mlsanityData to 7 Oct 2026

TimeFilter: Patch-Specific Spatial-Temporal Graph Filtration for Time Series Forecasting

arXiv 2501.13041 ↗First posted 22 Jan 2025cs.LGcited by 60

Not on the map. Why?

Abstract (arXiv)

Time series forecasting methods generally fall into two main categories: Channel Independent (CI) and Channel Dependent (CD) strategies. While CI overlooks important covariate relationships, CD captures all dependencies without distinction, introducing noise and reducing generalization. Recent advances in Channel Clustering (CC) aim to refine dependency modeling by grouping channels with similar characteristics and applying tailored modeling techniques. However, coarse-grained clustering struggles to capture complex, time-varying interactions effectively. To address these challenges, we propose TimeFilter, a GNN-based framework for adaptive and fine-grained dependency modeling. After constructing the graph from the input sequence, TimeFilter refines the learned spatial-temporal dependencies by filtering out irrelevant correlations while preserving the most critical ones in a patch-specific manner. Extensive experiments on 13 real-world datasets from diverse application domains demonstrate the state-of-the-art performance of TimeFilter. The code is available at https://github.com/TROUBADOUR000/TimeFilter.

Key facts. Cited by 60 papers in the map (data through 7 Oct 2026). First posted on arXiv 22 Jan 2025.

Builds on

0 of 0 references.

No references here yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 22; 2026: 38.

60 papers.

Sort:

By area:

  1. 2025A Comprehensive Survey of Deep Learning for Multivariate Time Series Forecasting: A Channel Strategy Perspectivecited by 58
  2. 2025Enhancing Time Series Forecasting through Selective Representation Spaces: A Patch Perspectivecited by 34
  3. 2025Aurora: Towards Universal Generative Multimodal Time Series Forecastingcited by 28
  4. 2025FinTSB: A Comprehensive and Practical Benchmark for Financial Time Series Forecastingcited by 23
  5. 2025Efficient Multivariate Time Series Forecasting via Calibrated Language Models with Privileged Knowledge Distillationcited by 20
  6. 2025Bridging Past and Future: Distribution-Aware Alignment for Time Series Forecastingcited by 19
  7. 2025DistDF: Time-Series Forecasting Needs Joint-Distribution Wasserstein Alignmentcited by 13
  8. 2026GCGNet: Graph-Consistent Generative Network for Time Series Forecasting with Exogenous Variablescited by 11
  9. 2025Glocal Information Bottleneck for Time Series Imputationcited by 9
  10. 2025K2VAE: A Koopman-Kalman Enhanced Variational AutoEncoder for Probabilistic Time Series Forecastingcited by 7
  11. 2025AutoHFormer: Efficient Hierarchical Autoregressive Transformer for Time Series Predictioncited by 7
  12. 2025Revisiting Multivariate Time Series Forecasting with Missing Valuescited by 7
38 more
  1. 2026From Observations to States: Latent Time Series Forecastingcited by 6
  2. 2025RDIT: Residual-based Diffusion Implicit Models for Probabilistic Time Series Forecastingcited by 4
  3. 2025FLAME: Flow Enhanced Legendre Memory Models for General Time Series Forecastingcited by 3
  4. 2026Rethinking the Role of LLMs in Time Series Forecastingcited by 3
  5. 2026Into the ORBIT for Time Series: Training Regimes for Foundation Modelscited by 3
  6. 2026PHAT: Modeling Period Heterogeneity for Multivariate Time Series Forecastingcited by 2
  7. 2026TEFL: Prediction-Residual-Guided Rolling Forecasting for Multi-Horizon Time Seriescited by 2
  8. 2026Routing Channel-Patch Dependencies in Time Series Forecasting with Graph Spectral Decompositioncited by 2
  9. 2025Towards Robust Real-World Multivariate Time Series Forecasting: A Unified Framework for Dependency, Asynchrony, and Missingnesscited by 1
  10. 2025Time Series Forecasting Through the Lens of Dynamicscited by 1
  11. 2025HGTS-Former: Hierarchical HyperGraph Transformer for Multivariate Time Series Analysiscited by 1
  12. 2025IndexNet: Timestamp and Variable-Aware Modeling for Time Series Forecastingcited by 1
  13. 2026Forecasting as Rendering: A 2D Gaussian Splatting Framework for Time Series Forecastingcited by 1
  14. 2026Forecasting with Guidance: Representation-Level Supervision for Time Series Forecastingcited by 1
  15. 2026DSPR: Dual-Stream Physics-Residual Networks for Trustworthy Industrial Time Series Forecastingcited by 1
  16. 2026What If We Let Forecasting Forget? A Sparse Bottleneck for Cross-Variable Dependenciescited by 1
  17. 2026One Step Closer to Ground Truth: A Multi-Scale Residual-Aware Representation Learning Pipeline for Predicting Time Series Datacited by 1
  18. 2026Existence Precedes Value: Joint Modeling of Observational Existence and Evolving States in Time Series Forecastingcited by 1
  19. 2026Multivariate Time Series Forecasting needs Cross Variable Losscited by 1
  20. 2026StationPDE: Station-Oriented Surface PDE Learning for Multi-Station Multivariate Weather Forecastingcited by 1
  21. 2025Dynamic Perturbed Adaptive Method for Infinite Task-Conflicting Time Seriescited by 0
  22. 2025Integrating Time Series into LLMs via Multi-layer Steerable Embedding Fusion for Enhanced Forecastingcited by 0
  23. 2025Dynamic Adaptive Parsing of Temporal and Cross-Variable Patterns for Network State Classificationcited by 0
  24. 2025KAIROS: Unified Training for Universal Non-Autoregressive Time Series Forecastingcited by 0
  25. 2025DecoKAN: Interpretable Decomposition for Forecasting Cryptocurrency Market Dynamicscited by 0
  26. 2026NavFormer: IGRF Forecasting in Moving Coordinate Framescited by 0
  27. 2026A Scalable Multi-Task Model for Virtual Sensorscited by 0
  28. 2026AltTS: A Dual-Path Framework with Alternating Optimization for Multivariate Time Series Forecastingcited by 0
  29. 2026MICA: Multivariate Infini Compressive Attention for Time Series Forecastingcited by 0
  30. 2026CombinationTS: A Modular Framework for Understanding Time-Series Forecasting Modelscited by 0
  31. 2026Perceive, Route and Modulate: Dynamic Pattern Recalibration for Time Series Forecastingcited by 0
  32. 2026PeakFocus: Bridging Peak Localization and Intensity Regression via a Unified Multi-Scale Framework for Electricity Load Forecastingcited by 0
  33. 2026Generalizing Multi-Scale Time-Series Modeling with a Single Operatorcited by 0
  34. 2026VLBM: Variational Latent Basis Modeling for OOD Robust Multivariate Time Series Forecastingcited by 0
  35. 2026Adaptive Patching Is Harder Than It Looks For Time-Series Forecastingcited by 0
  36. 2026GLAIM: Learning Global and Local Adaptive Inter-Variable Dependency for Multivariate Time Series Imputationcited by 0
  37. 2026SafeDivertor: Faithful Divertor Heat Flux Reconstruction from Macroscopic Plasma State Signals via Time-Frequency Prior Exploitationcited by 0
  38. 2026Role-Specific Predictive Geometries for Nonstationary Multivariate Graph-Signal Forecastingcited by 0

50 most cited of 60 All 60 (JSON)