aiqbench analysis
This research introduces CG-OMatG, a novel generative model that combines Riemannian flow methods with policy gradient reinforcement learning to tackle the complexities of molecular crystal structure prediction. By treating molecules as rigid bodies within a coarse-grained framework, the approach effectively navigates challenging polymorphism and complex packing geometries to generate low-energy candidate structures. The method demonstrates significant potential for accelerating polymorph screening and discovering new organic solid-state materials through advanced AI-driven simulation.
Riemannian Flow Models with Reinforcement Learning for Molecular Crystal Structure Prediction
arXiv:2609.39773v1 Announce Type: new Abstract: Crystal structure governs material properties, making crystal structure prediction (CSP) a fundamental problem in materials science. Generative models are a promising approach for solving this problem, but the prevalence of polymorphism, coupled with large unit cells and complex packing geometry, makes the molecular CSP task challenging for existing models. To address this, we introduce Coarse-Grained Open Materials Generation (CG-OMatG), an equivariant Riemannian flow-based generative model. CG-OMatG predicts molecular crystal structures \textit{via} a coarse-grained, hierarchical representation. CG-OMatG treats molecules as rigid bodies---performing both inter- and intra-molecular message passing to construct a geometric representation for molecular packings---and learns to reconstruct molecule centroid positions, orientations, and lattice parameters, conditioned on chemical species and conformer geometry. We train the model on subsets of the Open Molecular Crystals (OMC25) and Cambridge Structural Database (CSD) datasets. Further, we fine-tune the model \textit{via} policy gradient reinforcement learning to steer the model towards generating low-energy candidate structures. We validate the generated structures on the CSP blind test benchmark, assessing agreement with experimentally determined crystals using COMPACK packing-similarity analysis. CG-OMatG exhibits strong performance for generative molecular crystal structure prediction, paving the way for accelerated polymorph screening and organic solid-state materials discovery.
Authors:Thomas Egg, Harry Winston Sullivan, Maya M. Martirossyan, Philipp Höllmer, Cheng Zeng, Adrian Roitberg, Mingjie Liu, Richard Hennig, Sapna Sarupria, Ellad B. Tadmor, Stefano Martiniani
Abstract:Crystal structure governs material properties, making crystal structure prediction (CSP) a fundamental problem in materials science. Generative models are a promising approach for solving this problem, but the prevalence of polymorphism, coupled with large unit cells and complex packing geometry, makes the molecular CSP task challenging for existing models. To address this, we introduce Coarse-Grained Open Materials Generation (CG-OMatG), an equivariant Riemannian flow-based generative model. CG-OMatG predicts molecular crystal structures \textit{via} a coarse-grained, hierarchical representation. CG-OMatG treats molecules as rigid bodies---performing both inter- and intra-molecular message passing to construct a geometric representation for molecular packings---and learns to reconstruct molecule centroid positions, orientations, and lattice parameters, conditioned on chemical species and conformer geometry. We train the model on subsets of the Open Molecular Crystals (OMC25) and Cambridge Structural Database (CSD) datasets. Further, we fine-tune the model \textit{via} policy gradient reinforcement learning to steer the model towards generating low-energy candidate structures. We validate the generated structures on the CSP blind test benchmark, assessing agreement with experimentally determined crystals using COMPACK packing-similarity analysis. CG-OMatG exhibits strong performance for generative molecular crystal structure prediction, paving the way for accelerated polymorph screening and organic solid-state materials discovery.
Machine Learning (cs.LG); Computational Physics (physics.comp-ph)
arXiv:2609.39773 [cs.LG]
(or arXiv:2609.39773v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2609.39773 arXiv-issued DOI via DataCite (pending registration)
Full text under the publisher's open licence · 238 words
Related coverage
AI and Quantum Computing Briefing
livetradingnews.com · Aug 4, 2026
News
Quantum Computing Report · Jul 25, 2026
Open Materials Generation with Inference-Time Reinforcement Learning
arXiv · Jun 12, 2026
Where Should Physics Enter a Molecular Crystal Generator?
arXiv · Sep 30, 2026