SOTAVerified

Time Series Forecasting

Time Series Forecasting is the task of fitting a model to historical, time-stamped data in order to predict future values. Traditional approaches include moving average, exponential smoothing, and ARIMA, though models as various as RNNs, Transformers, or XGBoost can also be applied. The most popular benchmark is the ETTh1 dataset. Models are typically evaluated using the Mean Square Error (MSE) or Root Mean Square Error (RMSE).

( Image credit: ThaiBinh Nguyen )

Papers

Showing 51100 of 1609 papers

TitleStatusHype
TimeMachine: A Time Series is Worth 4 Mambas for Long-term ForecastingCode3
SparseTSF: Modeling Long-term Time Series Forecasting with 1k ParametersCode3
SOFTS: Efficient Multivariate Time Series Forecasting with Series-Core FusionCode3
Deep Frequency Derivative Learning for Non-stationary Time Series ForecastingCode3
The Rise of Diffusion Models in Time-Series ForecastingCode3
MMFNet: Multi-Scale Frequency Masking Neural Network for Multivariate Time Series ForecastingCode3
ElasTST: Towards Robust Varied-Horizon Forecasting with Elastic Time-Series TransformerCode3
The Tabular Foundation Model TabPFN Outperforms Specialized Time Series Forecasting Models Based on Simple FeaturesCode3
TSI-Bench: Benchmarking Time Series ImputationCode3
Amplifier: Bringing Attention to Neglected Low-Energy Components in Time Series ForecastingCode3
PRformer: Pyramidal Recurrent Transformer for Multivariate Time Series ForecastingCode2
Pre-training Enhanced Spatial-temporal Graph Neural Network for Multivariate Time Series ForecastingCode2
Proactive Model Adaptation Against Concept Drift for Online Time Series ForecastingCode2
DeTPP: Leveraging Object Detection for Robust Long-Horizon Event PredictionCode2
Predict, Refine, Synthesize: Self-Guiding Diffusion Models for Probabilistic Time Series ForecastingCode2
Learning Deep Time-index Models for Time Series ForecastingCode2
Ada-MSHyper: Adaptive Multi-Scale Hypergraph Transformer for Time Series ForecastingCode2
PGN: The RNN's New Successor is Effective for Long-Range Time Series ForecastingCode2
Advancing Spiking Neural Networks for Sequential Modeling with Central Pattern GeneratorsCode2
One Fits All:Power General Time Series Analysis by Pretrained LMCode2
Probabilistic Time Series Forecasting with Implicit Quantile NetworksCode2
DeepAR: Probabilistic Forecasting with Autoregressive Recurrent NetworksCode2
Non-stationary Diffusion For Probabilistic Time Series ForecastingCode2
N-HiTS: Neural Hierarchical Interpolation for Time Series ForecastingCode2
Non-stationary Transformers: Exploring the Stationarity in Time Series ForecastingCode2
PatchMixer: A Patch-Mixing Architecture for Long-Term Time Series ForecastingCode2
Decoupled Dynamic Spatial-Temporal Graph Neural Network for Traffic ForecastingCode2
Multivariate Probabilistic Time Series Forecasting via Conditioned Normalizing FlowsCode2
MTS-Mixers: Multivariate Time Series Forecasting via Factorized Temporal and Channel MixingCode2
MoFE-Time: Mixture of Frequency Domain Experts for Time-Series Forecasting ModelsCode2
Multi-Patch Prediction: Adapting LLMs for Time Series Representation LearningCode2
Deep Learning for Time Series Forecasting: Tutorial and Literature SurveyCode2
Patch-wise Structural Loss for Time Series ForecastingCode2
ProbTS: Benchmarking Point and Distributional Forecasting across Diverse Prediction HorizonsCode2
MambaTS: Improved Selective State Space Models for Long-term Time Series ForecastingCode2
Context is Key: A Benchmark for Forecasting with Essential Textual InformationCode2
MG-TSD: Multi-Granularity Time Series Diffusion Models with Guided Learning ProcessCode2
