PRISM: Position-encoded Regressive Inverse Spectral Model for Multilayer Thin-Film Design
This work provides a fast, accurate alternative to classical optimization for inverse design of optical coatings, a complex combinatorial-continuous problem.
PRISM introduces a decoder-only autoregressive transformer for multilayer thin-film design that jointly predicts material selection and thickness, achieving over 50% lower MAE than transformer baselines with one-fifth the parameters, and state-of-the-art MAE of 0.010 with a 44M-parameter variant.
The inverse problem of multilayer thin-film optical coatings design represents a complex combinatorial-continuous optimization challenge. We present PRISM (Position-encoded Regressive Inverse Spectral Model), a unified decoder-only autoregressive transformer that streamlines this process by jointly predicting discrete material selection and continuous thickness regression within a single backbone. PRISM introduces two primary architectural innovations: (1) spectrum prefix conditioning, which utilizes standard prefix tokens for in-context target injection, and (2) cumulative-depth Rotary Position Embeddings, which encode continuous thickness directly into the positional representation to preserve the physical spatial relationships of the stack. Our benchmarks demonstrate that a PRISM-13M model reduces MAE by over 50\% compared to other transformer baselines while utilizing only one-fifth of the parameters. Furthermore, a 44M-parameter variant achieves state-of-the-art performance (MAE = 0.010) on our in-distribution validation benchmark and operates significantly faster than simulated annealing, offering a highly efficient alternative to classical optimization methods.