AI Paper Atlas

Research map / Time series / Long-term forecasting

Dynamic Relational Priming Improves Transformer in Multivariate Time Series

Hunjae Lee, Corey Clark

ICML 2026 · Time series › Long-term forecasting · approach: Spatiotemporal forecasting models · 1 citations (OpenAlex)

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
replace standard attention with prime attentionTimer-XL · WeatherMSE: 0.277 -> 0.259Improved
replace standard attention with prime attentioniTransformer · Solar-EnergyMSE: 0.238 -> 0.229Improved
replace standard attention with prime attentioniTransformer · ECL (homogeneous channels)MSE: 0.174 -> 0.175No clear effect
replace standard attention with prime attentionTimer-XL, FreDF, iTransformer · short-term forecasting (PEMS)MSE: average gain of 5.5%Improved
use prime attention with shorter look-back (L=48 or 64) instead of standard attention at L=96 (ablation)iTransformer · sequence length analysis, H=96MSE: matches standard at L=96 using 40% and 33% less inputImproved
add an extra attention head to standard attention (about 38% more parameters) (ablation)iTransformer standard attention · ETTh1, Weather, SolarMSE: ETTh1 0.386 -> 0.385; Weather 0.176 -> 0.176; Solar 0.207 -> 0.209No clear effect
add prime attention instead of extra head (ablation)iTransformer · ETTh1, Weather, SolarMSE: ETTh1 0.386 -> 0.378; Weather 0.176 -> 0.170; Solar 0.207 -> 0.199Improved
use prime attention instead of DIFF Transformer on iTransformeriTransformer · ETTh2MSE: 0.386 -> 0.377Improved
use prime attention instead of DIFF TransformeriTransformer · ECLMSE: 0.173 -> 0.175Hurt

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