Uncertainty Estimation on GSM8K (1,000 samples, val)
0.668AUROC
Internal Confidence
Updated 1mo ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2025.06 | 0.668 | 28.2 | 5.7 | |
2025.06 | 0.664 | 32.6 | - | |
2025.06 | 0.661 | 28 | - | |
2025.06 | 0.657 | 34.9 | 3.1 | |
2025.06 | 0.64 | 32.3 | 32.4 | |
2025.06 | 0.636 | 25.7 | - | |
2025.06 | 0.63 | 30.9 | - | |
2025.06 | 0.613 | 18.5 | 33.5 | |
2025.06 | 0.604 | 17.9 | - | |
2025.06 | 0.575 | 13.2 | - | |
2025.06 | 0.561 | 13.5 | - | |
2025.06 | 0.55 | 11.3 | - | |
2025.06 | 0.55 | 11.7 | 6.4 | |
2025.06 | 0.549 | 3.1 | - | |
2025.06 | 0.543 | 13.5 | - | |
2025.06 | 0.54 | 7.3 | - | |
2025.06 | 0.539 | 6.4 | 19.9 | |
2025.06 | 0.536 | 6.9 | - | |
2025.06 | 0.533 | 9.4 | 11.3 | |
2025.06 | 0.533 | 10.4 | - | |
2025.06 | 0.532 | 4.3 | - | |
2025.06 | 0.527 | 6.7 | 45.9 | |
2025.06 | 0.526 | 5.2 | 11.9 | |
2025.06 | 0.524 | 4.4 | - | |
2025.06 | 0.505 | 9.3 | 25.4 | |
2025.06 | 0.479 | -2.6 | 35.2 | |
2025.06 | 0.475 | -4.6 | 33.1 |