Large language models can be zero-shot anomaly detectors for time series?Code2
Addressing Concept Shift in Online Time Series Forecasting: Detect-then-AdaptCode2
Model scale versus domain knowledge in statistical forecasting of chaotic systemsCode2
KAN4TSF: Are KAN and KAN-based models Effective for Time Series Forecasting?Code2
LibCity: An Open Library for Traffic PredictionCode2
Large Language Models Are Zero-Shot Time Series ForecastersCode2
Mamba-360: Survey of State Space Models as Transformer Alternative for Long Sequence Modelling: Methods, Applications, and ChallengesCode2
Learning Pattern-Specific Experts for Time Series Forecasting Under Patch-level Distribution ShiftCode2
CoST: Contrastive Learning of Disentangled Seasonal-Trend Representations for Time Series ForecastingCode2
Crossformer: Transformer Utilizing Cross-Dimension Dependency for Multivariate Time Series ForecastingCode2
MSGNet: Learning Multi-Scale Inter-Series Correlations for Multivariate Time Series ForecastingCode2
Minusformer: Improving Time Series Forecasting by Progressively Learning ResidualsCode2
HoTPP Benchmark: Are We Good at the Long Horizon Events Forecasting?Code2
Show:102550
← PrevPage 2 of 33Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1InformerMSE0.88Unverified
2QuerySelectorMSE0.85Unverified
3TransformerMSE0.83Unverified
4AarenMSE0.65Unverified
5RPMixerMSE0.52Unverified
6MOIRAILargeMSE0.51Unverified
7ATFNetMSE0.51Unverified
8AutoformerMSE0.51Unverified
9SCINetMSE0.5Unverified
10S-MambaMSE0.49Unverified
#ModelMetricClaimedVerifiedStatus
1QuerySelectorMSE1.12Unverified
2TransformerMSE1.11Unverified
3InformerMSE0.94Unverified
4GLinearMSE0.59Unverified
5SCINetMSE0.54Unverified
6MoLE-DLinearMSE0.51Unverified
7PRformerMSE0.49Unverified
8TEFNMSE0.48Unverified
9DLinearMSE0.47Unverified
10FiLMMSE0.47Unverified
#ModelMetricClaimedVerifiedStatus
1TransformerMSE2.66Unverified
2QuerySelectorMSE2.32Unverified
3InformerMSE1.67Unverified
4DLinearMSE0.45Unverified
5TEFNMSE0.42Unverified
6MoLE-DLinearMSE0.42Unverified
7FiLMMSE0.38Unverified
8MoLE-RLinearMSE0.37Unverified
9SCINetMSE0.37Unverified
10PRformerMSE0.36Unverified
#ModelMetricClaimedVerifiedStatus
1TransformerMSE3.18Unverified
2QuerySelectorMSE3.07Unverified
3InformerMSE2.34Unverified
4DLinearMSE0.61Unverified
5MoLE-DLinearMSE0.61Unverified
6SCINetMSE0.48Unverified
7FiLMMSE0.44Unverified
8TEFNMSE0.43Unverified
9TiDEMSE0.42Unverified
10MoLE-RLinearMSE0.41Unverified
#ModelMetricClaimedVerifiedStatus
1MoLE-DLinearMSE0.45Unverified
2TEFNMSE0.43Unverified
3FiLMMSE0.41Unverified
4PatchTST/64MSE0.41Unverified
5TiDEMSE0.41Unverified
6NLinearMSE0.41Unverified
7DiPE-LinearMSE0.41Unverified
8DLinearMSE0.41Unverified
9RLinearMSE0.4Unverified
10MoLE-RLinearMSE0.4Unverified
#ModelMetricClaimedVerifiedStatus
1DLinearMSE0.38Unverified
2TEFNMSE0.38Unverified
3MoLE-DLinearMSE0.36Unverified
4FiLMMSE0.36Unverified
5NLinearMSE0.34Unverified
6PatchTST/64MSE0.34Unverified
7MoLE-RLinearMSE0.34Unverified
8TiDEMSE0.33Unverified
9LTBoost (drop_last=false)MSE0.33Unverified
10PRformerMSE0.33Unverified
#ModelMetricClaimedVerifiedStatus
1DLinearMSE0.29Unverified
2TEFNMSE0.29Unverified
3MoLE-DLinearMSE0.29Unverified
4FiLMMSE0.28Unverified
5NLinearMSE0.28Unverified
6TSMixerMSE0.28Unverified
7DiPE-LinearMSE0.28Unverified
8PatchTST/64MSE0.27Unverified
9MoLE-RLinearMSE0.27Unverified
10TiDEMSE0.27Unverified
#ModelMetricClaimedVerifiedStatus
1TEFNMSE0.38Unverified
2MoLE-DLinearMSE0.38Unverified
3TiDEMSE0.38Unverified
4MoLE-RLinearMSE0.38Unverified
5FiLMMSE0.37Unverified
6PatchTST/64MSE0.37Unverified
7DiPE-LinearMSE0.37Unverified
8TSMixerMSE0.37Unverified
9RLinearMSE0.37Unverified
10TTMMSE0.36Unverified
#ModelMetricClaimedVerifiedStatus
1TEFNMSE0.23Unverified
2DLinearMSE0.22Unverified