Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2503.06512 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912500616790016 |
|---|---|
| author | Song, Zhilong Zhou, Qionghua Ren, Chunjin Ling, Chongyi Ju, Minggang Wang, Jinlan |
| author_facet | Song, Zhilong Zhou, Qionghua Ren, Chunjin Ling, Chongyi Ju, Minggang Wang, Jinlan |
| contents | Distilling underlying principles from data has historically driven scientific breakthroughs. However, conventional data-driven machine learning often produces complex models that lack interpretability and generalization due to insufficient domain expertise. Here, we present LLM-Feynman, a novel framework that leverages large language models (LLMs) alongside systematic optimization to derive concise, interpretable formulas from data and domain knowledge. Our method integrates automated feature engineering, LLM-guided symbolic regression with self-evaluation, and Monte Carlo tree search to enhance formula discovery and clarity. The embedding of domain knowledge simplifies the formula, while self-evaluation based on this knowledge further minimizes prediction errors, surpassing conventional symbolic regression in accuracy and interpretability. Our LLM-Feynman successfully rediscovered over 90% of fundamental physical formulas and demonstrated its efficacy in key materials science applications, including classification of two-dimensional material and perovskite synthesizability and determination of the Green's function and screened Coulomb interaction bandgaps, and prediction of ionic conductivity in lithium solid-state electrolytes. By transcending mere data fitting through the integration of deep domain knowledge, this LLM-Feynman offers a transformative paradigm for the automated discovery of generalizable scientific formulas and theories across disciplines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_06512 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | LLM-Feynman: Leveraging Large Language Models for Universal Scientific Formula and Theory Discovery Song, Zhilong Zhou, Qionghua Ren, Chunjin Ling, Chongyi Ju, Minggang Wang, Jinlan Materials Science Distilling underlying principles from data has historically driven scientific breakthroughs. However, conventional data-driven machine learning often produces complex models that lack interpretability and generalization due to insufficient domain expertise. Here, we present LLM-Feynman, a novel framework that leverages large language models (LLMs) alongside systematic optimization to derive concise, interpretable formulas from data and domain knowledge. Our method integrates automated feature engineering, LLM-guided symbolic regression with self-evaluation, and Monte Carlo tree search to enhance formula discovery and clarity. The embedding of domain knowledge simplifies the formula, while self-evaluation based on this knowledge further minimizes prediction errors, surpassing conventional symbolic regression in accuracy and interpretability. Our LLM-Feynman successfully rediscovered over 90% of fundamental physical formulas and demonstrated its efficacy in key materials science applications, including classification of two-dimensional material and perovskite synthesizability and determination of the Green's function and screened Coulomb interaction bandgaps, and prediction of ionic conductivity in lithium solid-state electrolytes. By transcending mere data fitting through the integration of deep domain knowledge, this LLM-Feynman offers a transformative paradigm for the automated discovery of generalizable scientific formulas and theories across disciplines. |
| title | LLM-Feynman: Leveraging Large Language Models for Universal Scientific Formula and Theory Discovery |
| topic | Materials Science |
| url | https://arxiv.org/abs/2503.06512 |