Math Reasoning on AQuA, GSM8K, SVAMP, MAWPS, SingleEq
30.31Accuracy (AQuA)
LoRA
Updated 1mo ago
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
2025.09 | 30.31 | 66.26 | 75.1 | 90.34 | 94.88 | 71.38 | -5.21 | |
2025.09 | 29.53 | 67.17 | 77.4 | 89.5 | 94.49 | 71.62 | -5.31 | |
2025.09 | 28.74 | 67.4 | 77.1 | 89.08 | 94.06 | 71.28 | -4.54 | |
2025.09 | 28.34 | 63.68 | 71.1 | 86.13 | 90.75 | 68 | - | |
2025.09 | 27.56 | 63.84 | 76.7 | 90.76 | 94.88 | 70.75 | -3.06 | |
2025.09 | 27.17 | 68.76 | 75.6 | 90.34 | 93.31 | 71.04 | -3.58 | |
2025.09 | 25.2 | 59.44 | 72.6 | 86.55 | 91.93 | 67.14 | 2.77 |