PhysMaster: Building an Autonomous AI Physicist for Theoretical and Computational Physics Research
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915876112957440 |
|---|---|
| author | Miao, Tingjia Dai, Jiawen Liu, Jingkun Tan, Jinxin Zhang, Muhua Jin, Wenkai Du, Yuwen Jin, Tian Pang, Xianghe Liu, Zexi Guo, Tu Zhang, Zhengliang Huang, Yunjie Chen, Shuo Ye, Rui Zhang, Yuzhi Zhang, Linfeng Chen, Kun Wang, Wei E, Weinan Chen, Siheng |
| author_facet | Miao, Tingjia Dai, Jiawen Liu, Jingkun Tan, Jinxin Zhang, Muhua Jin, Wenkai Du, Yuwen Jin, Tian Pang, Xianghe Liu, Zexi Guo, Tu Zhang, Zhengliang Huang, Yunjie Chen, Shuo Ye, Rui Zhang, Yuzhi Zhang, Linfeng Chen, Kun Wang, Wei E, Weinan Chen, Siheng |
| contents | 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 analytical reasoning with code-based computation. To address this, we propose PhysMaster, an LLM-based agent functioning as an autonomous theoretical and computational physicist. PhysMaster couples absract reasoning with numerical computation and leverages LANDAU, the Layered Academic Data Universe, which preserves retrieved literature, curated prior knowledge, and validated methodological traces, enhancing decision reliability and stability. It also employs an adaptive exploration strategy balancing efficiency and open-ended exploration, enabling robust performance in ultra-long-horizon tasks. We evaluate PhysMaster on problems from high-energy theory, condensed matter theory to astrophysics, including: (i) acceleration, compressing labor-intensive research from months to hours; (ii) automation, autonomously executing hypothesis-driven loops ; and (iii) autonomous discovery, independently exploring open problems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_19799 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | PhysMaster: Building an Autonomous AI Physicist for Theoretical and Computational Physics Research Miao, Tingjia Dai, Jiawen Liu, Jingkun Tan, Jinxin Zhang, Muhua Jin, Wenkai Du, Yuwen Jin, Tian Pang, Xianghe Liu, Zexi Guo, Tu Zhang, Zhengliang Huang, Yunjie Chen, Shuo Ye, Rui Zhang, Yuzhi Zhang, Linfeng Chen, Kun Wang, Wei E, Weinan Chen, Siheng Artificial Intelligence High Energy Physics - Lattice 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 analytical reasoning with code-based computation. To address this, we propose PhysMaster, an LLM-based agent functioning as an autonomous theoretical and computational physicist. PhysMaster couples absract reasoning with numerical computation and leverages LANDAU, the Layered Academic Data Universe, which preserves retrieved literature, curated prior knowledge, and validated methodological traces, enhancing decision reliability and stability. It also employs an adaptive exploration strategy balancing efficiency and open-ended exploration, enabling robust performance in ultra-long-horizon tasks. We evaluate PhysMaster on problems from high-energy theory, condensed matter theory to astrophysics, including: (i) acceleration, compressing labor-intensive research from months to hours; (ii) automation, autonomously executing hypothesis-driven loops ; and (iii) autonomous discovery, independently exploring open problems. |
| title | PhysMaster: Building an Autonomous AI Physicist for Theoretical and Computational Physics Research |
| topic | Artificial Intelligence High Energy Physics - Lattice |
| url | https://arxiv.org/abs/2512.19799 |