AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

GREAT: Generalizable Representation Enhancement via Auxiliary Transformations for Zero-Shot Environmental Prediction

Shiyuan Luo, Chonghao Qiu, Runlong Yu, Yiqun Xie, Xiaowei Jia

AAAI 2026 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Zero-shot robustness and transfer

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
use GREAT instead of DANNLSTM, Single ref 1% · Stream temperature zero-shotRMSE: 4.6434 -> 3.8306Improved
use GREAT instead of DANNLSTM, Multi ref 0.01% · Stream temperature zero-shotRMSE: 4.6090 -> 4.1555Improved
use GREAT instead of plain LSTM trained on source onlyLSTM · Stream temperature zero-shotRMSE: 6.2475 -> 3.8306Improved
make reference labels sparser (1% to 0.01%)GREAT, Single ref · Stream temperature zero-shotRMSE: 3.8306 -> 3.9026No clear effect
make reference labels sparser (1% to 0.01%)DroughtSet, Single ref · Stream temperature zero-shotRMSE: 6.8626 -> 7.2275Hurt
add feature transformation modules (ablation)GREAT · Stream temperature zero-shotRMSEImproved
add hidden-state transformation on top of input transformation (ablation)GREAT · Stream temperature zero-shotRMSEImproved
add pre-training (ablation)GREAT · Stream temperature zero-shotRMSEImproved
add bi-level training instead of joint training (ablation)GREAT, Single setting · Stream temperature zero-shotRMSEImproved
train other models on GREAT-augmented data (ablation)Transformer, Informer, TimeMixer, TimesNet · Stream temperature zero-shotRMSEImproved
use Lens learned augmentationLens · Stream temperature zero-shotRMSE: 6.2475 -> 7.5231Hurt

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