Image Editing on Image Editing Dataset (All Edits)
5.758Unchanged RegionsMLLM-as-a-Judge
Evaluation Results
| Method | Links | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| MLLM-as-a-JudgeJudge=Our Judge2026.02 | 5.758 | 5.948 | 6.517 | 6.406 | 6.204 | 5.839 | 6.276 | 6.094 | 5.87 | 6.327 | 6.283 | 6.702 | 6.236 | |
| HumanJudge=Human2026.02 | 4.769 | 5.432 | 5.327 | 5.826 | 5.667 | 5.545 | 5.668 | 5.553 | 5.687 | 5.241 | 5.376 | 5.528 | 5.652 |