Ranking on MMLU
0.94CorrelationBayesian
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Bayesian2025.05 | 0.94 | 100 | |
| Bradley-Terry2025.05 | 0.886 | 47.4 | |
| Bootstrap2025.05 | 0.884 | 47.4 | |
| Simple Average2025.05 | 0.884 | 21.1 | |
| Single Judge2025.05 | 0.855 | 21.1 |
| Method | Links | ||
|---|---|---|---|
| Bayesian2025.05 | 0.94 | 100 | |
| Bradley-Terry2025.05 | 0.886 | 47.4 | |
| Bootstrap2025.05 | 0.884 | 47.4 | |
| Simple Average2025.05 | 0.884 | 21.1 | |
| Single Judge2025.05 | 0.855 | 21.1 |