Hi there! I am Xiangyu Lu, a PhD student in the TMLR group at Hong Kong Baptist University, advised by Prof. Bo Han and working with Prof. Jiangchao Yao (SJTU).

My research interests lie in reasoning and learning in foundation models, with a particular focus on agentic intelligence. I am interested in how AI systems can develop the ability to solve complex problems, learn from experience, and improve through interaction with their environments. My goal is to understand the principles underlying these capabilities and develop methods that make reasoning agents more capable, reliable, and adaptable.

Previously, I received my master’s degree from the University of Copenhagen, where I worked with Dr. Abraham George Smith (DIKU), Dr. Mahoor Mehdikhani (KU Leuven), and Prof. Jon Sporring (DIKU). Before that, I obtained my bachelor’s degree from Nankai University.

I am always open to collaborations and new ideas. Feel free to email me for research discussions or a casual chat.

📣 News

• 2026.09, TACA was accepted to NeurIPS 2026. See you in Sydney!
• 2026.09, I began my PhD journey at Hong Kong Baptist University.
• 2026.05, CoDaPO was accepted to ICML 2026. See you in Seoul!
• 2026.04, We introduced AlphaDiana, a system for evaluating reasoning agents.
• 2025.10, We introduced AlphaApollo, a system for deep agentic reasoning.

📖 Education

• 2026 – present, Ph.D. Student, TMLR Group, Hong Kong Baptist University.
• 2022 – 2024, Master’s degree in IT & Cognition, University of Copenhagen.
• 2018 – 2022, Bachelor’s degree in Intelligent Science and Technology, Nankai University.

📝 Selected Publications

* Co-first author, ✉️ Corresponding author.

Transition-Aware Credit Assignment in Agentic Learning for LLM Reasoning.
Xiangyu Lu, Zhanke Zhou, Jiazhe Ning, Chentao Cao, Jiangchao Yao, Bo Han✉️
NeurIPS 2026.

The Easy, the Hard, and the Learnable: Confidence and Difficulty-Adaptive Policy Optimization for LLM Reasoning.
Zhanke Zhou*, Xiangyu Lu*, Chentao Cao, Brando Miranda, Tongliang Liu, Bo Han✉️, Sanmi Koyejo
ICML 2026. [paper] [code]

AlphaDiana: A System for Evaluating Reasoning Agents.
Zhanke Zhou*, Zongze Li*, Weikai Huang*, Xuan Li*, Chentao Cao*, Xiao Feng, Xiangyu Lu,
Jinbo Hu, Menghan Lu, Yi Xie, Nico Pelleriti, Shiyang Liu, Max Zimmer, Brando Miranda,
Jiangchao Yao, Bo Liu, Sanmi Koyejo, Sebastian Pokutta, Bo Han✉️
Technical Report. [paper] [code]

AlphaApollo: A System for Deep Agentic Reasoning.
Zhanke Zhou*, Chentao Cao*, Xiao Feng*, Xuan Li*, Zongze Li*, Xiangyu Lu*, Jiangchao Yao*,
Weikai Huang, Tian Cheng, Jianghangfan Zhang, Tangyu Jiang, Linrui Xu, Yiming Zheng,
Brando Miranda, Tongliang Liu, Sanmi Koyejo, Masashi Sugiyama, Bo Han✉️
Technical Report. [paper] [code] [website]

🎖 Selected Awards

• 2026.09, PhD Transdisciplinary Research Scholarship Scheme, Hong Kong Baptist University.

🤝 Services

Conference/Workshop reviewer for NeurIPS, ICML, ICLR, ACML, COLM.

📚 Teaching

Instructor for NDAK20002U Neural Information Retrieval, University of Copenhagen.

🌱 Miscellaneous