Zhiyi Li

I am a student in the Department of Electronic Engineering at Xiamen University. I build IrisFlow, a generative model for the inverse design of optical multilayer structures. It is a practical inverse-design method: engineers in my lab already use it for their designs, and we are working to integrate it into design software. We have used it to solve several design cases without any traditional optimization; the results are in our paper. I strongly believe it can make optical multilayer design much easier.

More broadly, I believe flow matching can drive inverse design and generation in other fields as well, especially problems where discrete and continuous variables are coupled.

Research

IrisFlow

Inverse design of multilayer optical coatings, with the design space as the input

You give IrisFlow the spectrum you want, the wavelength band, the materials you can deposit (only as their optical curves) and a layer count. It generates candidate coatings from noise, choosing the materials and the thicknesses together, and every candidate is re-simulated exactly.

Try the model How it works

Fiber-optic Fabry–Pérot acoustic sensing

Earlier work on digital signal demodulation for a fiber-optic Fabry–Pérot acoustic sensor. A first-author manuscript is under review.

Publications

Flexible Inverse Design of Optical Multilayer Coatings Using Discrete-Continuous Flow Matching

Zhiyi Li, Yuheng Jin, Yidan Huang, Nan Chen, Hongyan Fu, and Yikun Bu

arXiv preprint arXiv:2607.08392, 2026

First author · preprint under review

Abs arXiv Bib HTML Website

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.
@article{li2026irisflow, title = {Flexible Inverse Design of Optical Multilayer Coatings Using Discrete-Continuous Flow Matching}, author = {Li, Zhiyi and Jin, Yuheng and Huang, Yidan and Chen, Nan and Fu, Hongyan and Bu, Yikun}, journal = {arXiv preprint arXiv:2607.08392}, year = {2026}, note = {First author · preprint under review}, }

Contact

Email is the most reliable way to reach me. Write to me.