Browse State-of-the-Art › Time Series Forecasting

Time Series Forecasting

714 papers with code · 86 benchmarks · 46 datasets archive 2025-07-28

Time Series

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 )

Description from the archive archive 2025-07-28.

Benchmarks archive 2025-07-28

86 leaderboard tables shown for this task, 86 with rows (a “benchmark” on this site is a table with at least one row, as on /sota), ordered by row count. “Best model” is the first row in the archive's own order at snapshot; nothing is re-ranked here and metric direction is not recorded in the archive. PwC's Trend sparklines are not in the archive, so that column is omitted. 10 shown of 86 until expanded.

Syntology 60 newer papers were judged to report on this task's benchmarks so far (Claude Sonnet 4.5, reading their tables: a model's judgement, not a result; 5 placed): ETTh1 (336) Multivariate (32 judged, 0 placed), ETTh1 (720) Multivariate (32 judged, 0 placed), ETTh1 (192) Multivariate (34 judged, 0 placed), ETTh1 (96) Multivariate (34 judged, 0 placed), Weather (192) (38 judged, 3 placed), Weather (96) (44 judged, 0 placed), Electricity (96) (18 judged, 0 placed), Weather (336) (36 judged, 0 placed), Weather (720) (33 judged, 0 placed), Electricity (336) (1 judged, 0 placed), ETTm1 (192) Multivariate (1 judged, 0 placed), ETTm2 (192) Multivariate (11 judged, 0 placed), ETTm2 (96) Multivariate (29 judged, 0 placed), ETTm2 (720) Multivariate (1 judged, 0 placed), Traffic (192) (1 judged, 0 placed), Traffic (96) (1 judged, 0 placed), ETTh1 (11 judged, 2 placed), Solar (192) (1 judged, 0 placed), Solar (96) (1 judged, 0 placed), Weather (7 judged, 0 placed). Drawn from 8,886 of 18,637 newer papers checked so far (733 more had nothing to place; 41 could not be read).

