Output OOD Detection on Samsum
99.99AUROCAP-OOD
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| AP-OODBackbone=PEGASUS_LARGE, ID Dataset=XSUM2026.02 | 99.99 | 0 | |
| Binary logitsBackbone=PEGASUS_LARGE, ID Dataset=XSUM2026.02 | 99.95 | 4 | |
| Deep SADBackbone=PEGASUS_LARGE, ID Dataset=XSUM2026.02 | 99.95 | 8 | |
| Relative MahalanobisBackbone=PEGASUS_LARGE, ID Dataset=XSUM2026.02 | 99.52 | 95 | |
| AP-OODBackbone=PEGASUS-LARGE, ID Dataset=XSUM, Evaluation Protocol=Output OOD2026.02 | 98.89 | 4.09 | |
| KNNBackbone=PEGASUS-LARGE, ID Dataset=XSUM, Evaluation Protocol=Output OOD2026.02 | 97.28 | 10.99 | |
| MahalanobisBackbone=PEGASUS-LARGE, ID Dataset=XSUM, Evaluation Protocol=Output OOD2026.02 | 96.99 | 15.25 | |
| Deep SVDDBackbone=PEGASUS-LARGE, ID Dataset=XSUM, Evaluation Protocol=Output OOD2026.02 | 95.61 | 21.66 | |
| EntropyBackbone=PEGASUS-LARGE, ID Dataset=XSUM, Evaluation Protocol=Output OOD2026.02 | 87.07 | 50.83 | |
| PerplexityBackbone=PEGASUS-LARGE, ID Dataset=XSUM, Evaluation Protocol=Output OOD2026.02 | 77.9 | 47.35 |