验证驱动的闭环多智能体LLM框架实现合规结构设计
原标题：Verication-driven closed-loop multi-agent large language modelframework for code-compliant structural design
AI 摘要
该研究提出一种验证驱动的闭环多智能体大语言模型框架，用于符合规范的结构设计。框架将外部物理验证器的反馈注入修复循环，结合三层有限元验证系统与双节点修复机制，使代码合规率从56.8%提升至98.6%，综合评分从63.8提升至71.4。研究还发现性能提升主要归因于外部验证器而非骨干模型，并开源了基准测试和实验脚本。
正文节选
Computer Science > Software Engineering Title:Verication-driven closed-loop multi-agent large language modelframework for code-compliant structural design View PDF HTML (experimental) Abstract:Multi-agent large language model(LLM)systems are applied to structural design,yet most use one-shot generation and cannot verify their output,leaving themill-suited to safety-critical this http URL than trusting LLM self-correction,thisframework injects feedback from an external physics-based v