMax AI, Shanghai, contributing to Agentic RL team.
- 2024.10 - 2025.01, Shanghai AI Lab, Shanghai, advised by Jie Fu.
🤝 Service
- Reviewer: NeurIPS, ICLR, ACL, EMNLP, ARR, Math-AI@NeurIPS