Visual Consistency Assessment on EditReward-Bench 2025 (test)
80.88Subject Addition Accuracy
Gemini 3 Pro
Updated 19d ago
Evaluation Results
| Method | Links | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2026.03 | 80.88 | 83.56 | 90.57 | 91.3 | 90.16 | 88.89 | 86.67 | 92 | 78.95 | 94.12 | 84.93 | 87.84 | 85.48 | 87.33 | |
2026.03 | 80.88 | 63.01 | 77.36 | 86.96 | 88.52 | 87.5 | 83.33 | 83 | 50.88 | 89.71 | 75.34 | 75.68 | 80.65 | 78.68 | |
2026.03 | 77.94 | 67.12 | 84.91 | 94.2 | 83.61 | 86.11 | 80 | 87 | 71.93 | 89.71 | 87.67 | 82.43 | 79.03 | 82.44 | |
2026.03 | 74.38 | 89.9 | 81.55 | 83.49 | 67 | 74.39 | 72.61 | 77.15 | 93.66 | 69.02 | 84.38 | 75.83 | 75 | 78.34 | |
2026.03 | 72.06 | 61.64 | 73.58 | 72.46 | 75.41 | 76.39 | 78.33 | 74 | 56.14 | 85.29 | 65.75 | 74.32 | 74.19 | 72.27 | |
2026.03 | 57.65 | 51.91 | 54.85 | 77.83 | 51.23 | 61.38 | 55.65 | 63.25 | 76.06 | 29.41 | 51.56 | 58.75 | 62.26 | 57.83 |