A Self‐Evolving Agentic Framework for Metasurface Inverse Design
- 1Department of Electrical and Computer Engineering University of Massachusetts Lowell Massachusetts USA
Abstract
Metasurface inverse design can realize complex optical functionality, but turning a target optical response into executable optimization code still requires substantial expertise in computational electromagnetics and solver‐specific software engineering. We present a self‐evolving agentic framework that lowers this barrier by coupling a coding agent, explicit human‐readable skill files, and a deterministic physics‐based evaluator. Rather than updating model weights, it revises the skill files from solver‐grounded feedback, while the base model and differentiable solver, which provides the physics simulation and gradients, stay fixed. On a multi‐type benchmark, skill evolution raises same‐type task success from 38% to 74%, the fraction of physical criteria met from 0.51 to 0.87, and reduces average attempts from 4.10 to 2.30. On two new‐type families, success holds near ceiling on one (0.92 to 0.90) and rises from 0.20 to 0.90 on the other. Skill evolution offers a practical path toward autonomous and accessible inverse‐design workflows.
Related
- Inverse design of a fiber-to-fiber integrated metasurface system for multiplexing and demultiplexing of orbital angular momentum modes
- Design Framework for Multifunctional On-Chip Waveguide-Metasurface System for Circularly Polarized and Spatially Controlled Beams
- Deep learning-empowered metasurface design for complex target ISAR illusions
- NanoPhotoNet-Lase: Physics-Informed Neural Networks for Accelerated Design of Coherent Metasurface Nanolasers
- ETALON: Verified Execution and Persistent Research Memory for Agentic Computer-Aided Drug Design
BibTeX
@article{huang2026selfevolving,
title = {{A Self‐Evolving Agentic Framework for Metasurface Inverse Design}},
author = {Yi Huang and Bowen Zheng and Yunxi Dong and Hong Tang and Huan Zhao and S. M. Rakibul Hasan Shawon and Hualiang Zhang},
journal = {Laser \& Photonics Review},
year = {2026},
month = aug,
doi = {10.1002/lpor.71739},
url = {https://doi.org/10.1002/lpor.71739},
note = {CC BY},
}
RIS
TY - JOUR
TI - A Self‐Evolving Agentic Framework for Metasurface Inverse Design
AU - Yi Huang
AU - Bowen Zheng
AU - Yunxi Dong
AU - Hong Tang
AU - Huan Zhao
AU - S. M. Rakibul Hasan Shawon
AU - Hualiang Zhang
PY - 2026
DA - 2026/08/13
JO - Laser & Photonics Review
DO - 10.1002/lpor.71739
UR - https://doi.org/10.1002/lpor.71739
AB - Metasurface inverse design can realize complex optical functionality, but turning a target optical response into executable optimization code still requires substantial expertise in computational electromagnetics and solver‐specific software engineering. We present a self‐evolving agentic framework that lowers this barrier by coupling a coding agent, explicit human‐readable skill files, and a deterministic physics‐based evaluator. Rather than updating model weights, it revises the skill files from solver‐grounded feedback, while the base model and differentiable solver, which provides the physics simulation and gradients, stay fixed. On a multi‐type benchmark, skill evolution raises same‐type task success from 38% to 74%, the fraction of physical criteria met from 0.51 to 0.87, and reduces average attempts from 4.10 to 2.30. On two new‐type families, success holds near ceiling on one (0.92 to 0.90) and rises from 0.20 to 0.90 on the other. Skill evolution offers a practical path toward autonomous and accessible inverse‐design workflows.
SN - 1863-8880
SN - 1863-8899
ER -