AI Paper Atlas

Research map / Time series / Long-term forecasting

Are Global Dependencies Necessary? Scalable Time Series Forecasting via Local Cross-Variate Modeling

Kun Liu, Renjun Jia, Ruifeng Yang, Xirui Zeng, Yuqi Liang, Cen Chen

ICLR 2026 · Time series › Long-term forecasting · approach: Graph-based multivariate forecasting

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add local cross-variate mixing (variate kernel 1 -> 3) (ablation)VPNet · Electricityavg MSE: 0.184 -> 0.171Improved
add local cross-variate mixing (variate kernel 1 -> 3) (ablation)VPNet · Trafficavg MSE: 0.443 -> 0.431Improved
add local cross-variate mixing (variate kernel 1 -> 3) (ablation)VPNet · Solar-Energyavg MSE: 0.224 -> 0.203Improved
add local cross-variate mixing (variate kernel 1 -> 3) (ablation)VPNet · ETTh1avg MSE: 0.437 -> 0.435No clear effect
widen variate kernel from 3 to 27 (ablation)VPNet · Electricityavg MSE: 0.171 -> 0.160Improved
widen variate kernel from 3 to 17 (ablation)VPNet · Weatheravg MSE: 0.250 -> 0.239Improved
widen variate kernel from 17 to 27 (ablation)VPNet · Weatheravg MSE: 0.239 -> 0.243Hurt
random vs original variate ordering (ablation)VPNet (kv=3, 2 blocks) · Electricityavg MSE: 0.171 -> 0.171No clear effect
spectral ordering instead of original (ablation)VPNet (kv=3, 2 blocks) · Electricityavg MSE: 0.171 -> 0.174No clear effect
clustered (topology-aware) ordering instead of shuffled (ablation)VPNet · Hetero-Mix (Traffic+Electricity+Weather)avg MSE: 0.388 -> 0.385Improved
original (ordered) variate ordering instead of shuffled (ablation)VPNet (kernel 27) · Electricity + 10x noiseavg MSE: 0.178 -> 0.166Improved
replace depthwise 2D convolution with local-window attention (LANet) (ablation)VPNet · Weatheravg MSE: 0.238 -> 0.250Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy