Wide-Baseline Correspondence Reasoning on ReasonMatch-Bench Overall 90-sample high-divergence subset
94.7F1 Score
Human
Updated 1mo ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.06 | 94.7 | - | - | |
2026.06 | 84 | - | - | |
2026.06 | 84 | 87.5 | 81.3 | |
2026.06 | 70.6 | - | - | |
2026.06 | 52 | - | - | |
2026.06 | 52 | 48.3 | 56.8 | |
2026.06 | 49.7 | - | - | |
2026.06 | 45.7 | - | - | |
2026.06 | 44.1 | - | - | |
2026.06 | 37.2 | - | - | |
2026.06 | 37.2 | 34.5 | 40.7 | |
2026.06 | 30.5 | - | - | |
2026.06 | 29.9 | - | - | |
2026.06 | 29.9 | 27.5 | 33 | |
2026.06 | 29.5 | - | - | |
2026.06 | 29.5 | 28.3 | 31.1 | |
2026.06 | 24 | - | - | |
2026.06 | 24 | 22.1 | 26.7 |