Papers › Mixture-of-Linear-Experts for Long-term Time Series Forecasting

Mixture-of-Linear-Experts for Long-term Time Series Forecasting

11 Dec 2023arXiv:2312.06786archive 2025-07-28

Ronghao Ni, Zinan Lin, Shuaiqi Wang, Giulia Fanti

Long-term time series forecasting (LTSF) aims to predict future values of a time series given the past values. The current state-of-the-art (SOTA) on this problem is attained in some cases by linear-centric models, which primarily feature a linear mapping layer. However, due to their inherent simplicity, they are not able to adapt their prediction rules to periodic changes in time series patterns. To address this challenge, we propose a Mixture-of-Experts-style augmentation for linear-centric models and propose Mixture-of-Linear-Experts (MoLE). Instead of training a single model, MoLE trains multiple linear-centric models (i.e., experts) and a router model that weighs and mixes their outputs. While the entire framework is trained end-to-end, each expert learns to specialize in a specific temporal pattern, and the router model learns to compose the experts adaptively. Experiments show that MoLE reduces forecasting error of linear-centric models, including DLinear, RLinear, and RMLP, in over 78% of the datasets and settings we evaluated. By using MoLE existing linear-centric models can achieve SOTA LTSF results in 68% of the experiments that PatchTST reports and we compare to, whereas existing single-head linear-centric models achieve SOTA results in only 25% of cases.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 1 code sample harvested from 1 repository linked to this paper: 0 ran · 1 unverified.

By repository: official repository: 1 sample from 1 repository, 0 ran (0 with no instrument failure; 0 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

rogerni/mole officialmentioned in papermentioned on GitHubpytorchSyntology: 1 sample, 0 ran · 1 unverified 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

1 sample harvested: 0 ran · 1 unverified. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1unverified

Licence: 1 of the 1 sample is pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from rogerni/mole. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

rogerni/mole official repository 1 sample, 0 ran · 1 unverified
augmentation in 17 papers utils/augmentations.py unverified no licence file found · pointer only · 5c037d281a902e51 · report

