Multimodal Mathematical Reasoning on WeMath (test)
72.15Accuracy
GSPO + KAWHI (Qwen3)
Updated 2mo ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.03 | 72.15 | 55.54 | |
2026.03 | 71.55 | 54.14 | |
2026.03 | 71.23 | 53.06 | |
2026.03 | 70.98 | 51.98 | |
2026.03 | 70.31 | 51.17 | |
2026.03 | 69.67 | 51.8 | |
2026.03 | 69.43 | 50.53 | |
2026.03 | 59.71 | 52.94 | |
2026.03 | 59.02 | 51.4 | |
2026.03 | 58.21 | 51.85 | |
2026.03 | 58.1 | 50.18 | |
2026.03 | 57.98 | 50.79 | |
2026.03 | 57.84 | 51 | |
2026.03 | 57.12 | 50.29 | |
2026.03 | 56.78 | 49.37 | |
2026.03 | 55.32 | 47.82 | |
2026.03 | 54.32 | 47.68 |