CoChat

Artificial intelligence-driven approaches for materials design and discovery

The DOI matches its Crossref record; 99 of the 99 references it cites were found in a registry or an index.

Abstract

Materials design is an important component of modern science and technology, yet traditional approaches rely heavily on trial and error and can be inefficient. Computational techniques, enhanced by modern artificial intelligence, have reshaped the landscape of designing new materials. Among these approaches, inverse design has shown great promise in designing materials that meet specific property requirements. In this Review, we present key computational advances in materials design over the past few decades. We follow the evolution of relevant materials design techniques, from high-throughput forward machine learning methods and evolutionary algorithms, to advanced artificial intelligence strategies such as reinforcement learning and deep generative models. We highlight the paradigm shift from conventional screening approaches to inverse generation driven by deep generative models. Finally, we discuss current challenges and future perspectives of materials inverse design. This Review may serve as a brief guide to the approaches, progress and outlook of designing future functional materials with technological relevance.

Overview

Integrity signals

Registry

Crossref · 10.1038/s41563-025-02403-7

Cites

99 of 99 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Verification

Claims the paper makes

This paper’s full text isn’t openly available, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Citations

Cites

Show all 99 references

Research works printed but not linked: 32

  1. Mnih, V. Playing Atari with deep reinforcement learning. Preprint at https://arxiv.org/abs/1312.5602 (2013). · arXiv
  2. Christiano, P. F. et al. Deep reinforcement learning from human preferences. Adv. Neural Inf. Process. Syst. 30, (2017).
  3. Kingma, D. P. Auto-encoding variational Bayes. Preprint at https://arxiv.org/abs/1312.6114 (2013). · arXiv
  4. Goodfellow, I. et al. Generative adversarial nets. Adv. Neural Inf. Process. Syst. 27, (2014).
  5. Van Den Oord, A., Kalchbrenner, N. & Kavukcuoglu, K. Pixel recurrent neural networks. In International Conference on Machine Learning 1747–1756 (PMLR, 2016).
  6. Ho, J., Jain, A. & Abbeel, P. Denoising diffusion probabilistic models. Adv. Neural Inf. Process. Syst. 33, 6840–6851 (2020).
  7. Yang, H. et al. MatterSim: a deep learning atomistic model across elements, temperatures and pressures. Preprint at https://arxiv.org/abs/2405.04967 (2024). · arXiv
  8. Carrete, J., Li, W., Mingo, N., Wang, S. & Curtarolo, S. Finding unprecedentedly low-thermal-conductivity half-Heusler semiconductors via high-throughput materials modeling. Phys. Rev. X 4, 011019 (2014).
  9. Luo, Y., Liu, C. & Ji, S. Towards symmetry-aware generation of periodic materials. Adv. Neural Inf. Process. Syst. 36, (2024).
  10. Fernandez-Zelaia, P. et al. Denoising diffusion probabilistic models for generative alloy design. Addit. Manuf. 94, 104478 (2024).
  11. Xie, T., Fu, X., Ganea, O.-E., Barzilay, R. & Jaakkola, T. Crystal diffusion variational autoencoder for periodic material generation. Preprint at https://arxiv.org/abs/2110.06197 (2021). · arXiv
  12. Jiao, R. et al. Crystal structure prediction by joint equivariant diffusion. Adv. Neural Inf. Process. Syst. 36, 17464–17497 (2023).
  13. Tangsongcharoen, K. et al. CrystalGRW: generative modeling of crystal structures with targeted properties via geodesic random walks. Preprint at https://arxiv.org/abs/2501.08998 (2025). · arXiv
  14. Yang, S. et al. Generative hierarchical materials search. Adv. Neural Inf. Process. Syst. 37, 38799–38819 (2025).
  15. Jiao, R., Huang, W., Liu, Y., Zhao, D. & Liu, Y. Space group constrained crystal generation. Preprint at https://arxiv.org/abs/2402.03992 (2024). · arXiv
  16. Ruple, L., Torresi, L., Schopmans, H. & Friederich, P. Symmetry-aware Bayesian flow networks for crystal generation. Preprint at https://arxiv.org/abs/2502.03146 (2025). · arXiv
  17. Kelvinius, F. E. et al. WyckoffDiff—a generative diffusion model for crystal symmetry. Preprint at https://arxiv.org/abs/2502.06485 (2025). · arXiv
  18. Levy, D. et al. SymmCD: symmetry-preserving crystal generation with diffusion models. Preprint at https://arxiv.org/abs/2502.03638 (2025). · arXiv
  19. Lipman, Y., Chen, R. T., Ben-Hamu, H., Nickel, M. & Le, M. Flow matching for generative modeling. Preprint at https://arxiv.org/abs/2210.02747 (2022). · arXiv
  20. Luo, X. et al. CrystalFlow: a flow-based generative model for crystalline materials. Preprint at https://arxiv.org/abs/2412.11693 (2024). · arXiv
  21. Vaswani, A. Attention is all you need. Adv. Neural Inf. Process. Syst. 30, 6000–6010 (2017).
  22. Radford, A. et al. Language models are unsupervised multitask learners. OpenAI Blog 1, 9 (2019).
  23. Achiam, J. et al. GPT-4 technical report. Preprint at https://arxiv.org/abs/2303.08774 (2023). · arXiv
  24. Gruver, N. et al. Fine-tuned language models generate stable inorganic materials as text. Preprint at https://arxiv.org/abs/2402.04379 (2024). · arXiv
  25. Jia, S., Zhang, C. & Fung, V. LLMatDesign: autonomous materials discovery with large language models. Preprint at https://arxiv.org/abs/2406.13163 (2024). · arXiv
  26. Ansari, M., Watchorn, J., Brown, C. E. & Brown, J. S. dZiner: rational inverse design of materials with AI agents. Preprint at https://arxiv.org/abs/2410.03963 (2024). · arXiv
  27. Chen, Z. et al. ScienceAgentBench: toward rigorous assessment of language agents for data-driven scientific discovery. Preprint at https://arxiv.org/abs/2410.05080 (2024). · arXiv
  28. Jaech, A. et al. OpenAI o1 system card. Preprint at https://arxiv.org/abs/2412.16720 (2024). · arXiv
  29. Inizan, T. J. et al. System of agentic AI for the discovery of metal–organic frameworks. Preprint at https://arxiv.org/abs/2504.14110 (2025). · arXiv
  30. Ortega-Ochoa, R., Aspuru-Guzik, A., Vegge, T. & Buonassisi, T. A tomographic interpretation of structure–property relations for materials discovery. Preprint at https://arxiv.org/abs/2501.18163 (2025). · arXiv
  31. Antoniuk, E. R. et al. Active learning enables extrapolation in molecular generative models. Preprint at https://arxiv.org/abs/2501.02059 (2025). · arXiv
  32. Riebesell, J. et al. Matbench discovery—a framework to evaluate machine learning crystal stability predictions. Preprint at https://arxiv.org/abs/2308.14920 (2023). · arXiv