DatasetBest model (first row in archive order)PaperCodeSyntologyCompare
ETTh1 (336) Multivariate (72 rows) D-PAD D-PAD: Deep-Shallow Multi-Frequency Patterns Disentangling for... code — Compare
ETTh1 (720) Multivariate (22 rows) DiPE-Linear Disentangled Interpretable Representation for Efficient Long-term... code — Compare
ETTh2 (336) Multivariate (20 rows) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
ETTh2 (720) Multivariate (20 rows) RLinear Revisiting Long-term Time Series Forecasting: An Investigation on... code Syntology official: not harvested · 0 ran · 2 unverified Compare
ETTh1 (192) Multivariate (17 rows) PatchMixer PatchMixer: A Patch-Mixing Architecture for Long-Term Time Series... code Syntology official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified Compare
ETTh2 (192) Multivariate (16 rows) PatchMixer PatchMixer: A Patch-Mixing Architecture for Long-Term Time Series... code Syntology official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified Compare
ETTh2 (96) Multivariate (16 rows) PatchMixer PatchMixer: A Patch-Mixing Architecture for Long-Term Time Series... code Syntology official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified Compare
ETTh1 (96) Multivariate (15 rows) SegRNN SegRNN: Segment Recurrent Neural Network for Long-Term Time Series... code — Compare
Weather (192) (13 rows) SegRNN SegRNN: Segment Recurrent Neural Network for Long-Term Time Series... code — Compare
ETTh1 (720) Univariate (12 rows) AutoCon Self-Supervised Contrastive Learning for Long-term Forecasting code Syntology official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified Compare
Weather (96) (12 rows) SegRNN SegRNN: Segment Recurrent Neural Network for Long-Term Time Series... code — Compare
Electricity (96) (11 rows) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
ETTh2 (720) Univariate (11 rows) AutoCon Self-Supervised Contrastive Learning for Long-term Forecasting code Syntology official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified Compare
Weather (336) (11 rows) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
Weather (720) (11 rows) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
ETTh1 (336) Univariate (10 rows) SegRNN SegRNN: Segment Recurrent Neural Network for Long-Term Time Series... code — Compare
ETTh2 (336) Univariate (10 rows) PatchMixer PatchMixer: A Patch-Mixing Architecture for Long-Term Time Series... code Syntology official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified Compare
Electricity (336) (9 rows) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
ETTm1 (192) Multivariate (9 rows) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
ETTm1 (96) Multivariate (9 rows) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
ETTm2 (192) Multivariate (9 rows) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
ETTm2 (96) Multivariate (9 rows) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
Electricity (192) (8 rows) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
Electricity (720) (8 rows) MoLE-RMLP Mixture-of-Linear-Experts for Long-term Time Series Forecasting code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
ETTm1 (336) Multivariate (8 rows) LTBoost (drop_last=false) LTBoost: Boosted Hybrids of Ensemble Linear and Gradient... code — Compare
ETTm1 (720) Multivariate (8 rows) LTBoost (drop_last=false) LTBoost: Boosted Hybrids of Ensemble Linear and Gradient... code — Compare
ETTm2 (336) Multivariate (8 rows) LTBoost (drop_last=false) LTBoost: Boosted Hybrids of Ensemble Linear and Gradient... code — Compare
ETTm2 (720) Multivariate (8 rows) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
MLO-Cn2 (7 rows) GBRT Effective Benchmarks for Optical Turbulence Modeling code — Compare
PeMSD7 (7 rows) STGCN-Cov Conditional Temporal Neural Processes with Covariance Loss code — Compare
ETTh1 (96) Univariate (6 rows) SegRNN SegRNN: Segment Recurrent Neural Network for Long-Term Time Series... code — Compare
ETTh2 (192) Univariate (6 rows) PatchMixer PatchMixer: A Patch-Mixing Architecture for Long-Term Time Series... code Syntology official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified Compare
ETTh2 (96) Univariate (6 rows) PatchMixer PatchMixer: A Patch-Mixing Architecture for Long-Term Time Series... code Syntology official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified Compare
ETTh1 (192) Univariate (5 rows) PatchMixer PatchMixer: A Patch-Mixing Architecture for Long-Term Time Series... code Syntology official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified Compare
ETTh1 (24) Multivariate (5 rows) SCINet SCINet: Time Series Modeling and Forecasting with Sample... code Syntology official: no sample here; runs from other or unrecorded repositories · 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified Compare
ETTh1 (24) Univariate (5 rows) SCINet SCINet: Time Series Modeling and Forecasting with Sample... code Syntology official: no sample here; runs from other or unrecorded repositories · 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified Compare
ETTh2 (24) Univariate (5 rows) SCINet SCINet: Time Series Modeling and Forecasting with Sample... code Syntology official: no sample here; runs from other or unrecorded repositories · 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified Compare
USNA-Cn2 (short-duration) (5 rows) GBRT Effective Benchmarks for Optical Turbulence Modeling code — Compare
ETTh1 (168) Multivariate (4 rows) SCINet SCINet: Time Series Modeling and Forecasting with Sample... code Syntology official: no sample here; runs from other or unrecorded repositories · 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified Compare
ETTh1 (168) Univariate (4 rows) SCINet SCINet: Time Series Modeling and Forecasting with Sample... code Syntology official: no sample here; runs from other or unrecorded repositories · 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified Compare
ETTh1 (48) Multivariate (4 rows) SCINet SCINet: Time Series Modeling and Forecasting with Sample... code Syntology official: no sample here; runs from other or unrecorded repositories · 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified Compare
ETTh1 (48) Univariate (4 rows) SCINet SCINet: Time Series Modeling and Forecasting with Sample... code Syntology official: no sample here; runs from other or unrecorded repositories · 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified Compare
ETTh2 (168) Multivariate (4 rows) SCINet SCINet: Time Series Modeling and Forecasting with Sample... code Syntology official: no sample here; runs from other or unrecorded repositories · 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified Compare
ETTh2 (168) Univariate (4 rows) Informer Informer: Beyond Efficient Transformer for Long Sequence... code Syntology official: harvested, nothing ran · 62 ran (of which 53 constructed an object rather than computing a result; 62 with no instrument failure: 0 honoured, 1 violated, 61 with no contract checked; 0 where Syntology's instrument failed) · 14 unverified Compare
ETTh2 (24) Multivariate (4 rows) SCINet SCINet: Time Series Modeling and Forecasting with Sample... code Syntology official: no sample here; runs from other or unrecorded repositories · 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified Compare
ETTh2 (48) Multivariate (4 rows) SCINet SCINet: Time Series Modeling and Forecasting with Sample... code Syntology official: no sample here; runs from other or unrecorded repositories · 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified Compare
ETTh2 (48) Univariate (4 rows) SCINet SCINet: Time Series Modeling and Forecasting with Sample... code Syntology official: no sample here; runs from other or unrecorded repositories · 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified Compare
Traffic (192) (3 rows) PRformer PRformer: Pyramidal Recurrent Transformer for Multivariate Time... code Syntology official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified Compare
Traffic (336) (3 rows) TSMixer TSMixer: Lightweight MLP-Mixer Model for Multivariate Time Series... code Syntology official (archive's flag): 7 ran · 7 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified Compare
Traffic (720) (3 rows) PRformer PRformer: Pyramidal Recurrent Transformer for Multivariate Time... code Syntology official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified Compare
Traffic (96) (3 rows) PRformer PRformer: Pyramidal Recurrent Transformer for Multivariate Time... code Syntology official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified Compare
ETTh1 (96) (2 rows) Ada-MSHyper Ada-MSHyper: Adaptive Multi-Scale Hypergraph Transformer for Time... code Syntology official: harvested, nothing ran · 0 ran · 4 unverified Compare
Extreme Events > Natural Disasters > Hurricane (2 rows) UberNN Deep and Confident Prediction for Time Series at Uber code Syntology 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 2 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified Compare
Weather2K1786 (192) (2 rows) MoLE-RLinear Mixture-of-Linear-Experts for Long-term Time Series Forecasting code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
Weather2K1786 (720) (2 rows) MoLE-RLinear Mixture-of-Linear-Experts for Long-term Time Series Forecasting code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
Weather2K1786 (96) (2 rows) MoLE-DLinear Mixture-of-Linear-Experts for Long-term Time Series Forecasting code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
Weather2K850 (96) (2 rows) MoLE-RLinear Mixture-of-Linear-Experts for Long-term Time Series Forecasting code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
ETTh1 (1 row) BEAT BEAT: Balanced Frequency Adaptive Tuning for Long-Term Time-Series... — — Compare
ETTh1 (48) (1 row) Time-LLM Time-LLM: Time Series Forecasting by Reprogramming Large Language Models code Syntology 7 ran (of which 7 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified; every one of the 7 samples that ran constructed an object rather than computing a result Compare
Exchange (192) (1 row) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
Exchange (336) (1 row) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
Exchange (720) (1 row) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
Exchange (96) (1 row) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
Consumer Spendings (1 row) TSE-SC Traffic transformer: Capturing the continuity and periodicity of... — — Compare
Illness (24) (1 row) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
Illness (36) (1 row) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
Illness (48) (1 row) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
Illness (60) (1 row) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
Korea Composite Stock Price Index (1 row) GA-LSTM Genetic Algorithm-Optimized Long Short-Term Memory Network for... — — Compare
Solar (192) (1 row) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
Solar (336) (1 row) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
Solar (720) (1 row) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
Solar (96) (1 row) xPatch xPatch: Dual-Stream Time Series Forecasting with Exponential... code — Compare
Weather (1 row) BEAT BEAT: Balanced Frequency Adaptive Tuning for Long-Term Time-Series... — — Compare
Weather2K114 (192) (1 row) MoLE-DLinear Mixture-of-Linear-Experts for Long-term Time Series Forecasting code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
Weather2K114 (336) (1 row) MoLE-DLinear Mixture-of-Linear-Experts for Long-term Time Series Forecasting code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
Weather2K114 (720) (1 row) MoLE-DLinear Mixture-of-Linear-Experts for Long-term Time Series Forecasting code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
Weather2K114 (96) (1 row) MoLE-DLinear Mixture-of-Linear-Experts for Long-term Time Series Forecasting code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
Weather2K1786 (336) (1 row) MoLE-DLinear Mixture-of-Linear-Experts for Long-term Time Series Forecasting code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
Weather2K79 (192) (1 row) MoLE-DLinear Mixture-of-Linear-Experts for Long-term Time Series Forecasting code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
Weather2K79 (336) (1 row) MoLE-DLinear Mixture-of-Linear-Experts for Long-term Time Series Forecasting code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
Weather2K79 (720) (1 row) MoLE-DLinear Mixture-of-Linear-Experts for Long-term Time Series Forecasting code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
Weather2K79 (96) (1 row) MoLE-DLinear Mixture-of-Linear-Experts for Long-term Time Series Forecasting code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
Weather2K850 (192) (1 row) MoLE-DLinear Mixture-of-Linear-Experts for Long-term Time Series Forecasting code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
Weather2K850 (336) (1 row) MoLE-DLinear Mixture-of-Linear-Experts for Long-term Time Series Forecasting code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
Weather2K850 (720) (1 row) MoLE-DLinear Mixture-of-Linear-Experts for Long-term Time Series Forecasting code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare

