Pith. sign in

REVIEW 22 cited by

TSMixer: An All-MLP Architecture for Time Series Forecasting

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2303.06053 v5 pith:AREAWPEE submitted 2023-03-10 cs.LG cs.AI

TSMixer: An All-MLP Architecture for Time Series Forecasting

classification cs.LG cs.AI
keywords tsmixermodelsforecastingtimebenchmarksdeepseriestime-series
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Real-world time-series datasets are often multivariate with complex dynamics. To capture this complexity, high capacity architectures like recurrent- or attention-based sequential deep learning models have become popular. However, recent work demonstrates that simple univariate linear models can outperform such deep learning models on several commonly used academic benchmarks. Extending them, in this paper, we investigate the capabilities of linear models for time-series forecasting and present Time-Series Mixer (TSMixer), a novel architecture designed by stacking multi-layer perceptrons (MLPs). TSMixer is based on mixing operations along both the time and feature dimensions to extract information efficiently. On popular academic benchmarks, the simple-to-implement TSMixer is comparable to specialized state-of-the-art models that leverage the inductive biases of specific benchmarks. On the challenging and large scale M5 benchmark, a real-world retail dataset, TSMixer demonstrates superior performance compared to the state-of-the-art alternatives. Our results underline the importance of efficiently utilizing cross-variate and auxiliary information for improving the performance of time series forecasting. We present various analyses to shed light into the capabilities of TSMixer. The design paradigms utilized in TSMixer are expected to open new horizons for deep learning-based time series forecasting. The implementation is available at https://github.com/google-research/google-research/tree/master/tsmixer

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 22 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. JaGuard: Position Error Correction of GNSS Jamming with Deep Temporal Graphs

    cs.LG 2025-09 unverdicted novelty 7.0

    JaGuard recasts GNSS jamming mitigation as dynamic graph regression and uses a Heterogeneous Graph ConvLSTM to estimate 2D positional deviation, achieving MAEs of 2.26-2.61 cm on mixed-power real datasets and remainin...

  2. Deep Time Series Models: A Comprehensive Survey and Benchmark

    cs.LG 2024-07 unverdicted novelty 7.0

    This survey and benchmark of deep time series models using the released TSLib library finds that models with specific structures perform well only on distinct analysis tasks.

  3. A decoder-only foundation model for time-series forecasting

    cs.CL 2023-10 unverdicted novelty 7.0

    A pretrained decoder-only patched transformer achieves near state-of-the-art zero-shot forecasting performance across diverse time series datasets and settings.

  4. How Good Can Linear Models Be for Time-Series Forecasting?

    cs.LG 2026-06 unverdicted novelty 6.0

    Tuned Ridge regression with series-specific preprocessing beats prior linear forecasters and complex neural baselines on most dataset-horizon pairs across eight benchmarks.

  5. How Good Can Linear Models Be for Time-Series Forecasting?

    cs.LG 2026-06 conditional novelty 6.0

    Optimized Ridge regression with series-specific preprocessing beats prior linear forecasters and exceeds Transformer, MLP, and CNN baselines on six of eight time-series benchmarks.

  6. QuITE: Query-Based Irregular Time Series Embedding

    cs.LG 2026-05 unverdicted novelty 6.0

    QuITE is a query-token embedding layer that aggregates irregular observations via self-attention to produce backbone-compatible latents without generating artificial values.

  7. Three-Stage Learning Unlocks Strong Performance in Simple Models for Long-Term Time Series Forecasting

    cs.LG 2026-05 conditional novelty 6.0

    STAIR's three-stage training enables simple temporal models to match or exceed complex baselines on long-term forecasting benchmarks by combining shared learning, individual adaptation, and residual cross-variable modeling.

  8. Exploring the Potential of Probabilistic Transformer for Time Series Modeling: A Report on the ST-PT Framework

    cs.LG 2026-04 unverdicted novelty 6.0

    ST-PT turns transformers into explicit factor graphs for time series, enabling structural injection of symbolic priors, per-sample conditional generation, and principled latent autoregressive forecasting via MFVI iterations.

  9. Amortized Predictability-aware Training Framework for Time Series Forecasting and Classification

    cs.LG 2026-02 conditional novelty 6.0

    APTF reweights training samples by loss-based predictability buckets and uses an amortization model to stabilize the estimates, improving accuracy across TSF and TSC baselines.

