Question Answering on HotpotQA (train)
100AccuracyVanilla
Evaluation Results
| Method | Links | |
|---|---|---|
| VanillaModel=OLMo2025.02 | 100 | |
| VanillaModel=Pythia2025.02 | 100 | |
| VanillaModel=RPJ2025.02 | 100 | |
| Belief-SRModel=OLMo2025.02 | 96.1 | |
| Answer-SRModel=OLMo2025.02 | 93.9 | |
| Belief-SRModel=Pythia2025.02 | 91 | |
| Knowledge-SRModel=OLMo2025.02 | 89.6 | |
| Answer-SRModel=Pythia2025.02 | 89.4 | |
| Belief-SRModel=RPJ2025.02 | 89.4 | |
| Knowledge-SRModel=Pythia2025.02 | 85.6 | |
| Answer-SRModel=RPJ2025.02 | 85.1 | |
| Knowledge-SRModel=RPJ2025.02 | 84 |