TL;DR
A novel traffic forecasting framework, FusionGraphSAGE with Neural Networks (FGSNN), which combines a predefined static adjacency matrix based on node distances with a dynamic adaptive graph to better capture evolving traffic relationships and achieves collaborative spatiotemporal feature learning and refined traffic flow prediction.
TETRA is proposed, a hybrid spatio-temporal traffic forecasting model that integrates Graph Convolutional Networks (GCNs) with Extended Long Short-Term Memory (xLSTM) to capture complex multi-timescale temporal patterns, including congestion propagation and delayed recovery dynamics, which are not well represented by conventional recurrent models.
A prediction model that incorporates multiple attention mechanisms with spatiotemporal graph convolutional networks (HASTGCN) and designs a spatiotemporal map convolution module to collaboratively model the dynamic spatiotemporal connection of traffic flow collaboratively model is used.
A Spatial-Temporal Graph Neural Network framework that can learn a combination of spatial and temporal relationships in road networks and changing time-varying patterns in traffic flow is used, which implies that adaptive spatial learning, together with the temporal sequence modeling, can produce much better forecasting stability.
An improved Transformer prediction model that integrates a graph convolutional network (GCN) and a self-attention mechanism is proposed for traffic flow prediction, combining temporal self-attention and learnable temporal encoding to capture both long-term traffic evolution patterns and sudden fluctuations.
Traffic flow prediction is a critical foundational problem in intelligent transportation systems. Although Large Language Model (LLM) has shown promising potential in time series modeling tasks in recent years, existing LLM-based methods generally overlook the inherent multi-scale characteristics of traffic flow data, which significantly limits their ability to capture complex spatio-temporal evolution patterns. To address this issue, this paper proposes a traffic flow forecasting framework named Multi-Scale Graph Convolution Enhanced Large Language Model (MSG-LLM). Firstly, the traffic flow series are decomposed based on frequency-domain analysis to identify periodic components, enabling the adaptive partitioning of the original series into multiple time scales. Subsequently, adaptive graph structures are constructed at different time scales, and graph convolution operations are introduced to fully characterize the correlation dependencies of traffic nodes during multi-scale spatio-temporal evolution. On this basis, a bidirectional multi-scale fusion module is designed to obtain comprehensive and consistent multi-scale representations through information fusion from fine-to-coarse and coarse-to-fine scales. Finally, the fused multi-scale spatio-temporal features are integrated into a partially frozen pre-trained large language model. By fine-tuning only task-specific parameters, this approach preserves the LLM’s general time series modeling capabilities while effectively reducing training costs and mitigating overfitting risks. Extensive experimental results on the PEMS04 and PEMS08 datasets demonstrate that the proposed method significantly outperforms existing mainstream models in both short-term and long-term traffic flow forecasting tasks, validating the effectiveness and strong generalization ability of MSG-LLM in modeling complex traffic systems.
This framework introduces an adaptive graph learning module that dynamically infers meaningful connectivity relationships among traffic sensors—not relying on fixed geographic or distance-based assumptions—but instead leveraging real-time traffic correlations and node-level embeddings, enabling effective modeling of both localized spatial interactions and multi-scale temporal dependencies across varying prediction horizons.