Mathematical Reasoning on GSM8K, AddSub, AQuA, MultiArith (test)
79.03GSM8K Accuracy
NARA
Updated 5d ago
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
2026.05 | 79.03 | 88.53 | 57.27 | 98.72 | 80.89 | |
2026.05 | 78.96 | 88.38 | 54.45 | 98.5 | 80.07 | |
2026.05 | 77.69 | 82.95 | 49.74 | 97.72 | 77.02 | |
2026.05 | 77.41 | 89.87 | 54.33 | 96.94 | 79.64 | |
2026.05 | 75.68 | 87.12 | 50.2 | 97.5 | 77.63 | |
2026.05 | 75.28 | 84.73 | 49.61 | 95.56 | 76.29 | |
2026.05 | 75.01 | 89.62 | 53.94 | 93.72 | 78.07 | |
2026.05 | 73.09 | 87.09 | 49.74 | 93.83 | 75.94 | |
2026.05 | 68.61 | 82.28 | 40.55 | 87.17 | 69.65 | |
2026.05 | 65.96 | 57.22 | 50.39 | 45.83 | 54.85 | |
2026.05 | 37.76 | 32.91 | 35.43 | 37.33 | 35.86 | |
2026.05 | 20.92 | 15.7 | 30.31 | 26 | 23.23 |