Unsupervised OOD detection on The Pile ID MIMIC-III Clinical OOD (test)
86.44AUROCAP-OOD
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| AP-OODBase Model=Pythia-160M, ID Dataset=The Pile, OOD Type=Input OOD2026.02 | 86.44 | 57.38 | |
| PerplexityBase Model=Pythia-160M, ID Dataset=The Pile, OOD Type=Input OOD2026.02 | 85.86 | 65.22 | |
| MahalanobisBase Model=Pythia-160M, ID Dataset=The Pile, OOD Type=Input OOD2026.02 | 75.67 | 90.79 | |
| KNNBase Model=Pythia-160M, ID Dataset=The Pile, OOD Type=Input OOD2026.02 | 75.56 | 95.85 | |
| Deep SVDDBase Model=Pythia-160M, ID Dataset=The Pile, OOD Type=Input OOD2026.02 | 73.44 | 96.35 |