Mathematical Reasoning on Minerva Math (Accuracy, Generation Length)
68.4Accuracy
BCR-Qwen3-4B (Ours)
Updated 9d ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.04 | 68.4 | 3,338 | |
2026.04 | 66.2 | 4,576 | |
2026.04 | 57.7 | 5,714 | |
2026.04 | 54.4 | 5,201 | |
2026.04 | 49.1 | 4,205 | |
| 48.5 | 7,575 | ||
2026.04 | 48.5 | 2,494 | |
2026.04 | 46.7 | 5,678 | |
2026.04 | 45.6 | 3,806.2 | |
2026.04 | 44.9 | 2,593.1 | |
2026.04 | 43.8 | 3,027.3 | |
2026.04 | 43.8 | 2,742.5 | |
2026.04 | 43.4 | 5,413 | |
2026.04 | 43.4 | 2,429.9 | |
2026.04 | 40.8 | 816.7 | |
2026.04 | 40.4 | 1,957.3 | |
2026.04 | 40.4 | 700.8 | |
2026.04 | 40.1 | 948.6 | |
2026.01 | 39.7 | 3,909 | |
2026.04 | 39.3 | 965.9 | |
2026.04 | 39 | 742.2 | |
2026.01 | 38.2 | 1,504 | |
2026.04 | 37.9 | 815.2 | |
2026.01 | 37.5 | 1,761 | |
2026.01 | 36.8 | 1,024 | |
2026.04 | 34.6 | 832.7 | |
2026.01 | 33.8 | 1,181 | |
2026.04 | 33.5 | 1,522.3 | |
2026.04 | 33.5 | 1,059.5 | |
2026.04 | 33.1 | 1,168.4 | |
2026.01 | 32.7 | 956 | |
2025.12 | 32.08 | - | |
2026.04 | 31.6 | 1,733.2 | |
2026.04 | 30.9 | 4,903.2 | |
2026.01 | 30.7 | 4,948 | |
2025.12 | 30.24 | - | |
2026.04 | 30.1 | 1,657.1 | |
2026.04 | 30.1 | 1,026.5 | |
2026.04 | 30.1 | 1,044.2 | |
2026.04 | 29 | 853.6 | |
2026.04 | 28.3 | 1,673.2 | |
2026.04 | 27.9 | 3,849 | |
2026.04 | 27.9 | 1,263.9 | |
2026.01 | 27.6 | 862 | |
2026.04 | 27.6 | 857.9 | |
2026.04 | 27.2 | 2,638.1 | |
2025.12 | 26.93 | - | |
2026.01 | 23.5 | 3,021 | |
2026.04 | 23.5 | 777 | |
2026.01 | 22.1 | 872 | |
| 20.2 | 3,014.8 | ||
2026.04 | 16.5 | 1,315.8 | |
2026.01 | 11.4 | 1,037 | |
2026.04 | 11 | 1,032.2 |