Universal Magnetic Structure Prediction from Atomic Coordinates with Near-Experimental Accuracy

• URL: http://arxiv.org/abs/2605.16230v1
• Date: Fri, 15 May 2026 17:40:08 GMT
• Title: Universal Magnetic Structure Prediction from Atomic Coordinates with Near-Experimental Accuracy
• Abstract summary: Magnetic structure network (MSN) predicts collinear and non-collinear magnetic structures directly from atomic crystal structures.<n>MSN achieves strong performance across all modulation components and reconstructs experimental magnetic structures with high fidelity.
• Score: 3.574219561129704
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Magnetic order is a fundamental property of materials, governing collective behavior and enabling a broad range of functionalities. Yet magnetic structure remains difficult to determine: experiments are costly and specialized, while first-principles methods often struggle with the noncollinear and incommensurate orders found in real materials. Here we introduce magnetic structure network (MSN), an E(3) equivariant graph neural network that predicts both collinear and non-collinear magnetic structures directly from atomic crystal structures, trained directly on experimentally determined structures from MAGNDATA. By proposing the primitive modulated structure representation (PMSR), we are able to encode commensurate and incommensurate structures in a unified way without symmetry assumptions. The model achieves strong performance across all modulation components and reconstructs experimental magnetic structures with high fidelity. Our approach provides a scalable framework for rapid magnetic structure prediction and opens a route to data-driven discovery of magnetic materials.

Related papers

• Learning Hidden Structures in Open Quantum Dynamics [78.4675086615913]
  We introduce a machine-learning approach for identifying hidden structural features of open quantum dynamics under restricted experimental access.<n>Our framework targets the inference of invariant algebraic structures underlying the effective Markovian evolution.<n>The feasibility of the approach is illustrated on multiple synthetic models and a waveguide quantum electrodynamics system.
  arXiv  Detail & Related papers  (2026-04-01T10:20:38Z)
• RiboSphere: Learning Unified and Efficient Representations of RNA Structures [57.40815107640066]
  RNA backbones are highly flexible, non-canonical interactions are prevalent, and experimentally determined 3D structures are comparatively scarce.<n>We introduce emphRiboSphere, a framework that learns emphdiscrete geometric representations of RNA by combining vector quantization with flow matching.
  arXiv  Detail & Related papers  (2026-03-20T04:36:35Z)
• A first-principles linear response theory for open quantum systems and its application to Orbach and direct magnetic relaxation in Ln-based coordination polymers [39.146761527401424]
  Single-Molecule Magnets (SMMs) exhibit slow magnetic relaxation as a result of axial magnetic anisotropy inhibiting spin-phonon transitions.<n>We numerically implement a first-principles linear-response theory for open quantum systems that provides access to the complex a.c. magnetic susceptibility.
  arXiv  Detail & Related papers  (2026-03-19T10:19:50Z)
• Programmable Exploration of Magnetic States in Lieb-Kagome Interpolated Lattices [44.99833362998488]
  A quantum annealer is used to simulate magnetic interactions in molecular qubit lattices inspired by experimentally realizable systems.<n>The annealer provides access to observables such as the static structure factor and magnetization over a wide parameter space.<n>This framework defines a modular and scalable paradigm for probing the limits of engineered quantum matter across chemistry, condensed matter, and quantum information science.
  arXiv  Detail & Related papers  (2025-07-24T21:47:54Z)
• Observation of Magnetic Devil's Staircase-Like Behavior in Quasiperiodic Qubit Lattices [55.2480439325792]
  devil's staircase (DS) phenomenon is a fractal response of magnetization to external fields.<n>We uncover a wealth of abrupt magnetic transitions driven by increasing external magnetic fields within a simple yet effective Ising-model framework.<n>Our results challenge the prevailing view that DS behavior is limited to periodic systems.
  arXiv  Detail & Related papers  (2025-07-24T21:39:06Z)
• Advancing Magnetic Materials Discovery -- A structure-based machine learning approach for magnetic ordering and magnetic moment prediction [0.0]
  Accurately predicting magnetic behavior across diverse materials systems remains a longstanding challenge.<n>In this work, a refined descriptor is proposed that significantly improves the prediction of two critical magnetic properties.<n>The model achieves an accuracy of 82.4% for magnetic ordering classification and balanced recall across FM and FiM classes.
  arXiv  Detail & Related papers  (2025-07-02T17:24:50Z)
• AlphaFold Database Debiasing for Robust Inverse Folding [58.792020809180336]
  We introduce a Debiasing Structure AutoEncoder (DeSAE) that learns to reconstruct native-like conformations from intentionally corrupted backbone geometries.<n>At inference, applying DeSAE to AFDB structures produces debiased structures that significantly improve inverse folding performance.
  arXiv  Detail & Related papers  (2025-06-10T02:25:31Z)
• Graph-neural-network predictions of solid-state NMR parameters from spherical tensor decomposition [0.0]
  Nuclear magnetic resonance (NMR) is a powerful spectroscopic technique that is sensitive to the local atomic structure of matter.<n>Machine learning (ML) has emerged as an efficient route to making such predictions.
  arXiv  Detail & Related papers  (2024-12-19T17:11:07Z)
• Universal Reconstruction of Complex Magnetic Profiles with Minimum Prior Assumptions [12.951457419224631]
  We present a novel and efficient method for reconstructing spatially varying magnetization density from measured magnetic fields.<n>We validate our method by simulating diverse magnetic structures under realistic experimental conditions.<n>Our approach provides a versatile and universal tool for investigating complex magnetization profiles, paving the way for future quantum sensing experiments.
  arXiv  Detail & Related papers  (2024-11-28T03:15:54Z)
• Amorphous quantum magnets in a two-dimensional Rydberg atom array [44.99833362998488]
  We propose to explore amorphous quantum magnets with an analog quantum simulator. We first present an algorithm to generate amorphous quantum magnets, suitable for Rydberg simulators of the Ising model. We then use semiclassical approaches to get a preliminary insight of the physics of the model.
  arXiv  Detail & Related papers  (2024-02-05T10:07:10Z)
• AI-accelerated Discovery of Altermagnetic Materials [48.261668305411845]
  Altermagnetism, a new magnetic phase, has been theoretically proposed and experimentally verified to be distinct from ferromagnetism and antiferromagnetism.<n>We propose an automated discovery approach empowered by an AI search engine.<n>We successfully discovered 50 new altermagnetic materials that cover metals, semiconductors, and insulators.
  arXiv  Detail & Related papers  (2023-11-08T01:06:48Z)
• Unveiling Exotic Magnetic Phases in Fibonacci Quasicrystalline Stacking of Ferromagnetic Layers through Machine Learning [0.0]
  We study a Fibonacci quasicrystalline stacking of ferromagnetic layers, potentially realizable using van der Waals magnetic materials. We construct a model of this magnetic heterostructure, that displays a complex relationship between geometric frustration and magnetic order in this quasicrystalline system. We employ a machine learning approach, which proves to be a powerful tool in revealing the complex magnetic behavior of this system.
  arXiv  Detail & Related papers  (2023-07-29T19:03:12Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.