Tasks

Mixture-of-ExpertsTime SeriesTime Series Forecasting

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Time Series Forecasting ETTh1 (192) Multivariate MoLE-RLinear MSE 0.403 #8 of 17 Archive leaderboard report
Time Series Forecasting ETTh1 (192) Multivariate MoLE-DLinear MSE 0.453 #17 of 17 Archive leaderboard report
Time Series Forecasting ETTh1 (336) Multivariate MoLE-RLinear MSE 0.43 #30 of 72 Archive leaderboard report
Time Series Forecasting ETTh1 (336) Multivariate MoLE-DLinear MSE 0.469 #50 of 72 Archive leaderboard report
Time Series Forecasting ETTh1 (720) Multivariate MoLE-RLinear MSE 0.449 #11 of 22 Archive leaderboard report
Time Series Forecasting ETTh1 (720) Multivariate MoLE-DLinear MSE 0.505 #17 of 22 Archive leaderboard report
Time Series Forecasting ETTh1 (96) Multivariate MoLE-RLinear MSE 0.375 #13 of 15 Archive leaderboard report
Time Series Forecasting ETTh1 (96) Multivariate MoLE-DLinear MSE 0.377 #14 of 15 Archive leaderboard report
Time Series Forecasting ETTh2 (192) Multivariate MoLE-RLinear MSE 0.336 #10 of 16 Archive leaderboard report
Time Series Forecasting ETTh2 (192) Multivariate MoLE-DLinear MSE 0.362 #14 of 16 Archive leaderboard report
Time Series Forecasting ETTh2 (336) Multivariate MoLE-RLinear MSE 0.371 #13 of 20 Archive leaderboard report
Time Series Forecasting ETTh2 (336) Multivariate MoLE-DLinear MSE 0.419 #15 of 20 Archive leaderboard report
Time Series Forecasting ETTh2 (720) Multivariate MoLE-RLinear MSE 0.409 #11 of 20 Archive leaderboard report
Time Series Forecasting ETTh2 (720) Multivariate MoLE-DLinear MSE 0.605 #17 of 20 Archive leaderboard report
Time Series Forecasting ETTh2 (96) Multivariate MoLE-RLinear MSE 0.273 #8 of 16 Archive leaderboard report
Time Series Forecasting ETTh2 (96) Multivariate MoLE-DLinear MSE 0.287 #14 of 16 Archive leaderboard report
Time Series Forecasting ETTm1 (192) Multivariate MoLE-DLinear MSE 0.328 #5 of 9 Archive leaderboard report
Time Series Forecasting ETTm1 (336) Multivariate MoLE-DLinear MSE 0.38 #7 of 8 Archive leaderboard report
Time Series Forecasting ETTm1 (720) Multivariate MoLE-DLinear MSE 0.447 #7 of 8 Archive leaderboard report
Time Series Forecasting ETTm1 (96) Multivariate MoLE-DLinear MSE 0.286 #4 of 9 Archive leaderboard report
Time Series Forecasting ETTm2 (192) Multivariate MoLE-DLinear MSE 0.233 #7 of 9 Archive leaderboard report
Time Series Forecasting ETTm2 (336) Multivariate MoLE-DLinear MSE 0.289 #7 of 8 Archive leaderboard report
Time Series Forecasting ETTm2 (720) Multivariate MoLE-DLinear MSE 0.399 #7 of 8 Archive leaderboard report
Time Series Forecasting ETTm2 (96) Multivariate MoLE-DLinear MSE 0.168 #8 of 9 Archive leaderboard report
Time Series Forecasting Electricity (192) MoLE-DLinear MSE 0.147 #3 of 8 Archive leaderboard report
Time Series Forecasting Electricity (336) MoLE-DLinear MSE 0.162 #4 of 9 Archive leaderboard report
Time Series Forecasting Electricity (720) MoLE-RMLP MSE 0.178 #1 of 8 Archive leaderboard report
Time Series Forecasting Electricity (720) MoLE-DLinear MSE 0.18 #2 of 8 Archive leaderboard report
Time Series Forecasting Electricity (96) MoLE-RMLP MSE 0.129 #4 of 11 Archive leaderboard report
Time Series Forecasting Electricity (96) MoLE-DLinear MSE 0.131 #7 of 11 Archive leaderboard report
Time Series Forecasting Weather (192) MoLE-RMLP MSE 0.19 #7 of 13 Archive leaderboard report
Time Series Forecasting Weather (192) MoLE-DLinear MSE 0.203 #10 of 13 Archive leaderboard report
Time Series Forecasting Weather (336) MoLE-DLinear MSE 0.238 #4 of 11 Archive leaderboard report
Time Series Forecasting Weather (720) MoLE-DLinear MSE 0.314 #5 of 11 Archive leaderboard report
Time Series Forecasting Weather (96) MoLE-DLinear MSE 0.147 #8 of 12 Archive leaderboard report
Time Series Forecasting Weather2K114 (192) MoLE-DLinear MSE 0.405 #1 of 1 Archive leaderboard report
Time Series Forecasting Weather2K114 (336) MoLE-DLinear MSE 0.415 #1 of 1 Archive leaderboard report
Time Series Forecasting Weather2K114 (720) MoLE-DLinear MSE 0.425 #1 of 1 Archive leaderboard report
Time Series Forecasting Weather2K114 (96) MoLE-DLinear MSE 0.391 #1 of 1 Archive leaderboard report
Time Series Forecasting Weather2K1786 (192) MoLE-RLinear MSE 0.581 #1 of 2 Archive leaderboard report
Time Series Forecasting Weather2K1786 (192) MoLE-DLinear MSE 0.601 #2 of 2 Archive leaderboard report
Time Series Forecasting Weather2K1786 (336) MoLE-DLinear MSE 0.603 #1 of 1 Archive leaderboard report
Time Series Forecasting Weather2K1786 (720) MoLE-RLinear MSE 0.628 #1 of 2 Archive leaderboard report
Time Series Forecasting Weather2K1786 (720) MoLE-DLinear MSE 0.66 #2 of 2 Archive leaderboard report
Time Series Forecasting Weather2K1786 (96) MoLE-DLinear MSE 0.535 #1 of 2 Archive leaderboard report
Time Series Forecasting Weather2K1786 (96) MoLE-RLinear MSE 0.535 #2 of 2 Archive leaderboard report
Time Series Forecasting Weather2K79 (192) MoLE-DLinear MSE 0.566 #1 of 1 Archive leaderboard report
Time Series Forecasting Weather2K79 (336) MoLE-DLinear MSE 0.546 #1 of 1 Archive leaderboard report
Time Series Forecasting Weather2K79 (720) MoLE-DLinear MSE 0.535 #1 of 1 Archive leaderboard report
Time Series Forecasting Weather2K79 (96) MoLE-DLinear MSE 0.555 #1 of 1 Archive leaderboard report
Time Series Forecasting Weather2K850 (192) MoLE-DLinear MSE 0.484 #1 of 1 Archive leaderboard report
Time Series Forecasting Weather2K850 (336) MoLE-DLinear MSE 0.474 #1 of 1 Archive leaderboard report
Time Series Forecasting Weather2K850 (720) MoLE-DLinear MSE 0.461 #1 of 1 Archive leaderboard report
Time Series Forecasting Weather2K850 (96) MoLE-RLinear MSE 0.471 #1 of 2 Archive leaderboard report
Time Series Forecasting Weather2K850 (96) MoLE-DLinear MSE 0.474 #2 of 2 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.

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