AI Paper Atlas

Research map / Time series / Long-term forecasting

Beyond Point Predictions: Manifold Expansion and Dual Alignment for Robust Time Series Distillation

Junyao Hong, Zesheng Lai, Xinyi Xiao, Suyang Zhou, Aodong Shen, Youyong Kong

ICML 2026 · Time series › Long-term forecasting · approach: Spatiotemporal forecasting models

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 DSD distillation from PatchTSTLMP-Net · ETTh1MSE: 0.452 -> 0.430Improved
add DSD distillationLMP-Net · WeatherMSE: 0.254 -> 0.246Improved
add DSD distillationLMP-Net · TrafficMSE: 0.490 -> 0.479Improved
add manifold expansion (Expand-Evolve-Contract) (ablation)DSD LMP-Net · TrafficMSE: 0.615 -> 0.479Improved
add manifold expansion (ablation)DSD LMP-Net · ElectricityMSE: 0.211 -> 0.179Improved
add manifold expansion (ablation)DSD LMP-Net · ETTh1MSE: 0.447 -> 0.430Improved
add dual manifold alignment (ablation)DSD LMP-Net · ElectricityMSE: 0.183 -> 0.179Mixed
add dual manifold alignment (ablation)DSD LMP-Net · TrafficMSE: 0.491 -> 0.479No clear effect
add regime-aware prediction distillation (PredKD) (ablation)DSD LMP-Net · TrafficMSE: 0.491 -> 0.479Improved
add PredKD (ablation)DSD LMP-Net · ETTh1MSE: 0.446 -> 0.430Improved
distill from Crossformer teacherLMP-Net · ETTh1MSE: 0.452 -> 0.444Improved
distill from Pathformer teacherLMP-Net · ElectricityMSE: 0.182 -> 0.182No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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