  10. HieraMix: A Hierarchical MLP-Mixer for Large-Scale Traffic Forecasting

    cs.LG 2025-11 conditional novelty 6.0

    HSTMixer, a hierarchical all-MLP mixer with adaptive region mixing, reports state-of-the-art MAE/RMSE/MAPE on four LargeST datasets at 15-minute resolution.

  11. Dynamic Relational Priming Improves Transformer in Multivariate Time Series

    cs.LG 2025-09 conditional novelty 6.0

    Prime attention modulates attention keys and values per channel-pair and reports improved MTS forecasting accuracy across several benchmarks.

  12. GatedLinear: Adaptive Routing of Complementary Linear Bases for Time Series Forecasting

    cs.LG 2026-07 conditional novelty 5.5

    Adaptive soft routing among three complementary linear bases via a channel-horizon-phase gate yields competitive multivariate forecasting accuracy with a small, interpretable model.

  13. PIER: Physics-Informed Environmental Retrieval for Time-Series Modeling

    cs.LG 2026-07 conditional novelty 5.0

    PIER augments embedding-based retrieval for lake modeling with a physics-aware stream scored by local verifiers, improving water temperature and dissolved oxygen prediction across 356 lakes.

  14. CLOUDADV: Decision-Aligned Instance Sizing with Zero-Shot Foundation Models under Drift

    cs.AI 2026-06 unverdicted novelty 5.0

    CLOUDADV combines zero-shot forecasting with LLM-generated recommendations for cloud instance sizing, reporting 52.9% simulated monthly cost savings in a seven-VM Azure case study.

  15. Self-Adaptive Scale Handling for Forecasting Time Series with Scale Heterogeneity

    cs.LG 2026-06 unverdicted novelty 5.0

    The self-adaptive scale-handling (AS) module with scale calibrating and scaling selection learns per-input scale factors to improve forecasting on scale-heterogeneous time series.

  16. CHAM-net: A Contrastive Hierarchical Adaptive Meta-network for Robust Global Methane Flux Prediction

    cs.LG 2026-05 unverdicted novelty 5.0

    CHAM-net is a contrastive hierarchical adaptive meta-network that conditions predictions on historical site data to outperform baselines on methane flux tasks with nRMSE down to 0.43.

  17. GenHAR: Generalizing Cross-domain Human Activity Recognition for Last-mile Delivery

    cs.CV 2026-05 unverdicted novelty 5.0

    GenHAR generalizes cross-domain human activity recognition by 9.97% accuracy and 6.4x lower FLOPs via tokenized sensor data, frequency channel correlations, selective masking, and efficient attention, with deployment ...

  18. CaReTS: A Multi-Task Framework Unifying Classification and Regression for Time Series Forecasting

    cs.LG 2025-11 conditional novelty 5.0

    CaReTS forecasts multi-step time series by combining a trend classifier with a deviation regressor in a residual, uncertainty-weighted multi-task framework.

  19. FlowMixer: A Depth-Agnostic Neural Architecture for Interpretable Spatiotemporal Forecasting

    cs.LG 2025-05 unverdicted novelty 5.0

    A single-layer architecture called FlowMixer uses constrained matrix operations and a semi-group property to enable depth-agnostic, interpretable spatiotemporal forecasting with direct eigenmode extraction.

  20. Revisiting Long-term Time Series Forecasting: An Investigation on Linear Mapping

    cs.LG 2023-05 unverdicted novelty 5.0

    Affine mapping dominates LTSF benchmarks by learning similar input-to-output transition matrices, captures periodic signals well but struggles with non-periodic or cross-channel varying periods; reversible normalizati...

  21. Understanding Key Features of Time Series Foundation Models from Epidemic Forecasting

    cs.LG 2026-06 unverdicted novelty 4.0

    Mixture-of-experts fusing multiple pretrained forecasters achieves strongest performance on influenza time series, with pretraining gains largest at longer horizons when domain-aligned and LLM methods underperforming.

  22. SEMixer: Semantics Enhanced MLP-Mixer for Multiscale Mixing and Long-term Time Series Forecasting

    cs.LG 2026-02 conditional novelty 4.0

    SEMixer combines random-mask patch interactions with progressive adjacent-scale mixing and reports improved MSE/MAE on common long-term forecasting benchmarks.