A Pretrained Probabilistic Transformer for City-Scale Traffic Volume Prediction
This addresses the problem of uncertainty and generalizability in traffic prediction for intelligent transportation systems, though it is incremental as it builds on existing transformer and probabilistic methods.
The paper tackles city-scale traffic volume prediction by introducing TrafficPPT, a pretrained probabilistic transformer that models traffic as a distributional aggregation of trajectories, and it consistently surpasses state-of-the-art baselines, especially under extreme data sparsity conditions.
City-scale traffic volume prediction plays a pivotal role in intelligent transportation systems, yet remains a challenge due to the inherent incompleteness and bias in observational data. Although deep learning-based methods have shown considerable promise, most existing approaches produce deterministic point estimates, thereby neglecting the uncertainty arising from unobserved traffic flows. Furthermore, current models are typically trained in a city-specific manner, which hinders their generalizability and limits scalability across diverse urban contexts. To overcome these limitations, we introduce TrafficPPT, a Pretrained Probabilistic Transformer designed to model traffic volume as a distributional aggregation of trajectories. Our framework fuses heterogeneous data sources-including real-time observations, historical trajectory data, and road network topology-enabling robust and uncertainty-aware traffic inference. TrafficPPT is initially pretrained on large-scale simulated data spanning multiple urban scenarios, and later fine-tuned on target cities to ensure effective domain adaptation. Experiments on real-world datasets show that TrafficPPT consistently surpasses state-of-the-art baselines, particularly under conditions of extreme data sparsity. Code will be open.