EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Agentic, Autonomous Design Evolution of Machine-Learned Interatomic Potentials

Muyu Lu, Danyang Chen, Fan Yu, Jun Jiang, Linjiang Chen

Precision Chemistry · 29.10.2026

DOI: 10.1021/prechem.6c00099 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Precision Chemistry, online erschienen am 29.10.2026. Verfasst von Muyu Lu und 4 weiteren Autor:innen. Die Arbeit nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: University of Science and Technology of China, Chinese Academy of Sciences, National Advanced Talent Cultivation Center for Chemistry, USTC.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Recent AI systems can autonomously execute scientific workflows and empirical software tasks. We investigate whether an autonomous agent can use evaluation feedback to iteratively develop a scientific model. To address this question, we developed AutoResearch-MLIP, a research protocol in which the agent follows predefined instructions to propose, implement, and evaluate MLIP designs, with outcomes carried forward to guide subsequent iterations. Starting from a simple pair-distance MLIP, each generation selects a parent model by balancing exploration and exploitation, while structured design records guide subsequent proposals. Across 40 generations, the agent explores MLIP designs spanning local message passing, radial representations, scalar-vector states, body-order terms, residual readouts, long-range-inspired terms, and periodic image-offset graph construction. The final G40 model is competitive with established MLIP baselines in post hoc benchmarks. These results show that, within predefined rules, an autonomous agent can use evaluation feedback to iteratively improve MLIP implementations.

Förderung

  • University of Science and Technology of China
  • Chinese Academy of Sciences
  • National Advanced Talent Cultivation Center for Chemistry, USTC
  • National Natural Science Foundation of China (NSFC)
  • USTC Bihe Youth Program for Interdisciplinary Innovation

Zitieren

APA

Lu, M., Chen, D., Yu, F., Jiang, J., & Chen, L. (2026). Agentic, Autonomous Design Evolution of Machine-Learned Interatomic Potentials. Precision Chemistry. https://doi.org/10.1021/prechem.6c00099

Vancouver

Lu M, Chen D, Yu F, Jiang J, Chen L. Agentic, Autonomous Design Evolution of Machine-Learned Interatomic Potentials. Precision Chemistry. 2026. doi:10.1021/prechem.6c00099

BibTeX
@article{Lu2026,
  author = {Lu, Muyu and Chen, Danyang and Yu, Fan and Jiang, Jun and Chen, Linjiang},
  title = {Agentic, Autonomous Design Evolution of Machine-Learned Interatomic Potentials},
  journal = {Precision Chemistry},
  year = {2026},
  doi = {10.1021/prechem.6c00099}
}
Kostenlos starten →