Pith. sign in

REVIEW 11 cited by

Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term 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 2106.13008 v5 pith:YATVK45K submitted 2021-06-24 cs.LG cs.AI

Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting

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

Extending the forecasting time is a critical demand for real applications, such as extreme weather early warning and long-term energy consumption planning. This paper studies the long-term forecasting problem of time series. Prior Transformer-based models adopt various self-attention mechanisms to discover the long-range dependencies. However, intricate temporal patterns of the long-term future prohibit the model from finding reliable dependencies. Also, Transformers have to adopt the sparse versions of point-wise self-attentions for long series efficiency, resulting in the information utilization bottleneck. Going beyond Transformers, we design Autoformer as a novel decomposition architecture with an Auto-Correlation mechanism. We break with the pre-processing convention of series decomposition and renovate it as a basic inner block of deep models. This design empowers Autoformer with progressive decomposition capacities for complex time series. Further, inspired by the stochastic process theory, we design the Auto-Correlation mechanism based on the series periodicity, which conducts the dependencies discovery and representation aggregation at the sub-series level. Auto-Correlation outperforms self-attention in both efficiency and accuracy. In long-term forecasting, Autoformer yields state-of-the-art accuracy, with a 38% relative improvement on six benchmarks, covering five practical applications: energy, traffic, economics, weather and disease. Code is available at this repository: \url{https://github.com/thuml/Autoformer}.

discussion (0)

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

Forward citations

Cited by 11 Pith papers

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

  1. TelecomTS: A Multi-Modal Observability Dataset for Time Series and Language Analysis

    cs.AI 2025-10 conditional novelty 7.0

    TelecomTS is a new observability dataset from 5G networks that preserves absolute scale and supports multi-modal tasks, showing that current time series and language models struggle with abrupt noisy dynamics.

  2. Dynamic Linear Coregionalization for Realistic Synthetic Multivariate Time Series

    cs.LG 2026-04 unverdicted novelty 6.0

    DynLMC creates synthetic multivariate time series with dynamic inter-channel correlations that improve zero-shot forecasting performance when used to fine-tune foundation models across nine benchmarks.

  3. Dynamic Linear Coregionalization for Realistic Synthetic Multivariate Time Series

    cs.LG 2026-04 unverdicted novelty 6.0

    DynLMC creates synthetic time series data with dynamic inter-channel correlations that improve zero-shot forecasting in foundation models across multiple benchmarks.

  4. Neural CDEs as Correctors for Learned Time Series Models

    cs.LG 2025-12 unverdicted novelty 6.0

    Neural CDEs serve as correctors that reduce error accumulation in multi-step forecasts from learned time-series models across synthetic, physics, and real-world data.

  5. RhyMix: A Lightweight Adaptive Multi-Rhythm Network for Long-Term Time Series Forecasting

    cs.LG 2026-07 conditional novelty 5.5

    RhyMix reaches state-of-the-art long-term multivariate forecasting on 10 of 12 public benchmarks with a ~40K-parameter dual-path adaptive architecture of linear complexity.

  6. Exogenous Dropout: A Simple, Strong Baseline for Corruption-Robust Time Series Forecasting with Covariates

    cs.LG 2026-07 conditional novelty 5.0

    Exogenous dropout, a model-agnostic whole-channel training mask, restores corruption robustness better than a deliberately bounded gated-FiLM architecture across three forecasting domains.

  7. MSTN: A Lightweight and Fast Model for General TimeSeries Analysis

    cs.LG 2025-11 unverdicted novelty 5.0

    MSTN introduces a lightweight multi-scale temporal network using convolutional encoding, recurrent or attention-based modeling, and gated fusion to achieve claimed state-of-the-art results on 21 of 27 time series benc...

  8. Characteristic Root Analysis and Regularization for Linear Time Series Forecasting

    cs.LG 2025-09 unverdicted novelty 5.0

    Characteristic roots govern dynamics in linear forecasting models but noise induces spurious roots; rank reduction and Root Purge regularization mitigate this for more robust predictions.

  9. MR-ImagenTime: Multi-Resolution Time Series Generation through Dual Image Representations

    cs.LG 2026-03 unverdicted novelty 4.0

    MR-CDM uses hierarchical multi-resolution decomposition and multi-scale conditional diffusion to generate forecasts that reduce MAE and RMSE by 6-10% versus baselines like CSDI and Informer on four datasets.

  10. MSTN: A Lightweight and Fast Model for General TimeSeries Analysis

    cs.LG 2025-11 unverdicted novelty 4.0

    MSTN is a lightweight hybrid model that reports new state-of-the-art results on 33 of 40 time series benchmarks for imputation, forecasting, and classification while using under one million parameters and sub-second i...

  11. Signed Dual Attention: Capturing Signed Dependencies in Time Series Forecasting

    cs.LG 2026-06 unverdicted novelty 3.0

    Signed Dual Attention is a parameter-free attention module that models signed dependencies in time series via dual message passing to achieve two-head expressiveness in one block.