Autonomous system designs microwave absorbers using natural language
AbsorbEvo: An Agentic Framework for Autonomous Inverse Design of Microwave Absorbers
Artificial Intelligence
Summary
Designing materials that absorb microwaves well usually requires expert knowledge and lots of trial and error. The authors created AbsorbEvo, a system that can understand simple performance goals stated in natural language and automatically generate microwave absorber designs. It uses a clever mix of language-based reasoning, physics predictions, and simulations to find good designs more efficiently. Compared to other methods, AbsorbEvo finds better solutions more often and faster.
What this means in practice
- •For materials engineers: Automatically generate microwave absorber designs by specifying performance goals in plain language, reducing reliance on expert trial-and-error.
- •For antenna designers: Use natural language to guide autonomous design of absorber components for improved electromagnetic performance without manual tuning.
Authors
Zhicheng Feng, Yubo Zhao, Xuefeng Yao
Abstract
Designing high-performance microwave absorbers requires specialized expertise in electromagnetic theory, materials science and simulation programming, and entails time-consuming optimization. Here, we present AbsorbEvo, an agentic framework for autonomous inverse design that translates natural-language performance objectives into designs verified by full-wave simulations. Its candidate evolution strategy integrates language reasoning, physics-based prediction and historical feedback. A large language model proposes the directions and magnitudes of parameter adjustments based on task objectives and computational history. The system combines directed increments with global sampling to generate candidates and uses a low-cost predictive model as a physics prior to rank them. Only high-ranking designs undergo full-wave simulation. Results passing physical validity checks are used to evaluate performance and guide subsequent search. Experience from training tasks is further distilled into textual skills, which are independently validated before use in new tasks. Under identical proposal budgets on held-out AbsorbBench-36 tasks, AbsorbEvo achieved a task success rate of 79.17%, versus 25.00% for a generic agent and 12.50% for random search. Its mean best coverage was 0.7816, compared with 0.6434 and 0.6448, respectively. By integrating language reasoning and physics-based feedback into design decisions, AbsorbEvo provides a methodological foundation for natural-language-driven autonomous inverse design of microwave absorbers.