mlsanityData to 7 Oct 2026

Learning with Calibration: Exploring Test-Time Computing of Spatio-Temporal Forecasting

arXiv 2506.00635 ↗First posted 31 May 2025cs.LG cs.AI cs.ETcited by 9

Not on the map. Why?

Abstract (arXiv)

Spatio-temporal forecasting is crucial in many domains, such as transportation, meteorology, and energy. However, real-world scenarios frequently present challenges such as signal anomalies, noise, and distributional shifts. Existing solutions primarily enhance robustness by modifying network architectures or training procedures. Nevertheless, these approaches are computationally intensive and resource-demanding, especially for large-scale applications. In this paper, we explore a novel test-time computing paradigm, namely learning with calibration, ST-TTC, for spatio-temporal forecasting. Through learning with calibration, we aim to capture periodic structural biases arising from non-stationarity during the testing phase and perform real-time bias correction on predictions to improve accuracy. Specifically, we first introduce a spectral-domain calibrator with phase-amplitude modulation to mitigate periodic shift and then propose a flash updating mechanism with a streaming memory queue for efficient test-time computation. ST-TTC effectively bypasses complex training-stage techniques, offering an efficient and generalizable paradigm. Extensive experiments on real-world datasets demonstrate the effectiveness, universality, flexibility and efficiency of our proposed method.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Builds on 59 of its 104 references that are in the map. First posted on arXiv 31 May 2025.

Builds on

59 of 104 references.

