cs.AINov 6, 2026

SIMBAD: Spatio-Temporal Traffic Forecasting Robust to Aperiodicity

Authors: Daniel Y. Lee, Seungwon Shin, Seunghoon Han, Sungsu Lim, Susik Yoon

Organizations: Korea University · Chungnam National University

Abstract

Accurate traffic prediction is essential for urban planning but remains challenging due to the irregular patterns caused by unexpected events. While Spatio-Temporal Graph Neural Networks (STGNNs) effectively model periodic data, they often struggle with these aperiodic fluctuations. To address this, we propose SIMBAD, which improves generalization irregular patterns via two key mechanisms: 1) adaptive regulation of periodic signals to prioritize recent signals, and 2) dynamic spatial modeling that adjusts influence between connected nodes based on contextual relations. Built upon simple similarity metrics to adapt to aperiodic patterns, SIMBAD outperforms existing methods on real-world benchmark datasets, demonstrating superior performance in predicting unseen and irregular traffic events.

Figures & tables

Appendix figures & tables12 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. PatchSTG: Scalable Spatiotemporal Graph Transformers for Traffic Forecasting on Irregular Sensor Networks

    Jun 2, 2026Jichao Li, Xuanming ShiIntelligent Transportation SystemsSpatiotemporal Forecasting

  2. Efficient Prompt Learning for Traffic Forecasting

    May 8, 2026Qianru Zhang, Xinyi Gao, Alexander Zhou +3OOD GeneralizationPrompt Tuning

  3. ADMFormer: An Adaptive-Decomposition Transformer with Time-Varying Masked Spatial Attention for Traffic Forecasting

    May 25, 2026Ruiwen Gu, Qitai Tan, Yahao Liu +1Attention MechanismsSpatiotemporal Forecasting