mlsanityData to 7 Oct 2026

There are no Champions in Supervised Long-Term Time Series Forecasting

arXiv 2502.14045 ↗First posted 19 Feb 2025cs.LG cs.AIcited by 8

Not on the map. Why?

Abstract (arXiv)

Recent advances in long-term time series forecasting have introduced numerous complex supervised prediction models that consistently outperform previously published architectures. However, this rapid progression raises concerns regarding inconsistent benchmarking and reporting practices, which may undermine the reliability of these comparisons. In this study, we first perform a broad, thorough, and reproducible evaluation of the top-performing supervised models on the most popular benchmark and additional baselines representing the most active architecture families. This extensive evaluation assesses eight models on 14 datasets, encompassing ~5,000 trained networks for the hyperparameter (HP) searches. Then, through a comprehensive analysis, we find that slight changes to experimental setups or current evaluation metrics drastically shift the common belief that newly published results are advancing the state of the art. Our findings emphasize the need to shift focus away from pursuing ever-more complex models, towards enhancing benchmarking practices through rigorous and standardized evaluations that enable more substantiated claims, including reproducible HP setups and statistical testing. We offer recommendations for future research.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 50 of its 78 references that are in the map. First posted on arXiv 19 Feb 2025.

Builds on

50 of 78 references.

References · 50

  1. 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,120
  2. 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,616
  3. 2019Optuna: A Next-generation Hyperparameter Optimization Frameworkcited by 2,455
  4. 2020Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecastingcited by 2,141
  5. 2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 1,981
  6. 2021Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecastingcited by 1,654
  7. 2022Are Transformers Effective for Time Series Forecasting?cited by 1,400
  8. 2022FEDformer: Frequency Enhanced Decomposed Transformer for Long-term Series Forecastingcited by 1,200
  9. 2022TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysiscited by 1,030
  10. 2018MnasNet: Platform-Aware Neural Architecture Search for Mobilecited by 937
  11. 2023iTransformer: Inverted Transformers Are Effective for Time Series Forecastingcited by 906
  12. 2019N-BEATS: Neural basis expansion analysis for interpretable time series forecastingcited by 744
38 more
  1. 2024Chronos: Learning the Language of Time Seriescited by 712
  2. 2019Enhancing the Locality and Breaking the Memory Bottleneck of Transformer on Time Series Forecastingcited by 606
  3. 2024Unified Training of Universal Time Series Forecasting Transformerscited by 588
  4. 2023Large Language Models Are Zero-Shot Time Series Forecasterscited by 491
  5. 2022Non-stationary Transformers: Exploring the Stationarity in Time Series Forecastingcited by 385
  6. 2021SCINet: Time Series Modeling and Forecasting with Sample Convolution and Interactioncited by 335
  7. 2023Long-term Forecasting with TiDE: Time-series Dense Encodercited by 327
  8. 2024xLSTM: Extended Long Short-Term Memorycited by 320
  9. 2019Financial Time Series Forecasting with Deep Learning : A Systematic Literature Review: 2005-2019cited by 270
  10. 2024TFB: Towards Comprehensive and Fair Benchmarking of Time Series Forecasting Methodscited by 242
  11. 2024TimeXer: Empowering Transformers for Time Series Forecasting with Exogenous Variablescited by 241
  12. 2023When Do Neural Nets Outperform Boosted Trees on Tabular Data?cited by 218
  13. 2024Timer: Generative Pre-trained Transformers Are Large Time Series Modelscited by 198
  14. 2024Deep Time Series Models: A Comprehensive Survey and Benchmarkcited by 191
  15. 2024GIFT-Eval: A Benchmark For General Time Series Forecasting Model Evaluationcited by 188
  16. 2022PromptCast: A New Prompt-based Learning Paradigm for Time Series Forecastingcited by 170
  17. 2023TEMPO: Prompt-based Generative Pre-trained Transformer for Time Series Forecastingcited by 170
  18. 2023TEST: Text Prototype Aligned Embedding to Activate LLM's Ability for Time Seriescited by 165
  19. 2024Are Language Models Actually Useful for Time Series Forecasting?cited by 163
  20. 2023TimeGPT-1cited by 149
  21. 2020Current Time Series Anomaly Detection Benchmarks are Flawed and are Creating the Illusion of Progresscited by 148
  22. 2023LLM4TS: Aligning Pre-Trained LLMs as Data-Efficient Time-Series Forecasterscited by 127
  23. 2024Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-Shot Forecasting of Multivariate Time Seriescited by 120
  24. 2023Exploring Progress in Multivariate Time Series Forecasting: Comprehensive Benchmarking and Heterogeneity Analysiscited by 113
  25. 2024Is Mamba Effective for Time Series Forecasting?cited by 110
  26. 2024Inadequacies of Large Language Model Benchmarks in the Era of Generative Artificial Intelligencecited by 65
  27. 2025Can We Trust AI Benchmarks? An Interdisciplinary Review of Current Issues in AI Evaluationcited by 57
  28. 2022Forecast Evaluation for Data Scientists: Common Pitfalls and Best Practicescited by 54
  29. 2024Position: What Can Large Language Models Tell Us about Time Series Analysiscited by 47
  30. 2024Scaling Law for Time Series Forecastingcited by 31
  31. 2024Towards Neural Scaling Laws for Time Series Foundation Modelscited by 30
  32. 2024Specialized Foundation Models Struggle to Beat Supervised Baselinescited by 23
  33. 2024Position: Why We Must Rethink Empirical Research in Machine Learningcited by 19
  34. 2024xLSTM-Mixer: Multivariate Time Series Forecasting by Mixing via Scalar Memoriescited by 18
  35. 2024Position: Quo Vadis, Unsupervised Time Series Anomaly Detection?cited by 17
  36. 2024Position: Benchmarking is Limited in Reinforcement Learning Researchcited by 15
  37. 2024xLSTMTime : Long-term Time Series Forecasting With xLSTMcited by 13
  38. 2021Tune It or Don't Use It: Benchmarking Data-Efficient Image Classificationcited by 11

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 4; 2026: 4.

8 papers.

  1. 2026It's TIME: Towards the Next Generation of Time Series Forecasting Benchmarkscited by 25
  2. 2025Estimating Time Series Foundation Model Transferability via In-Context Learningcited by 2
  3. 2025ARIES: Relation Assessment and Model Recommendation for Deep Time Series Forecastingcited by 1
  4. 2026Superposition Is Not Necessary: A Mechanistic Interpretability Analysis of Transformer Representations for Time Series Forecastingcited by 1
  5. 2025PA-RNet: Perturbation-Aware Residual Network for Robust Multimodal Time Series Forecastingcited by 0
  6. 2025VFEM: Visual Feature Empowered Multivariate Time Series Forecasting with Cross-Modal Fusioncited by 0
  7. 2026The Spectrum Is Not Enough: When Context Helps Time-Series Forecastingcited by 0
  8. 2026MixBench-TS: A Multivariate Time Series Forecasting Benchmark Where Channel Mixing Pays Offcited by 0