Flexible Inverse Design of Optical Multilayer Coatings Using Discrete-Continuous Flow Matching
Abstract
Optical multilayer coatings are used in imaging optics, lasers, telecommunications, solar cells, and radiative cooling, where the spectral response required by each application is achieved by a well-designed layer structure. However, existing design methods have limitations. Local methods refine a starting design efficiently but depend on its quality, and global searches require up to millions of transfer-matrix evaluations per target. Existing deep-learning inverse-design models are fast, but they are trained for a fixed set of materials and a fixed wavelength band, and they either optimize the thicknesses for a fixed material sequence or choose them from a discrete grid. In this paper, we present IrisFlow, a generative model that addresses these limitations by taking each candidate material as its optical constants and the target spectrum together with its wavelengths as input. It generates the material and the continuous thickness of every layer jointly, by coupled discrete and continuous flow matching in a single network. We demonstrate its performance through design with materials outside the training library and on bands beyond the training range, and through comparisons with classical optimizers and an existing deep-learning model. We also demonstrate its usability in a practical design workflow by designing four colored solar-reflective coatings with the calibrated optical constants of an existing deposition system. Combined with a local refinement of its ten best designs, it reduces the number of transfer-matrix evaluations about 500-fold relative to needle optimization at comparable accuracy. The code, trained model, and training and test data are publicly available.
Related
- On the radiation damage tolerance of multilayer and single-layer coatings within the ZrN/SiNx system subjected to helium ion implantation
- Thermoplastic Dielectric Coatings for Flexible Electrodynamic Dust Shields: Structural and Functional Performance
- Disentangling discrete and continuous spectra of tidally forced internal waves in shear flow
- MIRRA: Fast, High-Fidelity Molecular Generation with Chemistry-Informed Discrete Flow Matching
- Continuous limit of a discrete stochastic model of cell migration
BibTeX
@misc{li2026flexible,
title = {{Flexible Inverse Design of Optical Multilayer Coatings Using Discrete-Continuous Flow Matching}},
author = {Zhiyi Li and Yuheng Jin and Yidan Huang and Nan Chen and Hongyan Fu and Yikun Bu},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2607.08392},
eprint = {2607.08392},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2607.08392},
}
RIS
TY - UNPB
TI - Flexible Inverse Design of Optical Multilayer Coatings Using Discrete-Continuous Flow Matching
AU - Zhiyi Li
AU - Yuheng Jin
AU - Yidan Huang
AU - Nan Chen
AU - Hongyan Fu
AU - Yikun Bu
PY - 2026
DA - 2026/10/09
PB - arXiv
DO - 10.48550/arxiv.2607.08392
UR - https://arxiv.org/abs/2607.08392
AB - Optical multilayer coatings are used in imaging optics, lasers, telecommunications, solar cells, and radiative cooling, where the spectral response required by each application is achieved by a well-designed layer structure. However, existing design methods have limitations. Local methods refine a starting design efficiently but depend on its quality, and global searches require up to millions of transfer-matrix evaluations per target. Existing deep-learning inverse-design models are fast, but they are trained for a fixed set of materials and a fixed wavelength band, and they either optimize the thicknesses for a fixed material sequence or choose them from a discrete grid. In this paper, we present IrisFlow, a generative model that addresses these limitations by taking each candidate material as its optical constants and the target spectrum together with its wavelengths as input. It generates the material and the continuous thickness of every layer jointly, by coupled discrete and continuous flow matching in a single network. We demonstrate its performance through design with materials outside the training library and on bands beyond the training range, and through comparisons with classical optimizers and an existing deep-learning model. We also demonstrate its usability in a practical design workflow by designing four colored solar-reflective coatings with the calibrated optical constants of an existing deposition system. Combined with a local refinement of its ten best designs, it reduces the number of transfer-matrix evaluations about 500-fold relative to needle optimization at comparable accuracy. The code, trained model, and training and test data are publicly available.
ER -