AI Paper Atlas

Research map / Time series / Causal and generative time series

Online time series prediction using feature adjustment

Xiannan Huang, Shuhan Qiu, Jiayuan Du, Chao Yang

ICLR 2026 · Time series › Causal and generative time series · approach: Robust forecasting and evaluation

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
Adapt features with ADAPT-Z adapter instead of updating parametersBase forecasters · ETTm1MSE: 0.2211 -> 0.1937Improved
Feature-space online gradient descent (fOGD) instead of plain OGDBase forecasters · 13 datasetsMSE: ETTm1 0.2180 -> 0.2178Improved
Add historical gradient input to adapter (ablation)ADAPT-Z (SOFTS) · ETTm1MSE: 0.1862 -> 0.1857Improved
Add current feature input to adapter (ablation)ADAPT-Z (SOFTS) · solarMSE: 0.0839 -> 0.0791Improved
Fine-tune base model and adapter on training set, keep online updating (Version1)ADAPT-Z (iTransformer) · solarMSE: 0.0940 -> 0.0885Improved
Freeze all parameters at deployment after training-set finetuning (Version2)ADAPT-Z (iTransformer) · ETTm1MSE: 0.2309 -> 0.2104Improved
Adapt raw input instead of internal features (ablation)iTransformer · feature location averagemean MSE: 0.1150 -> 0.1315Hurt
Use output of first transformer block as features (ablation)iTransformer · feature location averagemean MSE: 0.1150 -> 0.1041Improved
Combine DishTS normalization with ADAPT-ZiTransformer · ETTm1MSE: 0.2205 -> 0.1914Improved
Add DishTS normalization aloneiTransformer · exchangeMSE: 0.0394 -> 0.0792Hurt
Zero epochs of adapter tuning on validation set (ablation)ADAPT-Z (SOFTS) · 13 datasetsMSEImproved

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