The field of traffic forecasting and network analysis is moving towards the development of more sophisticated models that can effectively capture the complex spatial-temporal correlations and heterogeneity in traffic data. Recent research has focused on improving the accuracy of traffic forecasting models by incorporating multi-scale spatial interactions, dynamic graph learning, and causal reasoning. The use of graph neural networks and graph convolutional networks has shown promise in modeling complex traffic networks and predicting traffic flow. Noteworthy papers in this area include: PSIRAGCN, which proposes a novel graph convolutional network that captures pattern-spatial interactive fusion and regional awareness for traffic forecasting. MSRFormer, which presents a road network representation learning framework that integrates multi-scale spatial interactions and captures complex spatial dependencies across multiple scales. DyC-STG, which proposes a dynamic causal spatio-temporal graph network for real-time data credibility analysis in IoT. STALS, which proposes a spatiotemporal adaptive local search method for tracking congestion propagation in dynamic networks.