用大模型进化交通灯策略，兼顾效率与环境适配性
Evolutionary Discovery of Heuristic Policies for Traffic Signal Control
- 用大模型作为进化引擎，通过提示词生成专用启发式策略
- 在真实交通数据上表现优于传统规则和在线大模型策略
- 无需训练，适合部署在资源受限的交通管理系统
交通信号控制面临核心矛盾：经典启发式方法高效但过于简化，深度强化学习虽性能高却泛化能力差且策略不透明。在线大语言模型具备通用推理能力，但延迟高且缺乏环境优化。为此，我们提出时序策略进化框架（ extbf{/method{}}），利用大模型作为进化引擎，生成针对特定交通场景的专用启发式策略。该框架包含两个关键模块：(1) 结构化状态抽象（SSA），将高维交通数据转化为时序逻辑事实以支持推理；(2) 信用分配反馈（CAF），追踪错误微观决策与宏观结果之间的因果关系，实现精准批评。整个系统仅通过提示工程完成，无需训练，生成轻量、鲁棒的策略，在特定交通环境中表现超越传统启发式与在线大模型策略。
原文摘要 · Abstract (English)
Traffic Signal Control (TSC) involves a challenging trade-off: classic heuristics are efficient but oversimplified, while Deep Reinforcement Learning (DRL) achieves high performance yet suffers from poor generalization and opaque policies. Online Large Language Models (LLMs) provide general reasoning but incur high latency and lack environment-specific optimization. To address these issues, we propose Temporal Policy Evolution for Traffic (\textbf{\method{}}), which uses LLMs as an evolution engine to derive specialized heuristic policies. The framework introduces two key modules: (1) Structured State Abstraction (SSA), converting high-dimensional traffic data into temporal-logical facts for reasoning; and (2) Credit Assignment Feedback (CAF), tracing flawed micro-decisions to poor macro-outcomes for targeted critique. Operating entirely at the prompt level without training, \method{} yields lightweight, robust policies optimized for specific traffic environments, outperforming both heuristics and online LLM actors.
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。