EHR Multi-turn Reasoning on MIMIC Rare IV (OOD)
3.09Diagnoses
ReasoningBank
Updated 3mo ago
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
2026.01 | 3.09 | 7.83 | 5.4 | 3.49 | 25.7 | 18.59 | 10.68 | |
2026.01 | 8.63 | 7.57 | 5.5 | 5.16 | 26.45 | 21.96 | 12.55 | |
2026.01 | 9.47 | 4.46 | 5.51 | 3.61 | 28.26 | 21.23 | 12.09 | |
2026.01 | 12.48 | 6.07 | 4.47 | 5.74 | 24.92 | 18.48 | 12.03 | |
2026.01 | 16.53 | 4.44 | 5.97 | 4.65 | 25.06 | 21.37 | 13 | |
2026.01 | 18.07 | 6.09 | 5.79 | 8.57 | 28.12 | 22.82 | 14.91 | |
2026.01 | 19.5 | 5.3 | 5.19 | 9.21 | 27.19 | 24 | 15.06 | |
2026.01 | 23.25 | 7 | 5.36 | 9.33 | 29.91 | 21.94 | 16.13 |