English

Riemannian Denoising Model for Molecular Structure Optimization with Chemical Accuracy

Machine Learning 2026-01-21 v2 Chemical Physics

Abstract

We introduce a framework for molecular structure optimization using denoising model on a physics-informed Riemannian manifold (R-DM). Unlike conventional approaches operating in Euclidean space, our method leverages a Riemannian metric that better aligns with molecular energy change, enabling more robust modeling of potential energy surfaces. By incorporating internal coordinates reflective of energetic properties, R-DM achieves chemical accuracy with an energy error below 1 kcal/mol. Comparative evaluations on QM9, QM7-X, and GEOM datasets demonstrate improvements in both structural and energetic accuracy, surpassing conventional Euclidean-based denoising models. This approach highlights the potential of physics-informed coordinates for tackling complex molecular optimization problems, with implications for tasks in computational chemistry and materials science.

Keywords

Cite

@article{arxiv.2411.19769,
  title  = {Riemannian Denoising Model for Molecular Structure Optimization with Chemical Accuracy},
  author = {Jeheon Woo and Seonghwan Kim and Jun Hyeong Kim and Woo Youn Kim},
  journal= {arXiv preprint arXiv:2411.19769},
  year   = {2026}
}
R2 v1 2026-06-28T20:16:54.678Z