Cited by

See all 186

Metadata

Authors
Cheng, M., Fu, C.-L., Okabe, R., Chotrattanapituk, A., Boonkird, A., Hung, N. T., & Li, M.
Affiliations
  • Massachusetts Institute of Technology (US)
  • Frontier Research Institute for Interdisciplinary Sciences, Tohoku University (JP)
Corresponding author
  • Cheng, Mouyang
  • Li, Mingda
Venue
Nature Materials
Published
2026-01-02
Volume
25
Issue
2
Pages
174-190
ISSN
  • 1476-1122 (ISSN-L)
  • 1476-4660
OpenAlex
W7118011688
Cited by
109
References
96
Citations by year
2026: 108
FWCI
35.48
Type
journal article
Language
en
Open access
closed
Licence
Locations
  • Nature Materials (publishedVersion)
  • PubMed (publishedVersion)
Funders
  • U.S. Department of Energy (DE-SC0021940)
  • National Science Foundation (ITE-2345084)
Topics
  • Physical Sciences › Materials Science › Materials Chemistry › Machine Learning in Materials Science
  • Life Sciences › Pharmacology, Toxicology and Pharmaceutics › Drug Discovery › Computational Drug Discovery Methods
  • Physical Sciences › Materials Science › General Materials Science › Material Selection and Properties
Keywords
materials design, materials discovery, inverse design, deep generative models, reinforcement learning, evolutionary algorithms, machine learning, high-throughput screening, functional materials

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Artificial intelligence-driven approaches for materials design and discovery. https://app.cochat.ai/papers/10.1038/s41563-025-02403-7
The paper
Cheng, M., Fu, C.-L., Okabe, R., Chotrattanapituk, A., Boonkird, A., Hung, N. T., & Li, M. (2026). Artificial intelligence-driven approaches for materials design and discovery. Nature Materials. https://doi.org/10.1038/s41563-025-02403-7

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.