References · 59

  1. 2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 1,981
  2. 2022Are Transformers Effective for Time Series Forecasting?cited by 1,400
  3. 2017Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecastingcited by 1,125
  4. 2025Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?cited by 948
  5. 2020Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptationcited by 844
  6. 2019Graph WaveNet for Deep Spatial-Temporal Graph Modelingcited by 820
  7. 2022Continual Test-Time Domain Adaptationcited by 520
  8. 2023A Comprehensive Survey on Test-Time Adaptation under Distribution Shiftscited by 353
  9. 2023ClimaX: A foundation model for weather and climatecited by 325
  10. 2022Test-Time Training with Masked Autoencoderscited by 208
  11. 2023FITS: Modeling Time Series with 10k Parameterscited by 185
  12. 2020Spatial-Temporal Transformer Networks for Traffic Flow Forecastingcited by 179
38 more
  1. 2022Spatial-Temporal Identity: A Simple yet Effective Baseline for Multivariate Time Series Forecastingcited by 149
  2. 2023A Survey on Graph Neural Networks for Time Series: Forecasting, Classification, Imputation, and Anomaly Detectioncited by 145
  3. 2023Spatio-Temporal Graph Neural Networks for Predictive Learning in Urban Computing: A Surveycited by 140
  4. 2023FourierGNN: Rethinking Multivariate Time Series Forecasting from a Pure Graph Perspectivecited by 126
  5. 2013Learning by Transductioncited by 120
  6. 2023LargeST: A Benchmark Dataset for Large-Scale Traffic Forecastingcited by 116
  7. 2023Exploring Progress in Multivariate Time Series Forecasting: Comprehensive Benchmarking and Heterogeneity Analysiscited by 113
  8. 2024The Surprising Effectiveness of Test-Time Training for Few-Shot Learningcited by 88
  9. 2024UniST: A Prompt-Empowered Universal Model for Urban Spatio-Temporal Predictioncited by 86
  10. 2023Test-Time Training on Nearest Neighbors for Large Language Modelscited by 82
  11. 2024UrbanGPT: Spatio-Temporal Large Language Modelscited by 81
  12. 2023OneNet: Enhancing Time Series Forecasting Models under Concept Drift by Online Ensemblingcited by 71
  13. 2022Test-Time Adaptation via Self-Training with Nearest Neighbor Informationcited by 64
  14. 2020PM2.5-GNN: A Domain Knowledge Enhanced Graph Neural Network For PM2.5 Forecastingcited by 52
  15. 2023Deciphering Spatio-Temporal Graph Forecasting: A Causal Lens and Treatmentcited by 47
  16. 2023PastNet: Introducing Physical Inductive Biases for Spatio-temporal Video Predictioncited by 44
  17. 2022Learning Fast and Slow for Online Time Series Forecastingcited by 43
  18. 2021TrafficStream: A Streaming Traffic Flow Forecasting Framework Based on Graph Neural Networks and Continual Learningcited by 42
  19. 2023Taming Local Effects in Graph-based Spatiotemporal Forecastingcited by 37
  20. 2023Test-Time Training on Video Streamscited by 37
  21. 2020FC-GAGA: Fully Connected Gated Graph Architecture for Spatio-Temporal Traffic Forecastingcited by 30
  22. 2022Practical Adversarial Attacks on Spatiotemporal Traffic Forecasting Modelscited by 28
  23. 2024Wavelet Diffusion Neural Operatorcited by 28
  24. 2024OpenCity: Open Spatio-Temporal Foundation Models for Traffic Predictioncited by 27
  25. 2025TimeKAN: KAN-based Frequency Decomposition Learning Architecture for Long-term Time Series Forecastingcited by 27
  26. 2023Automated Spatio-Temporal Graph Contrastive Learningcited by 24
  27. 2024Efficient Large-Scale Traffic Forecasting with Transformers: A Spatial Data Management Perspectivecited by 24
  28. 2018Machine Learning for Spatiotemporal Sequence Forecasting: A Surveycited by 19
  29. 2023Disentangling Structured Components: Towards Adaptive, Interpretable and Scalable Time Series Forecastingcited by 19
  30. 2024FlashST: A Simple and Universal Prompt-Tuning Framework for Traffic Predictioncited by 18
  31. 2024BIGCity: A Universal Spatiotemporal Model for Unified Trajectory and Traffic State Data Analysiscited by 16
  32. 2022Long-term Spatio-temporal Forecasting via Dynamic Multiple-Graph Attentioncited by 15
  33. 2024Spatio-Temporal Few-Shot Learning via Diffusive Neural Network Generationcited by 14
  34. 2023Spatial-Temporal Graph Learning with Adversarial Contrastive Adaptationcited by 13
  35. 2023Seeing the Unseen: Learning Basis Confounder Representations for Robust Traffic Predictioncited by 12
  36. 2024Online Test-Time Adaptation of Spatial-Temporal Traffic Flow Forecastingcited by 11
  37. 2024Expand and Compress: Exploring Tuning Principles for Continual Spatio-Temporal Graph Forecastingcited by 10
  38. 2024TEAM: Topological Evolution-aware Framework for Traffic Forecasting--Extended Versioncited by 10

50 most cited of 59 All 59 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 1; 2026: 8.

9 papers.

  1. 2026STReasoner: Empowering LLMs for Spatio-Temporal Reasoning in Time Series via Spatial-Aware Reinforcement Learningcited by 7
  2. 2026Learning to Factorize and Adapt: A Versatile Approach Toward Universal Spatio-Temporal Foundation Modelscited by 2
  3. 2026Test-Time Learning of Causal Structure from Interventional Datacited by 2
  4. 2026UrbanFM: Scaling Urban Spatio-Temporal Foundation Modelscited by 1
  5. 2026From XXLTraffic to EvoXXLTraffic: Scaling Traffic Forecasting to Sensor-Evolving Networkscited by 1
  6. 2025Select, then Balance: Exploring Exogenous Variable Modeling of Spatio-Temporal Forecastingcited by 0
  7. 2026Learning from Complexity: Exploring Dynamic Sample Pruning of Spatio-Temporal Trainingcited by 0
  8. 2026A2TTA: Anchored-and-Agile Test-Time Adaptation for Evolving Traffic Sensor Networkscited by 0
  9. 2026More Sensors Only One Field: Rethinking Continual Spatio-Temporal Forecastingcited by 0