Hello 👋, I am Haixin Wang (汪海昕). I am a second-year Ph.D. student in Computer Science at UCLA, advised by Prof. Yizhou Sun and collaborating closely with Prof. Wei Wang. I received the M.S. degree from Peking University. My main research interests include LLM Post-training, Agentic Systems, and Multimodal Learning. I have published several papers at the top international AI Conferences/Journals with total . If you are interested in collaboration, feel free to email me at whx@cs.ucla.edu.

📖 Research Interests

Thesis: Towards Real-world Agentic Systems

• Agentic LLM Post-training — Large-scale fine-tuning (NeurIPS’23, NeurIPS’25, AAAI’24), Agentic RL (ICML’26), RL infrastructure acceleration
• Agentic System Deployment — Agent harness, Multi-agent coordination, Long-horizon agents
• Real-world Environments — Environment scaling, World model simulation (ACM MM’24), Physical AI (ICLR’24, KDD’26)

🔥 News

• 2026.05:  🎉🎉 One paper has been accepted by KDD 2026.
• 2026.04:  🎉🎉 Two papers (One Spotlight) have been accepted by ICML 2026.
• 2025.09:  🎉🎉 Five papers have been accepted by NeurIPS 2025.
• 2025.01:  🎉🎉 One paper has been accepted as oral by AAAI 2025.
• 2024.09:  🎉🎉 I joined the UCLA DATA MINING LAB as a PhD student.

💼 Industry Experience

• 2026.06 – Present, Applied Scientist Intern, Amazon (Rufus, Rufus-Product Team)
• 2026.03 – 2026.06, Research Scientist Intern, Microsoft (CoreAI, GitHub Copilot Team)
• 2025.06 – 2025.11, Applied Scientist Intern, Amazon (Rufus, Foundation Model Team)
• 2023.12 - 2024.03, AI4Science Research Intern, Microsoft Research Asia.

🔖 Selected Publications (* equal contribution, † corresponding author)

• T²PO: Uncertainty-Guided Exploration Control for Stable Multi-Turn Agentic Reinforcement Learning. Haixin Wang, Hejie Cui, Chenwei Zhang, Xin Liu, Shuowei Jin, Shijie Geng, Xinyang Zhang, Nasser Zalmout, Zhenyu Shi, and Yizhou Sun.

• ARLArena: A Unified Framework for Stable Agentic Reinforcement Learning. Xiaoxuan Wang*, Han Zhang*, Haixin Wang*, Yidan Shi, Ruoyan Li, Kaiqiao Han, Chenyi Tong, Haoran Deng, Renliang Sun, Alexander Taylor, Yanqiao Zhu, Jason Cong, Yizhou Sun, Wei Wang.

• FD-Bench: A Modular and Fair Benchmark for Data-driven Fluid Simulation. Haixin Wang, Ruoyan Li, Fred Xu, Fang Sun, Kaiqiao Han, Zijie Huang, Ching Chang, Xiao Luo, Wei Wang, Yizhou Sun.

• Omni-Mol: Multitask Molecular Model for Any-to-any Modalities. Chengxin Hu, Hao Li, Yihe Yuan, Zezheng Song, Chenyang Zhao, and Haixin Wang^†.

• Recent Advances on Machine Learning for Computational Fluid Dynamics: A Survey. Haixin Wang, Yadi Cao, Zijie Huang, Yuxuan Liu, Peiyan Hu, Xiao Luo, Zezheng Song, Wanjia Zhao, Jilin Liu, Jinan Sun, Shikun Zhang, Long Wei, Yue Wang, Tailin Wu, Zhi-Ming Ma, and Yizhou Sun.

• PastNet: Introducing Physical Inductive Biases for Spatio-temporal Video Prediction. Hao Wu, Fan Xu, Chong Chen, Xian-Sheng Hua, Xiao Luo, and Haixin Wang^†.

• BENO: Boundary-embedded Neural Operators for Elliptic PDEs. Haixin Wang*, Jiaxin Li*, Anubhav Dwivedi, Kentaro Hara, and Tailin Wu.

• Parameter-efficient Tuning of Large-scale Multimodal Foundation Model. Haixin Wang*, Xinlong Yang*, Jianlong Chang, Dian Jin, Jinan Sun, Shikun Zhang, Xiao Luo, and Qi Tian.

• Towards Effective Domain Adaptive Retrieval. Haixin Wang, Jinan Sun, Shikun Zhang, Wei Xiang, Chong Chen, Xiansheng Hua, and Xiao Luo.

• Stacking Networks Dynamically for Image Restoration Based on the Plug-and-Play Framework. Haixin Wang, Tianhao Zhang, Muzhi Yu, Jinan Sun, Wei Ye, Chen Wang, and Shikun Zhang.

📄 Academic Services

• Reviewer for Conferences: KDD’23, ICCV’23, NeurIPS’23, AAAI’24, ICLR’24, SDM’24, CVPR’24, ECCV’24, ACM MM’24, ICLR’25, ICML’25, KDD’25, ACM MM’25, NeurIPS’25, AAAI’26, ICLR’26, KDD’26, CVPR’26.
• Reviewer for Journals: IEEE TIFS, Physics of fluids, IEEE TIP, IEEE TPAMI.