Saved in:
Bibliographic Details
Main Authors: Jehanno, Emmanuel, Menegaux, Romain, Mairal, Julien, Grudinin, Sergei
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.00832
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911177603284992
author Jehanno, Emmanuel
Menegaux, Romain
Mairal, Julien
Grudinin, Sergei
author_facet Jehanno, Emmanuel
Menegaux, Romain
Mairal, Julien
Grudinin, Sergei
contents Crystalline structure prediction is an essential prerequisite for designing materials with targeted properties. Yet, it is still an open challenge in materials design and drug discovery. Despite recent advances in computational materials science, accurately predicting three-dimensional non-polymeric crystal structures remains elusive. In this work, we focus on the molecular assembly problem, where a set $\mathcal{S}$ of identical rigid molecules is packed to form a crystalline structure. Such a simplified formulation provides a useful approximation to the actual problem. However, while recent state-of-the-art methods have increasingly adopted sophisticated techniques, the underlying learning objective remains ill-posed. We propose a better formulation that introduces a loss function capturing key geometric molecular properties while ensuring permutation invariance over $\mathcal{S}$. Remarkably, we demonstrate that within this framework, a simple regression model already outperforms prior approaches, including flow matching techniques, on the COD-Cluster17 benchmark, a curated non-polymeric subset of the Crystallography Open Database (COD).
format Preprint
id arxiv_https___arxiv_org_abs_2509_00832
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Challenges in Non-Polymeric Crystal Structure Prediction: Why a Geometric, Permutation-Invariant Loss is Needed
Jehanno, Emmanuel
Menegaux, Romain
Mairal, Julien
Grudinin, Sergei
Machine Learning
Materials Science
Crystalline structure prediction is an essential prerequisite for designing materials with targeted properties. Yet, it is still an open challenge in materials design and drug discovery. Despite recent advances in computational materials science, accurately predicting three-dimensional non-polymeric crystal structures remains elusive. In this work, we focus on the molecular assembly problem, where a set $\mathcal{S}$ of identical rigid molecules is packed to form a crystalline structure. Such a simplified formulation provides a useful approximation to the actual problem. However, while recent state-of-the-art methods have increasingly adopted sophisticated techniques, the underlying learning objective remains ill-posed. We propose a better formulation that introduces a loss function capturing key geometric molecular properties while ensuring permutation invariance over $\mathcal{S}$. Remarkably, we demonstrate that within this framework, a simple regression model already outperforms prior approaches, including flow matching techniques, on the COD-Cluster17 benchmark, a curated non-polymeric subset of the Crystallography Open Database (COD).
title Challenges in Non-Polymeric Crystal Structure Prediction: Why a Geometric, Permutation-Invariant Loss is Needed
topic Machine Learning
Materials Science
url https://arxiv.org/abs/2509.00832