mlsanityData to 13 Oct 2026

Deciphering Spatio-Temporal Graph Forecasting: A Causal Lens and Treatment

arXiv 2309.13378 ↗First posted 23 Sep 2023cs.LG cs.AIcited by 47

Not on the map. Why?

Abstract (arXiv)

Spatio-Temporal Graph (STG) forecasting is a fundamental task in many real-world applications. Spatio-Temporal Graph Neural Networks have emerged as the most popular method for STG forecasting, but they often struggle with temporal out-of-distribution (OoD) issues and dynamic spatial causation. In this paper, we propose a novel framework called CaST to tackle these two challenges via causal treatments. Concretely, leveraging a causal lens, we first build a structural causal model to decipher the data generation process of STGs. To handle the temporal OoD issue, we employ the back-door adjustment by a novel disentanglement block to separate invariant parts and temporal environments from input data. Moreover, we utilize the front-door adjustment and adopt the Hodge-Laplacian operator for edge-level convolution to model the ripple effect of causation. Experiments results on three real-world datasets demonstrate the effectiveness and practicality of CaST, which consistently outperforms existing methods with good interpretability.

Key facts. Cited by 47 papers in the map (data through 13 Oct 2026). Builds on 36 of its 68 references that are in the map. First posted on arXiv 23 Sep 2023.

Builds on

36 of 68 references.

References · 36

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,093
  2. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,318
  3. 2017Graph Attention Networkscited by 8,888
  4. 2017Neural Discrete Representation Learningcited by 4,680
  5. 2018An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modelingcited by 1,460
  6. 2017Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecastingcited by 1,130
  7. 2019Graph WaveNet for Deep Spatial-Temporal Graph Modelingcited by 825
  8. 2018MINE: Mutual Information Neural Estimationcited by 682
  9. 2020Connecting the Dots: Multivariate Time Series Forecasting with Graph Neural Networkscited by 564
  10. 2013Generating Sequences With Recurrent Neural Networkscited by 553
  11. 2020Adaptive Graph Convolutional Recurrent Network for Traffic Forecastingcited by 527
  12. 2019GMAN: A Graph Multi-Attention Network for Traffic Predictioncited by 434
24 more
  1. 2016Deep Spatio-Temporal Residual Networks for Citywide Crowd Flows Predictioncited by 333
  2. 2020Causal Intervention for Weakly-Supervised Semantic Segmentationcited by 229
  3. 2021Spatial-Temporal Graph ODE Networks for Traffic Flow Forecastingcited by 210
  4. 2019Deep Learning for Spatio-Temporal Data Mining: A Surveycited by 172
  5. 2022CoST: Contrastive Learning of Disentangled Seasonal-Trend Representations for Time Series Forecastingcited by 144
  6. 2023Spatio-Temporal Graph Neural Networks for Predictive Learning in Urban Computing: A Surveycited by 140
  7. 2021Graph Neural Controlled Differential Equations for Traffic Forecastingcited by 133
  8. 2018Revisiting Spatial-Temporal Similarity: A Deep Learning Framework for Traffic Predictioncited by 120
  9. 2021AdaRNN: Adaptive Learning and Forecasting of Time Seriescited by 120
  10. 2021Causal Attention for Interpretable and Generalizable Graph Classificationcited by 108
  11. 2020Visual Commonsense R-CNNcited by 106
  12. 2018Random Walks on Simplicial Complexes and the normalized Hodge 1-Laplaciancited by 84
  13. 2023DiffSTG: Probabilistic Spatio-Temporal Graph Forecasting with Denoising Diffusion Modelscited by 81
  14. 2021OOD-GNN: Out-of-Distribution Generalized Graph Neural Networkcited by 78
  15. 2021DL-Traff: Survey and Benchmark of Deep Learning Models for Urban Traffic Predictioncited by 73
  16. 2022AirFormer: Predicting Nationwide Air Quality in China with Transformerscited by 73
  17. 2022STDEN: Towards Physics-Guided Neural Networks for Traffic Flow Predictioncited by 45
  18. 2022OOD Link Prediction Generalization Capabilities of Message-Passing GNNs in Larger Test Graphscited by 38
  19. 2018Flow Smoothing and Denoising: Graph Signal Processing in the Edge-Spacecited by 37
  20. 2021Edge Representation Learning with Hypergraphscited by 37
  21. 2023Do We Really Need Graph Neural Networks for Traffic Forecasting?cited by 20
  22. 2022Towards Out-of-Distribution Sequential Event Prediction: A Causal Treatmentcited by 10
  23. 2021Helmholtzian Eigenmap: Topological feature discovery & edge flow learning from point cloud datacited by 7
  24. 2023Heterogeneous Graph Convolutional Neural Network via Hodge-Laplacian for Brain Functional Datacited by 6

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2023: 2; 2024: 22; 2025: 16; 2026: 7.

47 papers.

