Mathematical Reasoning on GSM8K (Drop in Utility)
36.7Drop in Utility
KL
Updated 5d ago
Evaluation Results
| Method | Links | |
|---|---|---|
2026.04 | 36.7 | |
2026.04 | 5.3 | |
2026.04 | 1.9 | |
2026.04 | -6.2 | |
2026.04 | -8.8 | |
| -12.7 |
KL
| Method | Links | |
|---|---|---|
2026.04 | 36.7 | |
2026.04 | 5.3 | |
2026.04 | 1.9 | |
2026.04 | -6.2 | |
2026.04 | -8.8 | |
| -12.7 |