Pith. sign in

REVIEW 20 cited by

TSMixer: An All-MLP Architecture for Time Series Forecasting

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

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

Si-An Chen , Chun-Liang Li , Nate Yoder , Sercan O. Arik , Tomas Pfister This is my paper · ORCID
classification cs.LGcs.AI
keywords tsmixermodelsforecastingtimebenchmarksdeepseriestime-series
verification ladder T0 review T1 audit T2 compute T3 formal
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). Continue with ORCID to comment.

Forward citations

Cited by 20 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. OpenAlex reports about 122 citations worldwide. Full citation record

  1. Agoran: An Agentic Open Marketplace for 6G RAN Automation

    cs.NI 2025-08 unverdicted novelty 7.0 of 10

    Agoran uses AI agents with legislative, executive, and judicial branches to automatically negotiate and manage 6G network slices, achieving large performance gains on a 5G testbed.

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

    cs.LG 2026-02 conditional novelty 6.0 of 10

    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.

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

    cs.LG 2025-11 conditional novelty 6.0 of 10

    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.

  4. Dynamic Relational Priming Improves Transformer in Multivariate Time Series

    cs.LG 2025-09 conditional novelty 6.0 of 10

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

  5. TriForecaster: A Mixture of Experts Framework for Multi-Region Electric Load Forecasting with Tri-dimensional Specialization

    cs.LG 2025-08 conditional novelty 6.0 of 10

    TriForecaster, a mixture-of-experts multi-task model with region, context, and time-specialized experts, reports 22.4% lower MSE than prior multi-region load forecasting baselines across four datasets.

  6. Predicting Large-scale Urban Network Dynamics with Energy-informed Graph Neural Diffusion

    cs.LG 2025-07 conditional novelty 6.0 of 10

    A scalable spatiotemporal Transformer, ScaleSTF, matches the accuracy of much larger models on city-scale forecasting tasks at a fraction of the compute and memory cost.

  7. Jigsaw: Training Multi-Billion-Parameter AI Weather Models with Optimized Model Parallelism

    cs.LG 2025-07 conditional novelty 6.0 of 10

    An MLP-based weather model called WeatherMixer trains efficiently with a new Jigsaw parallelization scheme that shards data and model across GPUs, reaching 11 PFLOPs on 256 GPUs with 72% scaling efficiency.

  8. Cluster-Aware Causal Mixer for Online Anomaly Detection in Multivariate Time Series

    cs.LG 2025-05 conditional novelty 6.0 of 10

    A cluster-aware causal MLP-mixer with sequential evidence-accumulation scoring achieves state-of-the-art best-F1 on six multivariate time-series anomaly detection benchmarks.

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

    cs.LG 2026-07 conditional novelty 5.5 of 10

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

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

    cs.LG 2026-07 conditional novelty 5.0 of 10

    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.

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

    cs.LG 2025-11 conditional novelty 5.0 of 10

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

  12. FLDmamba: Integrating Fourier and Laplace Transform Decomposition with Mamba for Enhanced Time Series Prediction

    cs.LG 2025-07 conditional novelty 5.0 of 10

    FLDmamba combines a learnable Fourier filter on Mamba's step size with a damped-sinusoid output layer and reports superior long-term forecasting accuracy on standard benchmarks.

  13. Causality analysis of electricity market liberalization on electricity price using novel Machine Learning methods

    econ.GN 2025-07 conditional novelty 5.0 of 10

    Using machine-learning counterfactual forecasts, the paper estimates that US electricity market liberalization cut residential prices by about 7% in the first two years.

  14. Fourier Basis Mapping: A Time-Frequency Learning Framework for Time Series Forecasting

    cs.LG 2025-07 conditional novelty 5.0 of 10

    Using Fourier basis expansion to build time-frequency features improves long-term and short-term time series forecasting across linear, MLP, and transformer backbones.

  15. Evaluation of a Foundational Model and Stochastic Models for Forecasting Sporadic or Spiky Production Outages of High-Performance Machine Learning Services

    cs.LG 2025-06 conditional novelty 5.0 of 10

    On seven years of monthly production outage counts from a large ML service, a fine-tuned TimesFM foundation model beats moving-average and autoregressive baselines for total outages, but per root cause the best model varies.

  16. TFKAN: Time-Frequency KAN for Long-Term Time Series Forecasting

    cs.LG 2025-06 conditional novelty 5.0 of 10

    TFKAN places Kolmogorov-Arnold Networks directly on FFT coefficients alongside a time-domain KAN branch, improving long-term forecast accuracy on seven benchmark datasets.

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

    cs.LG 2026-02 conditional novelty 4.0 of 10

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

  18. User Trajectory Prediction Unifying Global and Local Temporal Information

    cs.LG 2025-08 conditional novelty 4.0 of 10

    A hybrid MLP, multi-scale CNN, and cross-attention model reduces trajectory prediction error by a few percent over ModernTCN on the GeoLife dataset, with the largest gain on the 15-second sampling interval data.

  19. DC-Mamber: A Dual Channel Prediction Model based on Mamba and Linear Transformer for Multivariate Time Series Forecasting

    cs.AI 2025-07 reject novelty 4.0 of 10

    A dual-channel forecasting architecture combining channel-independent Mamba and channel-mixing linear attention is proposed, but its claimed superiority is contradicted by its own experimental table on several dataset...

  20. W4S4: WaLRUS Meets S4 for Long-Range Sequence Modeling

    cs.LG 2025-06 conditional novelty 3.0 of 10

    W4S4 initializes S4 state space models with WaLRUS wavelet frames and reports better delay reconstruction and classification accuracy than HiPPO-based S4, with frozen (A,B).

Pith tools