mlsanityData to 14 Oct 2026

MixBench-TS: A Multivariate Time Series Forecasting Benchmark Where Channel Mixing Pays Off

arXiv 2609.32656 ↗First posted 26 Sep 2026cs.AI cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Multivariate Time Series Forecasting (MTSF) models that mix information across channels assume that the past of one channel carries information about the future of another. Yet they are evaluated on a small fixed set of standard datasets whose cross-channel structure is rarely examined. We ask two questions: "How can we reliably measure lagged, non-linear, and joint coupling in MTSF datasets?" and "Do the standard datasets actually have such coupling?" To answer the first, we test four candidate measures on synthetic datasets with planted ground-truth coupling: Granger Causality (GC), Transfer Entropy (TE), lagged Mutual Information (MI), and the CD gain, a model-based measure we introduce that compares a channel-dependent (CD) model to its channel-independent (CI) variant. Only lagged MI and the CD gain recover every planted coupling. For the second question, the answer is a definite no, as the standard datasets have a median of only 23% lagged-coupled channel pairs and a median CD gain of -4.9%, compared to 78% and +4.6% on chaotic ODE systems. We therefore propose MixBench-TS, a benchmark of 10 real-world datasets with a median of 55.5% lagged-coupled pairs and a median CD gain of +1.7%. Across six state-of-the-art models tuned under one protocol, CI models win on 10/10 (MSE) and 8/10 (MAE) standard datasets, but on only 3/10 and 2/10 MixBench-TS datasets. We recommend using our benchmark for evaluating new CD models. Moreover, we propose profiling new datasets with lagged MI and the CD gain before using them to evaluate multivariate models. Code and data are available at https://anonymous.4open.science/r/mixbench-ts-B027.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 30 of its 50 references that are in the map. First posted on arXiv 26 Sep 2026.

Builds on

30 of 50 references.

References · 30

  1. 2019Optuna: A Next-generation Hyperparameter Optimization Frameworkcited by 2,501
  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. 2023iTransformer: Inverted Transformers Are Effective for Time Series Forecastingcited by 931
  7. 2017Modeling Long- and Short-Term Temporal Patterns with Deep Neural Networkscited by 879
  8. 2024TimeMixer: Decomposable Multiscale Mixing for Time Series Forecastingcited by 346
  9. 2021Monash Time Series Forecasting Archivecited by 267
  10. 2024TFB: Towards Comprehensive and Fair Benchmarking of Time Series Forecasting Methodscited by 249
  11. 2023TSMixer: An All-MLP Architecture for Time Series Forecastingcited by 240
  12. 2024GIFT-Eval: A Benchmark For General Time Series Forecasting Model Evaluationcited by 212
18 more
  1. 2023Exploring Progress in Multivariate Time Series Forecasting: Comprehensive Benchmarking and Heterogeneity Analysiscited by 117
  2. 2024TimeMixer++: A General Time Series Pattern Machine for Universal Predictive Analysiscited by 116
  3. 2024DUET: Dual Clustering Enhanced Multivariate Time Series Forecastingcited by 107
  4. 2024CycleNet: Enhancing Time Series Forecasting through Modeling Periodic Patternscited by 106
  5. 2023The Capacity and Robustness Trade-off: Revisiting the Channel Independent Strategy for Multivariate Time Series Forecastingcited by 101
  6. 2021Chaos as an interpretable benchmark for forecasting and data-driven modellingcited by 88
  7. 2017Conditional independence testing based on a nearest-neighbor estimator of conditional mutual informationcited by 70
  8. 2025TimeFilter: Patch-Specific Spatial-Temporal Graph Filtration for Time Series Forecastingcited by 64
  9. 2024From Similarity to Superiority: Channel Clustering for Time Series Forecastingcited by 37
  10. 2026It's TIME: Towards the Next Generation of Time Series Forecasting Benchmarkscited by 33
  11. 2025CausalRivers -- Scaling up benchmarking of causal discovery for real-world time-seriescited by 26
  12. 2025OLinear: A Linear Model for Time Series Forecasting in Orthogonally Transformed Domaincited by 26
  13. 2024Cherry-Picking in Time Series Forecasting: How to Select Datasets to Make Your Model Shinecited by 10
  14. 2025Physiome-ODE: A Benchmark for Irregularly Sampled Multivariate Time Series Forecasting Based on Biological ODEscited by 10
  15. 2026QuitoBench: A High-Quality Open Time Series Forecasting Benchmarkcited by 10
  16. 2025Channel Dependence, Limited Lookback Windows, and the Simplicity of Datasets: How Biased is Time Series Forecasting?cited by 8
  17. 2025There are no Champions in Supervised Long-Term Time Series Forecastingcited by 8
  18. 2025NetBurst: Event-Centric Forecasting of Bursty, Intermittent Time Seriescited by 2

Built on by

Not cited yet.