Cross-Task Visual In-Context Learning on Second-tier group Light Enhancement to Colorization
22.72PSNRGemini 2.5 Flash
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Gemini 2.5 FlashPrompting Strategy=Fixed2025.11 | 22.72 | 86 | 5.05 | |
| T2T-VICL (Gemini 2.5 Flash)Model=Gemini 2.5 Flash, Prompting Strategy=Ours2025.11 | 21.04 | 75.3 | 6.86 | |
| T2T-VICL (Seedream 4.0)Model=Seedream 4.0, Prompting Strategy=Ours2025.11 | 12.09 | 42.2 | 7.28 | |
| Seedream 4.0Prompting Strategy=Fixed2025.11 | 11.01 | 43.6 | 4.59 |