Inverse Design of Optical Multilayer Thin Films using Robust Masked Diffusion Models

2026-04-01

AI Summary

Inverse design of multilayer thin-film structures to match target optical spectra has long been hindered by the vast design space and non-uniqueness of solutions. This work proposes OptoLlama, the first approach to introduce a masked diffusion language model into inverse photonic design. It encodes thin-film stacks as material–thickness sequences and generates high-probability structures conditioned on reflectance, absorptance, and transmittance spectra. By leveraging sequential representation and a spectrum-conditioned generation mechanism, OptoLlama effectively models the probabilistic mapping from spectra to structures. Evaluated on 3,000 test samples, it achieves an average spectral absolute error 2.9 times lower than a nearest-neighbor baseline and 3.45 times lower than the current state-of-the-art data-driven method, OptoGPT, while successfully reproducing canonical designs such as distributed Bragg reflectors.

Abstract

Inverse design of optical multilayer stacks seeks to infer layer materials, thicknesses, and ordering from a desired target spectrum. It is a long-standing challenge due to the large design space and non-unique solutions. We introduce \texttt{OptoLlama}, a masked diffusion language model for inverse thin-film design from optical spectra. Representing multilayer stacks as sequences of material-thickness tokens, \texttt{OptoLlama} conditions generation on reflectance, absorptance, and transmittance spectra and learns a probabilistic mapping from optical response to structure. Evaluated on a representative test set of 3,000 targets, \texttt{OptoLlama} reduces the mean absolute spectral error by 2.9-fold relative to a nearest-neighbor template baseline and by 3.45-fold relative to the state-of-the-art data-driven baseline, called \texttt{OptoGPT}. Case studies on designed and expert-defined targets show that the model reproduces characteristic spectral features and recovers physically meaningful stack motifs, including distributed Bragg reflectors. These results establish diffusion-based sequence modeling as a powerful framework for inverse photonic design.

Problem

• inverse design
• optical multilayer thin films
• non-unique solutions
• large design space
• optical spectra

Innovation

• masked diffusion model
• inverse design
• optical multilayer thin films
• sequence modeling
• photonic design