AI & ML New Capability

An autonomous AI agent that executes end-to-end theoretical and computational physics research, including hypothesis testing and discovery.

March 23, 2026

Original Paper

PhysMaster: Building an Autonomous AI Physicist for Theoretical and Computational Physics Research

Tingjia Miao, Jiawen Dai, Jingkun Liu, Jinxin Tan, Muhua Zhang, Wenkai Jin, Yuwen Du, Tian Jin, Xianghe Pang, Zexi Liu, Tu Guo, Zhengliang Zhang, Yunjie Huang, Shuo Chen, Rui Ye, Yuzhi Zhang, Linfeng Zhang, Kun Chen, Wei Wang, Weinan E, Siheng Chen

arXiv · 2512.19799

The Takeaway

Unlike standard LLM assistants that just retrieval-augmented, this system couples abstract reasoning with numerical tool execution over ultra-long horizons. It demonstrates the ability to compress months of labor-intensive physics research into hours.

From the abstract

Advances in LLMs have produced agents with knowledge and operational capabilities comparable to human scientists, suggesting potential to assist, accelerate, and automate research. However, existing studies mainly evaluate such systems on well-defined benchmarks or general tasks like literature retrieval, limiting their end-to-end problem-solving ability in open scientific scenarios. This is particularly true in physics, which is abstract, mathematically intensive, and requires integrating analy