硅酸盐学报

晶体结构生成式模型研究进展

PDF下载

魏巍 ¹^,² , 严壮 ¹^,³^,⁴ , 唐智勇 ¹^,²

硅酸盐学报 | AI for Materials专题—综合评述 2026,54(1): 106-121

收起

硅酸盐学报 |AI for Materials专题—综合评述 2026 , 54 (1) : 106 -121

晶体结构生成式模型研究进展

全屏

魏巍¹^,², 严壮¹^,³^,⁴, 唐智勇¹^,²

作者信息

^1．中国科学院纳米系统与多级次制造重点实验室，国家纳米科学中心，北京 100190

^2．中国科学院大学，北京 100049

^3．北京大学国际机器学习研究中心，北京 100871

^4．北京科学智能研究院，北京 100084

通讯作者:

唐智勇（1971—），男，博士，研究员

严壮（1998—），男，博士，助理研究员。

作者简介:

魏巍（1990—），男，博士研究生。

Review on Crystal Structure Generative Models

Wei WEI¹^,², Zhuang YAN¹^,³^,⁴, Zhiyong TANG¹^,²

Affiliations

^1.Cas Key Laboratory of Nano System and Hierarchical Fabrication, National Center for Nanoscience and Technology, Beijing 100190

^2.University of Chinese Academy of Sciences, Beijing 100049

^3.Center for Machine Learning Research, Peking University, Beijing 100871

^4.AI for Science Institute, Beijing 100084

Corresponding Author:

TANG Zhiyong (1971-), male, Ph.D., Professor. E-mail: zytang@nanoctr.cn

YAN Zhuang (1998-), male, Ph.D., Assistant Research Fellow. E-mail: yanz@pku.edu.cn

WEI Wei (1990-), male, Doctoral candidate. E-mail: weiw2025@nanoctr.cn

出版时间: 2026-01-20 doi: 10.14062/j.issn.0454-5648.20250875

文章导航

摘要

收起

晶体材料的物理化学性质根本上取决于其三维空间中原子周期性排列所形成的晶体结构。晶体结构的高效预测与合理设计是新材料研发的关键环节，传统基于密度泛函理论（DFT）的高通量筛选和全局优化方法虽在简单体系中表现突出，但计算开销随原子数增加呈指数增长、对复杂多组分体系适应性差。近年来，人工智能的发展为晶体结构研究提供了新思路，其中，生成式模型（generative models）通过学习晶体数据库中结构与成分的分布规律，可在条件约束下高效探索庞大的化学空间，生成具备目标性能的候选结构。该类模型兼具低推理成本与强泛化能力，为突破传统方法瓶颈提供了关键路径。本文综述生成式模型在晶体结构研究中的最新进展，涵盖生成任务定义、结构表示方法、主要数据集、模型演进及典型应用，并展望其在材料智能设计和理论计算-实验验证闭环系统方面的未来发展方向。

关键词

晶体结构预测 / 生成式模型 / 科学智能 / 晶体材料设计

Abstract

收起

Generative models for crystalline materials have emerged as a transformative paradigm in computational materials science, offering new possibilities for crystal structure prediction (CSP), inverse structural design, and functional materials discovery. Compared with traditional search-based CSP frameworks that depend on iterative global optimization and handcrafted physical constraints, generative approaches learn the statistical distribution of atomic arrangements directly from data, thereby enabling efficient sampling of chemically valid and symmetry-consistent structures. Recent advances in geometric deep learning, equivariant neural networks, variational autoencoders, and diffusion-based modeling have substantially enhanced both the fidelity and scalability of generative crystal models. However, despite the rapid development, a comprehensive and systematic review that integrates representations, generative mechanisms, and practical applications of these models is still sparse.

This review first analyzes the major classes of crystal representations. According to the level of geometric completeness and symmetry preservation, representations can be categorized into reciprocal-space encodings (e.g., diffraction-based descriptors), symmetry-aware Wyckoff-position formalisms, voxelized three-dimensional grids, point-cloud coordinate systems, and graph-based multirelational structures. The specific characteristics of each representation are summarized, including their capabilities for handling periodicity, long-range interactions, spatial boundary conditions, and compatibility with equivariant neural networks. Hybrid schemes—such as the point-cloud/graph combined formulations used in DiffCSP, MatterGen, and GNOME—provide balanced expressiveness and geometric accuracy, enabling efficient modeling of complex crystalline systems.

