aiqbench analysis
PatchSTG addresses the scalability bottleneck in traffic forecasting by introducing a hierarchical patch-based architecture that reduces computational complexity from quadratic to near-linear scaling. This approach effectively handles irregular sensor distributions through a dual attention mechanism that balances local interactions with global dependencies, offering a more efficient alternative to standard graph-based models. The method demonstrates stable performance on real-world datasets, validating its potential for large-scale intelligent transportation systems.
PatchSTG: Scalable Spatiotemporal Graph Transformers for Traffic Forecasting on Irregular Sensor Networks
arXiv:2606.09872v1 Announce Type: new Abstract: Traffic forecasting is a fundamental component of intelligent transportation systems, yet remains challenging in real-world settings due to irregular sensor distributions and the high computational cost of modeling large-scale spatiotemporal dependencies. In practical traffic networks, sensors are unevenly distributed across regions, leading to non-uniform spatial structures that limit the effectiveness and scalability of existing graph-based and attention-based models. To address these challenges, we propose PatchSTG, a patch-based spatiotemporal graph Transformer designed for efficient forecasting on irregular sensor networks. The key idea is to introduce a hierarchical spatial representation that partitions sensors into balanced, locality-preserving patches based on geographic information. On top of this structure, a dual attention encoder alternates between intra-patch attention for capturing local interactions and inter-patch attention for modeling global dependencies, reducing computational complexity from quadratic to near-linear scaling. We evaluate PatchSTG on real-world traffic data from Rhode Island and additional large-scale datasets. Experimental results demonstrate that the proposed model achieves stable and competitive forecasting performance across multiple horizons, while significantly improving computational efficiency. Ablation studies further validate the effectiveness of spatial partitioning and dual attention in capturing both local and long-range traffic dynamics. These results suggest that patch-based spatiotemporal modeling provides a scalable and effective framework for traffic forecasting under irregular spatial settings.
Abstract:Traffic forecasting is a fundamental component of intelligent transportation systems, yet remains challenging in real-world settings due to irregular sensor distributions and the high computational cost of modeling large-scale spatiotemporal dependencies. In practical traffic networks, sensors are unevenly distributed across regions, leading to non-uniform spatial structures that limit the effectiveness and scalability of existing graph-based and attention-based models. To address these challenges, we propose PatchSTG, a patch-based spatiotemporal graph Transformer designed for efficient forecasting on irregular sensor networks. The key idea is to introduce a hierarchical spatial representation that partitions sensors into balanced, locality-preserving patches based on geographic information. On top of this structure, a dual attention encoder alternates between intra-patch attention for capturing local interactions and inter-patch attention for modeling global dependencies, reducing computational complexity from quadratic to near-linear scaling. We evaluate PatchSTG on real-world traffic data from Rhode Island and additional large-scale datasets. Experimental results demonstrate that the proposed model achieves stable and competitive forecasting performance across multiple horizons, while significantly improving computational efficiency. Ablation studies further validate the effectiveness of spatial partitioning and dual attention in capturing both local and long-range traffic dynamics. These results suggest that patch-based spatiotemporal modeling provides a scalable and effective framework for traffic forecasting under irregular spatial settings.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2606.09872 [cs.LG]
(or arXiv:2606.09872v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2606.09872 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 301 words
Related coverage
Spatiotemporal Graph Transformer for Traffic Intelligence in Edge Computing
arXiv · Aug 6, 2026
TWIG: A Time-Causal Wavelet Operator for Autoregressive Forecasting on Irregular Graphs
arXiv · Sep 22, 2026
Scene-Conditioned Relation Routing for urban cellular activity forecasting
arXiv · Sep 18, 2026
Vertical Advantage: Transforming Industries with Lakebase and Agentic AI
Databricks · Aug 27, 2026