Image Editing on Image Editing Dataset (Relation)
5.833Unchanged Regions ScoreMLLM-as-a-Judge
Evaluation Results
| Method | Links | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| MLLM-as-a-JudgeJudge=Our Judge2026.02 | 5.833 | 6.167 | 6.5 | 6 | 6.667 | 5.667 | 6.333 | 6.167 | 5.667 | 6.167 | 6.167 | 6.5 | 6.153 | |
| HumanJudge=Human2026.02 | 4.992 | 5.444 | 5.714 | 6.033 | 5.728 | 5.639 | 5.683 | 5.553 | 5.578 | 5.178 | 5.556 | 5.586 | 5.557 |