Papers › Unified Training of Universal Time Series Forecasting Transformers

Unified Training of Universal Time Series Forecasting Transformers

4 Feb 2024arXiv:2402.02592archive 2025-07-28

Gerald Woo, Chenghao Liu, Akshat Kumar, Caiming Xiong, Silvio Savarese, Doyen Sahoo

Deep learning for time series forecasting has traditionally operated within a one-model-per-dataset framework, limiting its potential to leverage the game-changing impact of large pre-trained models. The concept of universal forecasting, emerging from pre-training on a vast collection of time series datasets, envisions a single Large Time Series Model capable of addressing diverse downstream forecasting tasks. However, constructing such a model poses unique challenges specific to time series data: i) cross-frequency learning, ii) accommodating an arbitrary number of variates for multivariate time series, and iii) addressing the varying distributional properties inherent in large-scale data. To address these challenges, we present novel enhancements to the conventional time series Transformer architecture, resulting in our proposed Masked Encoder-based Universal Time Series Forecasting Transformer (Moirai). Trained on our newly introduced Large-scale Open Time Series Archive (LOTSA) featuring over 27B observations across nine domains, Moirai achieves competitive or superior performance as a zero-shot forecaster when compared to full-shot models. Code, data, and model weights can be found at https://github.com/SalesforceAIResearch/uni2ts.

Paper PDF Code

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

SalesforceAIResearch/uni2ts officialmentioned in papermentioned on GitHubjaxApache-2.0 report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Time Series Time Series Forecasting

Results from the paper archive 2025-07-28

Task	Dataset	Model	Metric	Value	Rank at snapshot	Leaderboard	Report
Time Series Forecasting	ETTh1 (336) Multivariate	MOIRAISmall	MAE	0.429	#13 of 72	Archive leaderboard	report
Time Series Forecasting	ETTh1 (336) Multivariate	MOIRAISmall	MSE	0.412	#13 of 72	Archive leaderboard	report
Time Series Forecasting	ETTh1 (336) Multivariate	MOIRAIBase	MAE	0.450	#42 of 72	Archive leaderboard	report
Time Series Forecasting	ETTh1 (336) Multivariate	MOIRAIBase	MSE	0.456	#42 of 72	Archive leaderboard	report
Time Series Forecasting	ETTh1 (336) Multivariate	MOIRAILarge	MAE	0.474	#64 of 72	Archive leaderboard	report
Time Series Forecasting	ETTh1 (336) Multivariate	MOIRAILarge	MSE	0.514	#64 of 72	Archive leaderboard	report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Absolute Position Encodings Adam Attention BPE Dense Connections Dropout Label Smoothing Layer Normalization Linear Layer Multi-Head Attention Position-Wise Feed-Forward Layer Residual Connection Softmax Transformer

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