Abstention Classification on NaturalQA (test)
100Accuracy (Abs=0)OTC
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| OTC2025.10 | 100 | 0 | |
| Ternary2025.10 | 100 | 0 | |
| MASH w/ OTC2025.10 | 99.9 | 0.1 | |
| AFHmode=Multisample2025.10 | 87.9 | 52.1 | |
| MASH w/ EXP2025.10 | 85.6 | 62.7 | |
| MASH w/ OTC-ST2025.10 | 85.4 | 66.1 | |
| DPO2025.10 | 84.5 | 71.6 | |
| AFHmode=Absolute2025.10 | 67.7 | 48.1 | |
| Promptingshots=5-shot2025.10 | 60.2 | 44.6 |