mlsanityData to 14 Oct 2026

Dynamic TMoE: A Drift-Aware Dynamic Mixture of Experts Framework for Non-Stationary Time Series Forecasting

arXiv 2605.20678 ↗First posted 20 May 2026cs.LG cs.AIcited by 0

On the map: Mixture-of-experts → MoE for time series and graphs

Abstract (arXiv)

Non-stationary time series forecasting is challenged by evolving distribution shifts that static models struggle to capture. While Mixture-of-Experts (MoE) architectures offer a promising paradigm for decoupling complex drift patterns, existing approaches are limited by fixed expert pools and memoryless routing, hampering their ability to adapt to abrupt regime shifts. To address this, we propose Dynamic TMoE, a framework that unifies architectural evolution with temporal continuity during learning phase. By detecting distribution shifts via Maximum Mean Discrepancy (MMD), we dynamically instantiate heterogeneous experts and prune redundant ones to optimize capacity. Additionally, a temporal memory router leverages recurrent states and an anomaly repository to ensure stable, context-aware expert selection without requiring test-time updates. Experiments on nine benchmarks demonstrate state-of-the-art performance, reducing MSE by 10.4% and MAE by 7.8%. Code is available at https://github.com/andone-07/Dynamic-TMoE.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in MoE for time series and graphs, one of 12 directions in Mixture-of-experts. Builds on 30 of its 42 references that are in the map. First posted on arXiv 20 May 2026.

Builds on

30 of 42 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025Wavelet Mixture of Experts for Time Series Forecasting12 Aug
  2. 2025FreqMoE: Enhancing Time Series Forecasting through Frequency Decomposition Mixture of Experts25 Jan
  3. 2024Time-MoE: Billion-Scale Time Series Foundation Models with Mixture of Experts24 Sep

Other references · 27

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  2. 2020Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecastingcited by 2,169
  3. 2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 2,039
  4. 2021Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecastingcited by 1,676
  5. 2022Are Transformers Effective for Time Series Forecasting?cited by 1,424
  6. 2022FEDformer: Frequency Enhanced Decomposed Transformer for Long-term Series Forecastingcited by 1,210
  7. 2022TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysiscited by 1,058
  8. 2023iTransformer: Inverted Transformers Are Effective for Time Series Forecastingcited by 931
  9. 2017Modeling Long- and Short-Term Temporal Patterns with Deep Neural Networkscited by 879
  10. 2022Non-stationary Transformers: Exploring the Stationarity in Time Series Forecastingcited by 391
  11. 2024TimeMixer: Decomposable Multiscale Mixing for Time Series Forecastingcited by 346
  12. 2023FITS: Modeling Time Series with 10k Parameterscited by 195
15 more
  1. 2021AdaRNN: Adaptive Learning and Forecasting of Time Seriescited by 121
  2. 2023Koopa: Learning Non-stationary Time Series Dynamics with Koopman Predictorscited by 113
  3. 2024CycleNet: Enhancing Time Series Forecasting through Modeling Periodic Patternscited by 106
  4. 2023Dish-TS: A General Paradigm for Alleviating Distribution Shift in Time Series Forecastingcited by 93
  5. 2024TSLANet: Rethinking Transformers for Time Series Representation Learningcited by 74
  6. 2024Frequency Adaptive Normalization For Non-stationary Time Series Forecastingcited by 47
  7. 2024Retrieval Augmented Time Series Forecastingcited by 46
  8. 2024TimeBridge: Non-Stationarity Matters for Long-term Time Series Forecastingcited by 40
  9. 2023Mixture-of-Linear-Experts for Long-term Time Series Forecastingcited by 29
  10. 2024Learning Pattern-Specific Experts for Time Series Forecasting Under Patch-level Distribution Shiftcited by 24
  11. 2024Deep Frequency Derivative Learning for Non-stationary Time Series Forecastingcited by 13
  12. 2024IN-Flow: Instance Normalization Flow for Non-stationary Time Series Forecastingcited by 7
  13. 2024Peri-midFormer: Periodic Pyramid Transformer for Time Series Analysiscited by 5
  14. 2025TimeExpert: Boosting Long Time Series Forecasting with Temporal Mix of Expertscited by 4
  15. 2025ST-MTM: Masked Time Series Modeling with Seasonal-Trend Decomposition for Time Series Forecastingcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026AME-TS: Anchored Mixture-of-Experts for Time Series Forecasting
  2. 2025DriftMoE: A Mixture of Experts Approach to Handle Concept Drifts
  3. 2026WaveMoE: A Wavelet-Enhanced Mixture-of-Experts Foundation Model for Time Series Forecasting
  4. 2026RAVEN: A Regime-Aware Variable-context Expert Network for Financial Time Series Forecasting
  5. 2026NEST: Tackling Dataset-Level Distribution Shifts via Regime-Oriented Mixture-of-Experts
  6. 2026MemCast: Memory-Driven Time Series Forecasting with Experience-Conditioned Reasoning
  7. 2026DynG-Diff: A State-Aware Dynamic Guidance Diffusion Framework for Probabilistic Time Series Forecasting
  8. 2026FAME: Forecastability-Aware Mixture of Experts for Heterogeneous Time Series Forecasting