GAS-PFN: Graph-Aware Spatiotemporal Prior-Data Fitted Networks for Zero-Shot Traffic Forecasting

Joseph Natter, Long Vu, Yifan Zhang, Bobin Deng, Kun Suo et al.|October 27, 2026arXiv

Key Takeaway

By explicitly encoding road network structure into the training process and using a two-stage inference design, you can build traffic forecasters that work on new networks without retraining—a major practical advantage for deploying to evolving city infrastructure.

Summary

This paper introduces GAS-PFN, a neural network that forecasts traffic flow without needing to be retrained for new road networks. It learns from synthetic traffic patterns during training, using graph structure (road topology) to understand how traffic moves between connected sensors.

reasoning architecture applications

Key Terms

zero-shot-learning prior-data-fitted-networks random-walk-diffusion mask-conditioned-inference spatiotemporal-modeling