arXiv:2606.04036
About
About Me
I am a PhD candidate at Princeton University and a Princeton AI Lab Fellow, working with Prof. Mengdi Wang, Prof. Andrew Yao, and Prof. Quanquan Gu, where my research focuses on building scalable and capable large language models (LLMs) and multimodal foundation models. My work explores methods to develop new attention mechanisms, positional encodings, and model architectures, improve LLM reasoning and agentic AI via reinforcement learning, and advance the data and algorithms behind foundation models.
I am working with the NVIDIA Nemotron team on Agentic RL Scaling and Model Architectures. Previously, I was a Top Seed researcher with the Seed Foundation Model Team, working on LLM and MLLM pretraining and scaling, and a visiting PhD student at the UCLA AGI Lab.
For a concise overview of my research, see my interactive research-talk slides.
If you are interested in my research or projects, I would be happy to discuss potential collaborations via email.
Focus
Research Interests
01
Model Architectures and Continual Learning
02
Agentic Reinforcement Learning Scaling
03
AutoResearch, Automating AI Research
You can find my publications on Google Scholar, and my writing at Yifan's Blog.
Research
Selected Works
Preprint
arXiv:2603.15854
arXiv:2601.00417
International Conference on Learning Representations (ICLR 2026); see also OpenAI's GPT-4 Turbo and GPT-5, xAI's Grok 4, as well as Thinking Machines' Inkling and Jane Street's blog.
International Conference on Learning Representations (ICLR 2026); see also Thinking Machines Tinker and DeepSeek V3.2
Conference on Neural Information Processing Systems (NeurIPS 2025 Spotlight)
International Conference on Machine Learning (ICML 2025)
Transactions on Machine Learning Research (TMLR)
(* denotes equal contribution, † denotes corresponding authors)
Papers
Recent Publications
International Conference on Learning Representations (ICLR 2026)
International Conference on Learning Representations (ICLR 2026); see also Thinking Machines Tinker and DeepSeek V3.2
Annual Meeting of the Association for Computational Linguistics (ACL 2026)
Conference on Neural Information Processing Systems (NeurIPS 2025 Spotlight)
Findings of the Association for Computational Linguistics (ACL 2025 Findings)
International Conference on Machine Learning (ICML 2025)
AAAI Conference on Artificial Intelligence (AAAI 2025)
International Conference on Learning Representations (ICLR 2025 Spotlight)
(* denotes equal contribution)
Writing
Blog Highlights
Service
Professional Activities
Teaching
- Teaching Assistant, Machine Learning for Yao class, IIIS, Tsinghua University
Academic Services
- Conference Reviewer: NeurIPS, ICLR, ICML, COLM, AAAI, AISTATS
- Journal Reviewer: TMLR, IEEE TDSC, ACM TKDD, ACM TIST, Neuralcomputing, Neural Networks
Awards
- William G. Bowen Merit Fellowship at Princeton University (only one in each academic division)
- 2025 Stanford University-Elsevier World's Top 2% Scientist