Generating Symmetric Materials using Latent Flow Matching

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Karmush, Anmar, Brandenburg, Cedric Mathieu, Ershadrad, Soheil, Rosén, Johanna, Felsberg, Michael, Kelvinius, Filip Ekström
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913110809378816
author Karmush, Anmar
Brandenburg, Cedric Mathieu
Ershadrad, Soheil
Rosén, Johanna
Felsberg, Michael
Kelvinius, Filip Ekström
author_facet Karmush, Anmar
Brandenburg, Cedric Mathieu
Ershadrad, Soheil
Rosén, Johanna
Felsberg, Michael
Kelvinius, Filip Ekström
contents Tackling the task of materials generation, we aim to enhance the previously proposed All-atom Diffusion Transformer (ADiT) by introducing SymADiT, a symmetry-aware variant. To do so, we use a representation of materials based on Wyckoff positions. We follow ADiT and perform generative modelling in latent space, adapted to our symmetry-aware representation. By forcing the output of the generative model to adhere to the symmetry restrictions imposed by the generated crystal's space group and each atom's Wyckoff-position, the generated materials exhibit more realistic symmetry properties. We benchmark our method against both symmetry-aware and symmetry-agnostic models for materials generation and show competitive performance, generating stable, symmetric materials with a simple Transformer architecture.
format Preprint
id arxiv_https___arxiv_org_abs_2605_10115
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Generating Symmetric Materials using Latent Flow Matching
Karmush, Anmar
Brandenburg, Cedric Mathieu
Ershadrad, Soheil
Rosén, Johanna
Felsberg, Michael
Kelvinius, Filip Ekström
Machine Learning
Materials Science
Tackling the task of materials generation, we aim to enhance the previously proposed All-atom Diffusion Transformer (ADiT) by introducing SymADiT, a symmetry-aware variant. To do so, we use a representation of materials based on Wyckoff positions. We follow ADiT and perform generative modelling in latent space, adapted to our symmetry-aware representation. By forcing the output of the generative model to adhere to the symmetry restrictions imposed by the generated crystal's space group and each atom's Wyckoff-position, the generated materials exhibit more realistic symmetry properties. We benchmark our method against both symmetry-aware and symmetry-agnostic models for materials generation and show competitive performance, generating stable, symmetric materials with a simple Transformer architecture.
title Generating Symmetric Materials using Latent Flow Matching
topic Machine Learning
Materials Science
url https://arxiv.org/abs/2605.10115