用多智能体进化搜索，高效设计复杂物理性能的微观结构。
AutoMS: Multi-Agent Evolutionary Search for Cross-Physics Inverse Microstructure Design
- 用大模型做语义导航，协调智能体分工完成逆向设计。
- 在17个跨物理任务中成功率达83.8%，显著优于传统方法。
- 适合需要高精度物理仿真与复杂目标协同的设计场景。
设计具有耦合跨物理特性的微结构是一项基础挑战，传统拓扑优化计算成本高，深度生成模型常出现物理幻觉。我们提出AutoMS，一种多智能体神经符号框架，将逆向设计重构为大语言模型驱动的进化搜索。AutoMS利用大语言模型作为语义导航器，分解复杂需求并协调智能体流程；同时引入新颖的模拟感知进化搜索（SAES）机制，通过局部梯度近似和定向参数更新实现低层数值优化。该架构在17个多样化的跨物理任务中达到83.8%的最高成功率，显著优于传统进化算法和现有代理基线。通过解耦开放式语义编排与基于仿真的数值搜索，AutoMS为探索难以处理的复杂物理空间提供了稳健路径，超越了标准生成或纯语言方法的局限。
原文摘要 · Abstract (English)
Designing microstructures with coupled cross-physics objectives is a fundamental challenge where traditional topology optimization is often computationally prohibitive and deep generative models frequently suffer from physical hallucinations. We introduce AutoMS, a multi-agent neuro-symbolic framework that reformulates inverse design as an LLM-driven evolutionary search. AutoMS leverages LLMs as semantic navigators to decompose complex requirements and coordinate agent workflows, while a novel Simulation-Aware Evolutionary Search (SAES) mechanism handles low-level numerical optimization via local gradient approximation and directed parameter updates. This architecture achieves a state-of-the-art 83.8% success rate on 17 diverse cross-physics tasks, significantly outperforming both traditional evolutionary algorithms and existing agentic baselines. By decoupling open-ended semantic orchestration from simulation-grounded numerical search, AutoMS provides a robust pathway for navigating complex physical landscapes that remain intractable for standard generative or purely linguistic approaches.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。