Vision-language reasoning on MathVista (test)
34.6Accuracy
Top Probability + Confidence Modulation
Updated 6d ago
Evaluation Results
| Method | Links | |
|---|---|---|
| 34.6 | ||
| 33.2 | ||
2026.05 | 29.2 | |
| 29.1 | ||
2026.05 | 28.8 | |
2026.05 | 27 | |
2026.05 | 24.8 |
Top Probability + Confidence Modulation
| Method | Links | |
|---|---|---|
| 34.6 | ||
| 33.2 | ||
2026.05 | 29.2 | |
| 29.1 | ||
2026.05 | 28.8 | |
2026.05 | 27 | |
2026.05 | 24.8 |