Ranking on GPQA
0.922CorrelationBootstrap
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Bootstrap2025.05 | 0.922 | 55.6 | |
| Simple Average2025.05 | 0.922 | 13 | |
| Bayesian2025.05 | 0.916 | 88.9 | |
| Bradley-Terry2025.05 | 0.901 | 38.9 | |
| Single Judge2025.05 | 0.869 | 13 |
| Method | Links | ||
|---|---|---|---|
| Bootstrap2025.05 | 0.922 | 55.6 | |
| Simple Average2025.05 | 0.922 | 13 | |
| Bayesian2025.05 | 0.916 | 88.9 | |
| Bradley-Terry2025.05 | 0.901 | 38.9 | |
| Single Judge2025.05 | 0.869 | 13 |