发表机构
University of Shanghai for Science and Technology（上海理工大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对交通流预测中传统神经网络精度受限及大语言模型时空信息捕获不足的问题，提出动态融合大语言模型DF-LLM，通过时空嵌入、融合及残差连接等组件，在四个数据集上取得更优性能。
AI 中文摘要
交通流预测是智能交通系统的核心支撑技术，它利用历史数据推断特定区域未来的交通动态，从而有助于缓解拥堵并提高资源分配效率。传统神经网络由于依赖单一特征建模，难以突破精度极限，而大语言模型（LLM）在空间拓扑信息捕获和时空相关性挖掘方面存在不足。本研究提出了一种用于交通流预测的动态融合大语言模型（DF-LLM）。该模型包含三个核心组件：时空嵌入模块、时空融合模块和LLM主干。时空嵌入模块能够实现多尺度时空特征的协同表示。时空融合模块通过图卷积整合空间拓扑和动态依赖关系。LLM主干采用差异化参数自适应策略，以平衡训练效率和交通数据适应性。此外，它还引入了上下文聚合注意力模块以增强全局依赖关系。更重要的是，LLM主干采用残差连接以缓解深度网络中的梯度消失问题。实验表明，DF-LLM在所有四个数据集上的指标比较中均取得了更好的性能。
英文摘要
Traffic flow prediction is a core supporting technology for intelligent transportation systems. It uses historical data to infer future traffic dynamics in specific areas, thereby helping to alleviate congestion and improve resource allocation efficiency. Traditional neural networks struggle to break through accuracy limits due to their reliance on singular feature modeling, while large language models (LLMs) suffer from insufficient capture of spatial topological information and mining spatiotemporal correlation. This study proposes a Dynamic Fusion Large Language Model (DF-LLM) for traffic flow prediction. The model incorporates three core components: spatiotemporal embedding module, spatiotemporal fusion module, and LLM backbone. The spatiotemporal embedding module enables synergistic representation of multi-scale spatiotemporal features. The spatiotemporal fusion module integrates spatial topology and dynamic dependencies via graph convolution. The LLM backbone adopts a differentiated parameter adaptation strategy to balance training efficiency and traffic data adaptability. Additionally, it introduces a context aggregation attention module to strengthens global dependencies. More importantly, the LLM backbone takes the residual connections to mitigate the gradient vanishing in deep networks. Experiments show that DF-LLM has achieved better performance by comparing the metrics on all the four datasets.
CommentsAccepted by WISA 2026
