EHR Prediction on eICU
0.785AUROCFeatEHR-LLM
Evaluation results
| Method | Links | |
|---|---|---|
| FeatEHR-LLMFeatures=Baseline + FeatEHR-LLM2026.04 | 0.785 | |
| TFFeatures=Baseline + TF2026.04 | 0.782 | |
| ReMedTrain mode=Multi, Tokenization Strategy=LLM Align2026.03 | 0.78 | |
| GenHPFTrain mode=Multi, Tokenization Strategy=LLM Align2026.03 | 0.775 | |
| YAIB*Train mode=Single, Tokenization Strategy=Code-base2026.03 | 0.774 | |
| ReMedTrain mode=Multi, Tokenization Strategy=MLTok2026.03 | 0.773 | |
| ReMedTrain mode=Multi, Tokenization Strategy=EnTok2026.03 | 0.772 | |
| ReMedTrain mode=Single, Tokenization Strategy=EnTok2026.03 | 0.771 | |
| ReMedTrain mode=Single, Tokenization Strategy=LLM Align2026.03 | 0.771 | |
| CAAFEFeatures=Baseline + CAAFE2026.04 | 0.77 | |
| FeatLLMFeatures=Baseline + FeatLLM2026.04 | 0.769 | |
| BaselineFeatures=BL2026.04 | 0.769 | |
| ReMedTrain mode=Single, Tokenization Strategy=MLTok2026.03 | 0.767 | |
| OpenFEFeatures=Baseline + OpenFE2026.04 | 0.762 | |
| GenHPFTrain mode=Multi, Tokenization Strategy=EnTok2026.03 | 0.761 | |
| GenHPFTrain mode=Multi, Tokenization Strategy=MLTok2026.03 | 0.761 | |
| Rajikomar*Train mode=Single, Tokenization Strategy=Code-base2026.03 | 0.76 | |
| YAIB*Train mode=Multi, Tokenization Strategy=Code-base2026.03 | 0.757 | |
| GenHPFTrain mode=Single, Tokenization Strategy=EnTok2026.03 | 0.754 | |
| GenHPFTrain mode=Single, Tokenization Strategy=MLTok2026.03 | 0.754 | |
| GenHPFTrain mode=Single, Tokenization Strategy=LLM Align2026.03 | 0.754 | |
| Rajikomar*Train mode=Multi, Tokenization Strategy=Code-base2026.03 | 0.753 |