Visual Question Answering (Edit Inspection) on EditInspector 1.0 (test)
70.3Accuracy
Gemini-1.5
Updated 4d ago
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
2025.06 | 70.3 | 51.1 | 46.3 | 58.5 | 66.3 | |
2025.06 | 70.2 | 49.2 | 46.7 | 49.8 | 58.2 | |
2025.06 | 67.8 | 55.7 | 55 | 65.7 | 64.3 | |
2025.06 | 67.7 | 49.5 | 48.4 | 50 | 58.2 | |
2025.06 | 67.3 | 50.4 | 53.7 | 50.7 | 63.9 | |
2025.06 | 67.2 | - | - | 51.7 | 54.5 | |
2025.06 | 66.9 | 48.2 | 49.3 | 52.8 | 64 | |
2025.06 | 58.9 | 52 | 50.1 | 47.6 | 50 | |
2025.06 | 49.9 | 50.4 | 50.1 | 49.4 | 53.9 |