Pith. sign in

REVIEW 16 cited by

Crystal Diffusion Variational Autoencoder for Periodic Material Generation

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2110.06197 v3 pith:F63YULAF submitted 2021-10-12 cs.LG cond-mat.mtrl-sciphysics.comp-ph

Tian Xie , Xiang Fu , Octavian-Eugen Ganea , Regina Barzilay , Tommi Jaakkola This is my paper
classification cs.LGcond-mat.mtrl-sciphysics.comp-ph
keywords materialsperiodicdiffusiongeneratingmaterialstablestructureatom
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Generating the periodic structure of stable materials is a long-standing challenge for the material design community. This task is difficult because stable materials only exist in a low-dimensional subspace of all possible periodic arrangements of atoms: 1) the coordinates must lie in the local energy minimum defined by quantum mechanics, and 2) global stability also requires the structure to follow the complex, yet specific bonding preferences between different atom types. Existing methods fail to incorporate these factors and often lack proper invariances. We propose a Crystal Diffusion Variational Autoencoder (CDVAE) that captures the physical inductive bias of material stability. By learning from the data distribution of stable materials, the decoder generates materials in a diffusion process that moves atomic coordinates towards a lower energy state and updates atom types to satisfy bonding preferences between neighbors. Our model also explicitly encodes interactions across periodic boundaries and respects permutation, translation, rotation, and periodic invariances. We significantly outperform past methods in three tasks: 1) reconstructing the input structure, 2) generating valid, diverse, and realistic materials, and 3) generating materials that optimize a specific property. We also provide several standard datasets and evaluation metrics for the broader machine learning community.

Discussion (0). Sign in to comment.

Forward citations

Cited by 16 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Extracting Atomic Environments for Machine Learning Interatomic Potentials

    cond-mat.mtrl-sci 2026-07 conditional novelty 6.0 of 10

    For building DFT-sized training cells from large bulk simulations, deleting atoms that collide across periodic boundaries after a cubic cut gives the most accurate forces with low energy artifacts.

  2. Crystalite: A Lightweight Transformer for Efficient Crystal Modeling

    cs.LG 2026-04 conditional novelty 6.0 of 10

    A lightweight diffusion Transformer with Subatomic Tokenization and periodic geometry attention biases reaches SOTA crystal structure prediction and the best reported S.U.N. de novo discovery score among compared base...

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

    cs.LG 2026-02 unverdicted novelty 6.0 of 10

    MCFlow uses decoupled flow time axes for atom types and crystal structures so a single model handles crystal structure prediction, de novo generation, and atom-type generation.

  4. Score-based diffusion models for accurate crystal-structure inpainting and reconstruction of hydrogen positions

    cond-mat.mtrl-sci 2026-01 conditional novelty 6.0 of 10

    Adapting TD-Paint to crystal diffusion models reconstructs hydrogen positions with a LES success rate above 97%, beating unconditioned diffusion and DFT-based inpainting.

  5. Discovery and recovery of crystalline materials with property-conditioned transformers

    cond-mat.mtrl-sci 2025-11 conditional novelty 6.0 of 10

    Conditioning the attention layers of a crystal-writing transformer on continuous property values enables XRD-based structure recovery and targeted generation of photovoltaic candidates.

  6. MiAD: Mirage Atom Diffusion for De Novo Crystal Generation

    cs.LG 2025-11 conditional novelty 6.0 of 10

    Mirage infusion lets crystal diffusion models vary atom counts during generation and raises the S.U.N. rate on MP-20 to 8.2%.

  7. Enhancing Materials Discovery with Valence Constrained Design in Generative Modeling

    cond-mat.mtrl-sci 2025-07 conditional novelty 6.0 of 10

    CrysVCD generates valence-balanced compositions with an elemental language model and then constructs their crystal structures with a diffusion model, reporting improved stability and functional property targeting.

  8. Universal crystal material property prediction via multi-view geometric fusion in graph transformers

    cs.LG 2025-07 conditional novelty 6.0 of 10

    MGT fuses SE3-invariant and SO3-equivariant crystal graph representations with a task-adaptive mixture-of-experts router, achieving state-of-the-art accuracy on multiple crystal property prediction benchmarks.

  9. Hybrid DiffractGPT-Rietveld Refinement Framework for Automated X-ray Diffraction Analysis

    cond-mat.mtrl-sci 2026-07 conditional novelty 5.0 of 10

    A hybrid DiffractGPT–database–Rietveld pipeline returns candidate structures for 93.8% of filtered RRUFF minerals and valid lattice parameters for 79.7% of them (and ~95–98% on an Alexandria subset).

  10. How Far Can You Grow? Characterizing the Extrapolation Frontier of Graph Generative Models for Materials Science

    cond-mat.mtrl-sci 2026-02 conditional novelty 5.0 of 10

    A size-resolved benchmark shows well-behaved crystal generative models degrade by ~13% beyond training radii and follow RMSD ~ N^{1/3}, but the task is deterministic and the headline 'all models' claim is not supporte...

  11. AtomBench: A Benchmarking Framework for Generative Crystal Reconstruction Models in Conventional Superconductors

    cs.LG 2025-10 conditional novelty 5.0 of 10

    On two superconductor datasets, CDVAE best reproduces lattice parameters while AtomGPT (full text) or MatterGen (abstract) best reproduces atomic coordinates, but the comparison is confounded by unequal input information.

  12. AMD-Mamba: A Phenotype-Aware Multi-Modal Framework for Robust AMD Prognosis

    eess.IV 2025-08 unverdicted novelty 5.0 of 10

    AMD-Mamba claims a new multi-modal model and biomarker for early prediction of high-risk age-related macular degeneration, but the submitted manuscript text does not match the abstract and contains placeholder passages.

  13. Property-Guided Diffusion for Inverse Design of Crystalline Materials

    cond-mat.mtrl-sci 2026-07 conditional novelty 4.0 of 10

    An adapter-based classifier-free guidance framework steers a pre-trained crystal diffusion model toward target properties and higher symmetry, with MLIP-based screening reporting modest success rates for stable magnet...

  14. DiffractGPT: Atomic Structure Determination from X-ray Diffraction Patterns using Generative Pre-trained Transformer

    cond-mat.mtrl-sci 2025-08 unverdicted novelty 4.0 of 10

    DiffractGPT predicts crystal structures directly from simulated powder XRD patterns, and its accuracy improves when the chemical composition is provided.

  15. TopoMAS: Large Language Model Driven Topological Materials Multiagent System

    cond-mat.mtrl-sci 2025-07 conditional novelty 4.0 of 10

    TopoMAS is a multi-agent LLM framework that automates retrieval, generation, and first-principles validation for topological materials, reporting 94.55% accuracy with a lightweight Qwen2.5-72B model.

  16. A Survey of AI for Materials Science: Foundation Models, LLM Agents, Datasets, and Tools

    cs.LG 2025-06 unverdicted novelty 4.0 of 10

    This survey organizes foundation models, LLM agents, datasets, and tools in materials science into six task areas.

Pith tools