The generative mechanisms of crystal models are more intricate than those of molecular generative models. According to their learning paradigm, recent models may be divided into four categories. Multimodal-aligned models (e.g., Chemeleon) employ contrastive learning to align textual descriptions with crystal structures, enabling text-conditioned generation across complex chemical spaces. Experiment-driven models (e.g., XtalNet) directly map PXRD spectra to crystal structures via integrated pretraining and conditional generation, circumventing traditional database matching. Twin-model collaborative frameworks (e.g., DAO) combine generative modules with high-precision property predictors, achieving concurrent optimization of structural stability and targeted properties. Dynamics-informed models incorporate nonlinear kinetic effects such as impurity-induced frustration and non-equilibrium crystallization into sampling strategies. These mechanisms collectively provide diverse pathways for addressing the unique challenges posed by crystalline systems.

The applications of generative crystal models are further examined across pharmaceuticals, quantum materials, and energy-related functional materials. In pharmaceutical solid-state chemistry, pocket-conditioned molecular generative models provide insights into integrating solid-form considerations with molecular design. In quantum materials research, constraint-embedded generators such as SCIGEN enable large-scale exploration of lattice-specific structures—including Archimedean and Lieb lattices—yielding candidates for quantum spin liquids, flat-band materials, and exotic magnetic phases. In energy materials, the DAO framework demonstrates the potential of "generation-prediction" coupling by successfully designing superconductors and high-capacity layered cathodes, showing strong coherence between computational predictions and experimental verification.

Summary and prospects

Although notable progress has been made in generative crystal models, significant challenges remain in both scientific understanding and practical deployment. Existing datasets are limited in structural diversity, physical fidelity, and multimodal coverage, making it necessary to construct large-scale, high-quality datasets integrating experimental observations, first-principles calculations, and model-generated structures. The physical interpretability and reliability of generated structures, particularly in multicomponent and nonequilibrium systems, require systematic investigation. Establishing closed-loop platforms that integrate model generation, autonomous synthesis, high-throughput characterization, and iterative retraining remains a key step toward practical materials discovery. Physics-aware constraints, uncertainty quantification, and robust generation under sparse data conditions are essential for improving model reliability. Furthermore, industrial applications demand models capable of handling large unit cells, defect chemistry, compositional tunability, and synthesis feasibility. With the development of large materials models, automated laboratories, and data-centric infrastructures, generative crystalline modeling is expected to become a core driver in building an intelligent, automated, and industrialized materials innovation ecosystem.

Key words

crystal structure prediction / generative models / artificial intelligence for science / crystalline material design

引用本文

魏巍, 严壮, 唐智勇. 晶体结构生成式模型研究进展. 硅酸盐学报, 2026 , 54 (1) : 106 -121 . DOI: 10.14062/j.issn.0454-5648.20250875

Wei WEI, Zhuang YAN, Zhiyong TANG. Review on Crystal Structure Generative Models[J]. Journal of the Chinese Ceramic Society, 2026 , 54 (1) : 106 -121 . DOI: 10.14062/j.issn.0454-5648.20250875

基金

收起

北京市科学技术委员会中关村科技园区管理委员会资助(Z251100007525007)

参考文献

收起

文献

收起

参考文献引证文献

排序方式：

2026年第54卷第1期

PDF下载

引用本文

BibTeX

文章信息

doi: 10.14062/j.issn.0454-5648.20250875

接收时间：2025-12-01
首发时间：2026-10-27
出版时间：2026-01-20

补充材料

相关文章

文章信息

作者

出版历史

收稿日期：2025-12-01
修回日期：2025-12-25

基金

北京市科学技术委员会中关村科技园区管理委员会资助(Z251100007525007)

作者信息

^1．中国科学院纳米系统与多级次制造重点实验室，国家纳米科学中心，北京 100190

^2．中国科学院大学，北京 100049

^3．北京大学国际机器学习研究中心，北京 100871

^4．北京科学智能研究院，北京 100084

通讯作者:

唐智勇（1971—），男，博士，研究员

严壮（1998—），男，博士，助理研究员。

参考文献

分享链接

https://castjournals.cast.org.cn/joweb/gsyxb/CN/10.14062/j.issn.0454-5648.20250875

分享至

全文二维码

扫描看全文

引用本文

BibTeX

本文的引用情况

2种不同金属材料的力学参数

科 Family	属数 Number of genus	种数 Number of species	占总种数比例 Percentage of total species (%)	属 Genus	种数 Number of species	占总种数比例 Percentage of total species (%)
鹅膏菌科Amanitaceae	2	11	5.26	鹅膏菌属 Amanita	10	4.78
小菇科 Mycenaceae	2	12	5.74	丝盖伞属 Inocybe	5	2.39
多孔菌科 Polyporaceae	8	14	6.70	蜡蘑属 Laccaria	5	2.39
红菇科 Russulaceae	3	23	11.00	小皮伞属 Marasmius	6	2.87
				小菇属 Mycena	11	5.26
				光柄菇属 Pluteus	5	2.39
				红菇属 Russula	17	8.13
				栓菌属 Trametes	5	2.39

关闭全屏

BibTeX
EndNote
RefWorks
TxT