Human Preference Evaluation on GenBlemish-27K (test)
48.8Preference Share: Significantly BetterAgentic Retoucher
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Agentic Retoucher2026.01 | 48.8 | 34.4 | 10.2 | 5.8 | 0.8 | |
| Baseline2026.01 | 4.2 | 22.8 | 60.8 | 9.2 | 3 |
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Agentic Retoucher2026.01 | 48.8 | 34.4 | 10.2 | 5.8 | 0.8 | |
| Baseline2026.01 | 4.2 | 22.8 | 60.8 | 9.2 | 3 |