Cross-Task Visual In-Context Learning on Second-tier group Inpainting to Colorization
21.31PSNRGemini 2.5 Flash
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Gemini 2.5 FlashPrompting Strategy=Fixed2025.11 | 21.31 | 84.3 | 5.23 | |
| T2T-VICL (Gemini 2.5 Flash)Model=Gemini 2.5 Flash, Prompting Strategy=Ours2025.11 | 20.75 | 74.2 | 7.39 | |
| Seedream 4.0Prompting Strategy=Fixed2025.11 | 13.53 | 44.7 | 5.86 | |
| T2T-VICL (Seedream 4.0)Model=Seedream 4.0, Prompting Strategy=Ours2025.11 | 12.08 | 39.7 | 6.96 |