有状态多智能体LLM实现汽车MBSE跨视图接口对齐
原标题：Stateful Multi-Agent LLMs for Cross-View Interface Alignment in Automotive Model-Based Systems Engineering
AI 摘要
该研究提出了一种有状态的多智能体LLM验证流水线，用于汽车基于模型的系统工程中的跨视图接口对齐。该框架采用顺序生成矩阵和基于车辆信号规范的检索增强生成，并通过独立的AI验证器代理进行对抗性审计，以消除幻觉。在高级驾驶辅助系统场景中，该方法实现了97%的实体可追溯性和85%的F1分数，而标准RAG的实体可追溯性为0%。
正文节选
Computer Science > Software Engineering Title:Stateful Multi-Agent LLMs for Cross-View Interface Alignment in Automotive Model-Based Systems Engineering View PDF HTML (experimental) Abstract:While Large Language Models (LLMs) can accelerate Model-Based Systems Engineering (MBSE) for software-defined vehicles, their probabilistic nature causes "architectural drift", fabricating interfaces in behavioral views that lack structural foundations. To enforce deterministic interface alignmen