CbLDM: A diffusion model for recovering nanostructure from atomic pair distribution functions
APL Machine Learning · 01.12.2026 · 4 (4)
DOI: 10.1063/5.0332941 ↗Kurz gefasst
Fachartikel in der Zeitschrift APL Machine Learning, online erschienen am 18.11.2026. Verfasst von Jiarui Cao und 7 weiteren Autor:innen. Die Arbeit nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Key Research and Development Program of China, Natural Science Foundation of Tianjin Municipality, Fundamental Research Funds for the Central Universities.
Abstract (Originalsprache)
The nanostructure inverse problem is a compelling challenge that helps researchers to understand the structure–property relationships of nanomaterials. This work focuses on recovering a model system of monometallic nanoparticles from their pair distribution function (PDF) data, framing the task as a highly ill-posed conditional generation problem. We introduce a Condition-based Latent Diffusion Model as a practical solution. The model incorporates an accelerated sampling strategy that initializes generation from a conditional-prior sample rather than from pure Gaussian noise, allowing the sampler to skip a contiguous block of intermediate denoising steps, reducing inference time by 8% under the default conservative setting, with a maximum speedup of 27% achievable at the cost of marginal accuracy loss. In addition, we employ a Laplacian matrix in place of a conventional distance matrix to represent nanostructures, which enhances recovery stability. Our results demonstrate that a latent diffusion model with a conditional prior can generate physically meaningful nanostructures consistent with PDF observations, laying the groundwork for solving more complex inverse problems in the future work.
Förderung
- National Key Research and Development Program of China
- Natural Science Foundation of Tianjin Municipality
- Fundamental Research Funds for the Central Universities
- National Natural Science Foundation of China
- U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences
Zitieren
APA
Cao, J., Zhang, Z., Wang, H., Xu, J., Lan, L., Kapasi, N., Billinge, S. J. L., & Gu, R. (2026). CbLDM: A diffusion model for recovering nanostructure from atomic pair distribution functions. APL Machine Learning, 4(4). https://doi.org/10.1063/5.0332941
Vancouver
Cao J, Zhang Z, Wang H, Xu J, Lan L, Kapasi N, et al. CbLDM: A diffusion model for recovering nanostructure from atomic pair distribution functions. APL Machine Learning. 2026;4(4). doi:10.1063/5.0332941
BibTeX
@article{Cao2026,
author = {Cao, Jiarui and Zhang, Zhiyang and Wang, Heming and Xu, Jun and Lan, Ling and Kapasi, Nikhil and Billinge, Simon J. L. and Gu, Ran},
title = {CbLDM: A diffusion model for recovering nanostructure from atomic pair distribution functions},
journal = {APL Machine Learning},
year = {2026},
volume = {4},
number = {4},
doi = {10.1063/5.0332941}
}
EUVIMED