Structure-Guided Spatiotemporal Attention Graph Neural Network for Traffic Flow Prediction

2026-08-14

AI Summary

This study addresses the lack of transparency in traffic flow prediction models and the inability of existing diagnostic methods to reveal underlying decision mechanisms. We propose SGSAN, a novel framework that learns a static directed dependency graph and integrates an InfoNCE-based soft coupling mechanism to anchor dynamic attention onto structural priors. Furthermore, a decoupled optimization strategy is employed to synergize predictive performance with interpretability. Extensive experiments on multiple real-world datasets demonstrate that SGSAN achieves state-of-the-art accuracy while providing intrinsic interpretability consistent with the physical logic of traffic networks. Consequently, this work effectively resolves the persistent challenge of balancing high-precision forecasting with transparent internal mechanisms, offering a robust solution for interpretable traffic modeling.

Abstract

Deep spatiotemporal models integrating graph convolutions and attention mechanisms have demonstrated excellent performance in network-level traffic flow prediction, owing to their exceptional ability to capture complex spatiotemporal dependencies. Despite their predictive success, deployment of such models in safety-critical urban systems remains constrained by their inherent lack of transparency. Existing post-hoc diagnostic methods often struggle with spurious correlations and fail to unveil the intrinsic decision-making mechanisms governing traffic dynamics, resulting in suboptimal interpretability and limited operational trustworthiness. To address these challenges, this paper proposes the Structure-Guided Spatiotemporal Attention Graph Neural Network (SGSAN). Departing from traditional architectures that rely on unconstrained adaptive graphs, SGSAN explicitly learns a static Directed Dependency Graph (DDG) to identify the invariant macroscopic propagation paths of traffic states. We further introduce an InfoNCE-based soft-coupling mechanism that anchors the model's dynamic spatiotemporal attention to this structural prior, offering a mechanistic account of the model's decision-making process while ensuring robust forecasting by aligning attention-based reasoning with identified macroscopic dependencies and preventing over-reliance on ephemeral local noise. Furthermore, a decoupled two-stage optimization framework is developed to resolve the fundamental conflict between structural discovery and predictive error minimization. Extensive experiments on multiple real-world datasets demonstrate that SGSAN achieves state-of-the-art predictive accuracy while providing built-in interpretability that organically aligns with the physical logic of traffic networks.

Problem

• Traffic Flow Prediction
• Interpretability
• Spatiotemporal Dependencies
• Trustworthiness
• Spurious Correlations

Innovation

• Structure-Guided GNN
• Directed Dependency Graph
• InfoNCE Soft-Coupling
• Decoupled Optimization
• Built-in Interpretability