Sort:

By area:

  1. 2024Foundation Models for Time Series Analysis: A Tutorial and Surveycited by 237
  2. 2023Large Models for Time Series and Spatio-Temporal Data: A Survey and Outlookcited by 109
  3. 2024UniST: A Prompt-Empowered Universal Model for Urban Spatio-Temporal Predictioncited by 86
  4. 2024Deep Learning for Cross-Domain Data Fusion in Urban Computing: Taxonomy, Advances, and Outlookcited by 33
  5. 2024MolTC: Towards Molecular Relational Modeling In Language Modelscited by 22
  6. 2024Causal Evaluation of Language Modelscited by 18
  7. 2024Heterogeneity-Informed Meta-Parameter Learning for Spatiotemporal Time Series Forecastingcited by 18
  8. 2024Spatio-Temporal Few-Shot Learning via Diffusive Neural Network Generationcited by 15
  9. 2025Unified Spatial-Temporal Edge-Enhanced Graph Networks for Pedestrian Trajectory Predictioncited by 13
  10. 2025Learning with Calibration: Exploring Test-Time Computing of Spatio-Temporal Forecastingcited by 9
  11. 2023Exploring Causal Learning through Graph Neural Networks: An In-depth Reviewcited by 8
  12. 2024CaT-GNN: Enhancing Credit Card Fraud Detection via Causal Temporal Graph Neural Networkscited by 7
35 more
  1. 2024Evaluating the Generalization Ability of Spatiotemporal Model in Urban Scenariocited by 7
  2. 2025Causal Time Series Generation via Diffusion Modelscited by 7
  3. 2024Spatio-Temporal Fluid Dynamics Modeling via Physical-Awareness and Parameter Diffusion Guidancecited by 5
  4. 2025Invariant Link Selector for Spatial-Temporal Out-of-Distribution Problemcited by 5
  5. 2024Towards Robust Trajectory Representations: Isolating Environmental Confounders with Causal Learningcited by 4
  6. 2024DualCast: A Model to Disentangle Aperiodic Events from Traffic Seriescited by 4
  7. 2025FineCausal: A Causal-Based Framework for Interpretable Fine-Grained Action Quality Assessmentcited by 4
  8. 2025Causal Spatio-Temporal Prediction: An Effective and Efficient Multi-Modal Approachcited by 4
  9. 2025Reimagining Urban Science: Scaling Causal Inference with Large Language Modelscited by 3
  10. 2025Spatiotemporal Causal Decoupling Model for Air Quality Forecastingcited by 3
  11. 2025AquaSentinel: Next-Generation AI System Integrating Sensor Networks for Urban Underground Water Pipeline Anomaly Detection via Collaborative MoE-LLM Agent Architecturecited by 3
  12. 2026Policy4OOD: A Knowledge-Guided World Model for Policy Intervention Simulation against the Opioid Overdose Crisiscited by 3
  13. 2025AirRadar: Inferring Nationwide Air Quality in China with Deep Neural Networkscited by 2
  14. 2024Predicting Parking Availability in Singapore with Cross-Domain Data: A New Dataset and A Data-Driven Approachcited by 1
  15. 2024Robust Traffic Forecasting against Spatial Shift over Yearscited by 1
  16. 2024Causal Adjacency Learning for Spatiotemporal Prediction Over Graphscited by 1
  17. 2024Spatio-temporal Causal Learning for Streamflow Forecastingcited by 1
  18. 2024STDCformer: A Transformer-Based Model with a Spatial-Temporal Causal De-Confounding Strategy for Crowd Flow Predictioncited by 1
  19. 2024Memory-enhanced Invariant Prompt Learning for Urban Flow Prediction under Distribution Shiftscited by 1
  20. 2025SynEVO: A neuro-inspired spatiotemporal evolutional framework for cross-domain adaptationcited by 1
  21. 2025CaPulse: Detecting Anomalies by Tuning in to the Causal Rhythms of Time Seriescited by 1
  22. 2025AI-Driven Forecasting and Monitoring of Urban Water Systemcited by 1
  23. 2026A General ReLearner: Empowering Spatiotemporal Prediction by Re-learning Input-label Residualcited by 1
  24. 2026A Generalizable Physics-guided Causal Model for Trajectory Prediction in Autonomous Drivingcited by 1
  25. 2024Towards Causal Relationship in Indefinite Data: Baseline Model and New Datasetscited by 0
  26. 2024ComS2T: A complementary spatiotemporal learning system for data-adaptive model evolutioncited by 0
  27. 2024Incorporating Long-term Data in Training Short-term Traffic Prediction Modelcited by 0
  28. 2024Unveiling the Inflexibility of Adaptive Embedding in Traffic Forecastingcited by 0
  29. 2025Soft causal learning for generalized molecule property prediction: An environment perspectivecited by 0
  30. 2025Solar Forecasting with Causality: A Graph-Transformer Approach to Spatiotemporal Dependenciescited by 0
  31. 2025Towards Resilient Transportation: A Conditional Transformer for Accident-Informed Traffic Forecastingcited by 0
  32. 2026Efficient Prompt Learning for Traffic Forecastingcited by 0
  33. 2026Every Client Is an Environment: Federated De-confounding for Spatio-Temporal Forecastingcited by 0
  34. 2026When Correlations Mislead: Confounder-Aware Multi-View Urban Region Representation Learningcited by 0
  35. 2026AdaST: Adaptive Coupling for Spatial-Temporal Forecastingcited by 0