AI Paper Atlas

Research map / Time series / Long-term forecasting

APT: Affine Prototype-Timestamp for Time Series Forecasting Under Distribution Shift

Yujie Li, Zezhi Shao, Chengqing Yu, Yisong Fu, Tao Sun, Yongjun Xu, Fei Wang

AAAI 2026 · Time series › Long-term forecasting · approach: Attention for time series forecasting · 4 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 APT plug-inInformer+RevIN · ETTh2MSE: 0.658 -> 0.531Improved
add APT plug-inInformer+RevIN · ExchangeMSE: 0.587 -> 0.460Improved
add APT plug-iniTransformer+RevIN · ETTh1MSE: 0.485 -> 0.441Improved
add APT plug-inSparseTSF+RevIN · ETTh2MSE: 0.399 -> 0.470Hurt
add APT plug-inCATS+RevIN · ECLMSE: 0.165 -> 0.166No clear effect
add APT plug-inSparseTSF+RevIN · ETTh1MSE: 0.427 -> 0.424No clear effect
add APT plug-inSparseTSF+SAN · ETTh2MSE: 0.619 -> 0.478Improved
add pretraining losses (full APT) vs skipping APT pretraining (ablation)APT · Exchange iTransformerMSE: 0.785 -> 0.445Improved
add loss weight lambda (ablation)APT · Exchange iTransformerMSE: 0.6549 -> 0.445Improved
add load balancing loss (ablation)APT · Exchange iTransformerMSE: 0.531 -> 0.445Improved
add orthogonal loss (ablation)APT · ETTh1 InformerMSE: 1.215 -> 1.121Improved
add affine regularization loss (ablation)APT · ETTh1 InformerMSE: 1.121 -> 1.121No clear effect

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