mlsanityData to 7 Oct 2026

Multimodal Crystal Flow: Any-to-Any Modality Generation for Unified Crystal Modeling

arXiv 2602.20210 ↗First posted 23 Feb 2026cs.LG cs.AIcited by 4

Not on the map. Why?

Abstract (arXiv)

Crystal modeling spans a family of conditional and unconditional generation tasks, including crystal structure prediction (CSP) and de novo generation (DNG). While recent deep generative models have shown promising performance, they remain largely task-specific, lacking a unified framework that shares crystal representations across tasks. To address this limitation, we propose Multimodal Crystal Flow (MCFlow), a unified multimodal flow model that realizes multiple crystal generation tasks as distinct inference trajectories via independent time variables for atom types and crystal structures. To enable multimodal flow in a standard transformer model, we introduce a composition- and symmetry-aware atom ordering with hierarchical permutation augmentation, injecting compositional and crystallographic priors without explicit structural templates. Experiments on the MP-20 and MPTS-52 benchmarks show that a single MCFlow model is competitive with task-specific baselines across CSP, DNG, and structure-conditioned atom type generation.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 44 of its 81 references that are in the map. First posted on arXiv 23 Feb 2026.

Builds on

44 of 81 references.

References · 44

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  3. 2022Scalable Diffusion Models with Transformerscited by 5,931
  4. 2022Flow Matching for Generative Modelingcited by 5,426
  5. 2018Neural Ordinary Differential Equationscited by 3,733
  6. 2017MoleculeNet: A Benchmark for Molecular Machine Learningcited by 1,037
  7. 2024Generative Flows on Discrete State-Spaces: Enabling Multimodal Flows with Applications to Protein Co-Designcited by 294
  8. 2022Rectified Flow: A Marginal Preserving Approach to Optimal Transportcited by 277
  9. 2021Crystal Diffusion Variational Autoencoder for Periodic Material Generationcited by 268
  10. 2023Crystal Structure Prediction by Joint Equivariant Diffusioncited by 142
  11. 2024FlowMM: Generating Materials with Riemannian Flow Matchingcited by 105
  12. 2025UMA: A Family of Universal Models for Atomscited by 99
32 more
  1. 2023MatterGen: a generative model for inorganic materials designcited by 98
  2. 2024Fine-Tuned Language Models Generate Stable Inorganic Materials as Textcited by 92
  3. 2024Generator Matching: Generative modeling with arbitrary Markov processescited by 82
  4. 2025All-atom Diffusion Transformers: Unified generative modelling of molecules and materialscited by 75
  5. 2024Space Group Constrained Crystal Generationcited by 71
  6. 2023Scalable Diffusion for Materials Generationcited by 52
  7. 2024FlowLLM: Flow Matching for Material Generation with Large Language Models as Base Distributionscited by 52
  8. 2025SymmCD: Symmetry-Preserving Crystal Generation with Diffusion Modelscited by 48
  9. 2019Generative adversarial networks (GAN) based efficient sampling of chemical space for inverse design of inorganic materialscited by 46
  10. 2024Space Group Informed Transformer for Crystalline Materials Generationcited by 40
  11. 2019Data-Driven Approach to Encoding and Decoding 3-D Crystal Structurescited by 34
  12. 2022Physics Guided Deep Learning for Generative Design of Crystal Materials with Symmetry Constraintscited by 30
  13. 2023Crystal-GFN: sampling crystals with desirable properties and constraintscited by 29
  14. 2025Open Materials Generation with Stochastic Interpolantscited by 26
  15. 2025WyckoffDiff -- A Generative Diffusion Model for Crystal Symmetrycited by 25
  16. 2025Wyckoff Transformer: Generation of Symmetric Crystalscited by 24
  17. 2025LeMat-GenBench: A Unified Evaluation Framework for Crystal Generative Modelscited by 20
  18. 2025Diffuse Everything: Multimodal Diffusion Models on Arbitrary State Spacescited by 19
  19. 2024MatExpert: Decomposing Materials Discovery by Mimicking Human Expertscited by 18
  20. 2022Crystal structure prediction with machine learning-based element substitutioncited by 17
  21. 2025A Periodic Bayesian Flow for Material Generationcited by 15
  22. 2025Periodic Materials Generation using Text-Guided Joint Diffusion Modelcited by 15
  23. 2025Space Group Equivariant Crystal Diffusioncited by 15
  24. 2025LLM Meets Diffusion: A Hybrid Framework for Crystal Material Generationcited by 12
  25. 2025PLaID++: A Preference Aligned Language Model for Targeted Inorganic Materials Designcited by 11
  26. 2025All that structure matches does not glittercited by 10
  27. 2025Flexible MOF Generation with Torsion-Aware Flow Matchingcited by 9
  28. 2025Space Group Conditional Flow Matchingcited by 7
  29. 2025CrystalDiT: A Diffusion Transformer for Crystal Generationcited by 6
  30. 2025CrystalGRW: Generative Modeling of Crystal Structures with Targeted Properties via Geodesic Random Walkscited by 5
  31. 2025Sampling 3D Molecular Conformers with Diffusion Transformerscited by 5
  32. 2025CrystalFormer-CSP: Thinking Fast and Slow for Crystal Structure Predictioncited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 4.

4 papers.

  1. 2026Zatom-1: Towards a Multimodal Foundation Model for 3D Molecules and Materialscited by 3
  2. 2026Discovering Crystal Structure Prediction Algorithms with an AI Co-Scientistcited by 1
  3. 2026uFlowCSP: Crystal Structure Prediction using Mean flow generative modelscited by 0
  4. 2026GLASS: Global Latent Aggregation with Slot-based Set Decoding for Scalable All-Atom Crystal Generationcited by 0