Syntology column: samples harvested from the paper's repositories and executed on synthesized fixtures; “ran” is not a correctness claim and does not order the table. Each line reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. When the archive marks a repository official for the row's paper, the line starts with that repository's state (“official” is the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from. A dash means no Syntology record for that paper, not a recorded non-run. Each “ran” line links to that paper's sample list. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

Libraries

Not in the archive: the export carries no per-task library table, so there is nothing to show at snapshot 2025-07-28.

Datasets archive 2025-07-28

46 datasets whose archive record lists this task, ordered by the archive's paper count. 30 shown of 46 until expanded.

Subtasks archive 2025-07-28

6 subtasks in the archive's task tree.

Parent tasks archive 2025-07-28

Papers archive 2025-07-28

archive papers tagged: 1,609 · with a code link: 714 · where Syntology ran a sample: 211 (185 with a run with no instrument failure, 26 where every run was a failure of Syntology's instrument) Syntology

Show: all tagged papersonly where code ran (211 of 1,609 tagged: 185 with a run with no instrument failure, 26 where every run was a failure of Syntology's instrument)

Page 1 of 17: papers 1 to 100 of 1,609, in archive order: by repositories listed in the archive (most first), then newest first, not by stars (the archive holds no stars, so PwC's “Social” and “Latest” sorts cannot be reproduced). Papers that list no repository come after every paper that lists one.

Papers without a page here are shown as plain text. A Syntology line reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the figure “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code); hover it for the repositories the samples that ran came from. The first sentence of the abstract is shown for the first 30 papers; every abstract is on its paper's page.

Syntology lines on 65 of the papers shown; no Syntology record for the others (a paper without an arXiv id cannot be joined to the graph, and absence from the graph layer is not a recorded non-run). “Ran” means the sample executed on a synthesized fixture, not that the paper's result was reproduced; each line links to that paper's sample list. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections