AI Paper Atlas

Research map / Time series / Long-term forecasting

HMformer: Unleashing Transformer’s Potential for Time Series Forecasting via Hierarchical Multi-Scale Modeling

Renjun Huang, Han Xiao, Bingqing Li, Baili Zhang, Jianhua Lyu

AAAI 2026 · Time series › Long-term forecasting · approach: Transformers for long-term forecasting · 0 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
add cross-scale mixing (ablation)HMformer · TrafficMSE: 0.429 -> 0.420Improved
add cross-scale mixing (ablation)HMformer · ETTh1MSE: 0.428 -> 0.422Improved
add cross-scale mixing (ablation)HMformer · ETTh2MSE: 0.370 -> 0.363Improved
use Scale-Adaptive Feature Expansion instead of uniform dk (ablation)HMformer · TrafficMSE: 0.425 -> 0.420Improved
use SAFE instead of uniform dk=32 (ablation)HMformer · ETTh2MSE: 0.371 -> 0.363Improved
add predictions from all branches (ablation)HMformer · TrafficMSE: 0.442 -> 0.420Improved
add second-branch prediction to first (ablation)HMformer · ETTh1MSE: 0.429 -> 0.422Improved
sum predictions from deeper branches (ablation)HMformer · TrafficMSE: 0.423 -> 0.420Improved
HMformer versus best baseline on ECLHMformer · ECLMSE: 0.178 -> 0.183Hurt
HMformer versus TimeMixer on WeatherHMformer · WeatherMSE: 0.240 -> 0.252Hurt
HMformer versus best baselineHMformer · TrafficMSE: 0.428 -> 0.420Improved
HMformer versus TimesNetHMformer · M4OWA: 0.851 -> 0.848Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 4 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