EHR prediction on UMCdb
76.2AUROCReMed
Evaluation results
| Method | Links | |
|---|---|---|
| ReMedTrain mode=Multi, Tokenization Strategy=LLM Align2026.03 | 76.2 | |
| YAIB*Train mode=Single, Tokenization Strategy=Code-base2026.03 | 76.1 | |
| GenHPFTrain mode=Multi, Tokenization Strategy=LLM Align2026.03 | 75.9 | |
| ReMedTrain mode=Multi, Tokenization Strategy=EnTok2026.03 | 75.7 | |
| ReMedTrain mode=Single, Tokenization Strategy=LLM Align2026.03 | 75.7 | |
| ReMedTrain mode=Multi, Tokenization Strategy=MLTok2026.03 | 75.6 | |
| ReMedTrain mode=Single, Tokenization Strategy=EnTok2026.03 | 75.5 | |
| YAIB*Train mode=Multi, Tokenization Strategy=Code-base2026.03 | 75.2 | |
| ReMedTrain mode=Single, Tokenization Strategy=MLTok2026.03 | 75.1 | |
| GenHPFTrain mode=Multi, Tokenization Strategy=EnTok2026.03 | 74.8 | |
| GenHPFTrain mode=Multi, Tokenization Strategy=MLTok2026.03 | 74.6 | |
| GenHPFTrain mode=Single, Tokenization Strategy=LLM Align2026.03 | 74.5 | |
| GenHPFTrain mode=Single, Tokenization Strategy=EnTok2026.03 | 74.4 | |
| Rajikomar*Train mode=Single, Tokenization Strategy=Code-base2026.03 | 74.2 | |
| GenHPFTrain mode=Single, Tokenization Strategy=MLTok2026.03 | 73.8 | |
| Rajikomar*Train mode=Multi, Tokenization Strategy=Code-base2026.03 | 73.6 |