Mathematical Problem Solving on GSM8K
59.4Accuracy (Exact Match)
DCDM (MoE)
Updated 2mo ago
Evaluation Results
| Method | Links | |
|---|---|---|
2026.05 | 59.4 | |
2026.05 | 58.72 | |
2026.05 | 57.13 | |
2026.05 | 53.55 | |
2026.05 | 50.62 | |
2026.05 | 50.37 | |
2026.05 | 50.21 | |
2026.05 | 45.32 |
DCDM (MoE)
| Method | Links | |
|---|---|---|
2026.05 | 59.4 | |
2026.05 | 58.72 | |
2026.05 | 57.13 | |
2026.05 | 53.55 | |
2026.05 | 50.62 | |
2026.05 | 50.37 | |
2026.05 | 50.21 | |
2026.05 | 45.32 |