I’m an M.S. student in Statistics and Data Science at Yale University, where I work as a research assistant with Prof. Zhuoran Yang. Before Yale, I received my B.S. in Computer Science from Nanjing University and spent a semester as an exchange student at UC Berkeley. In summer 2026, I was a Research Intern (AI/ML) at Snowflake AI Research.

My research focuses on large language models, in particular:

• LLM post-training and agentic RL: reinforcement learning for multi-agent systems such as deep research agents.
• Evaluation of LLM agents: benchmarks for agents on realistic, end-to-end tasks in cybersecurity and digital forensics.
• Mechanistic interpretability: understanding what Transformers learn and how they learn it.

Here’s my CV. Feel free to reach out at xintian.pan@yale.edu.

News

• Aug 2026: Finished my research internship at Snowflake AI Research.
• May 2026: Started a Research Intern (AI/ML) position at Snowflake AI Research in Bellevue, WA, working on multi-agent RL for deep research.
• Oct 2025: Released Build Your Personalized Research Group (freephdlabor) on arXiv.
• Aug 2025: Started my M.S. in Statistics and Data Science at Yale University.
• Jul 2025: Graduated from Nanjing University with a B.S. in Computer Science.
• Mar 2025: Released In-Context Linear Regression Demystified on arXiv.

Publications

• Jianliang He, Xintian Pan, Siyu Chen, Zhuoran Yang. In-Context Linear Regression Demystified: Training Dynamics and Mechanistic Interpretability of Multi-Head Softmax Attention. arXiv preprint arXiv:2503.12734, 2025. [arXiv]
• Ed Li, Junyu Ren, Xintian Pan, Cat Yan, Chuanhao Li, Dirk Bergemann, Zhuoran Yang. Build Your Personalized Research Group: A Multiagent Framework for Continual and Interactive Science Automation. arXiv preprint arXiv:2510.15624, 2025. [arXiv] [Code]

Research Experience

Snowflake AI Research, Bellevue, WA — Research Intern (AI/ML), May 2026 – Aug 2026
Manager: Zhewei Yao, Mentor: Boyi Liu. LLM post-training and agentic RL; developed SWARM, multi-agent RL training for deep research.

Yale University, New Haven, CT — Research Assistant, Aug 2025 – Present
Supervisor: Prof. Zhuoran Yang. LLM pre-/post-training, agentic RL, and evaluation of LLM agents (ForensicsBench, AD-Bench, freephdlabor).

Yale University, New Haven, CT — Student Intern, Jul 2024 – Jan 2025
Supervisor: Prof. Zhuoran Yang. Mechanistic interpretability: identified the kernel estimator learned by one-layer Transformers for in-context linear regression, along with its training dynamics.

See Projects for details.