Question Answering on HotpotQA D (eval)
46.2AccuracyBelief-SR
Evaluation Results
| Method | Links | |
|---|---|---|
| Belief-SRModel=OLMo2025.02 | 46.2 | |
| VanillaModel=OLMo2025.02 | 42.9 | |
| Knowledge-SRModel=OLMo2025.02 | 42.9 | |
| Answer-SRModel=OLMo2025.02 | 39.6 | |
| Belief-SRModel=Pythia2025.02 | 38.5 | |
| Belief-SRModel=RPJ2025.02 | 38.5 | |
| VanillaModel=RPJ2025.02 | 36.5 | |
| Knowledge-SRModel=RPJ2025.02 | 35.6 | |
| VanillaModel=Pythia2025.02 | 34.3 | |
| Answer-SRModel=RPJ2025.02 | 34.1 | |
| Knowledge-SRModel=Pythia2025.02 | 33.5 | |
| Answer-SRModel=Pythia2025.02 | 31.4 |