Cross-Task Visual In-Context Learning on Second-tier group Harmonization to Style Transfer
13.32PSNRGemini 2.5 Flash
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Gemini 2.5 FlashPrompting Strategy=Fixed2025.11 | 13.32 | 46.5 | 4.79 | |
| T2T-VICL (Gemini 2.5 Flash)Model=Gemini 2.5 Flash, Prompting Strategy=Ours2025.11 | 13.14 | 45.8 | 5.66 | |
| Seedream 4.0Prompting Strategy=Fixed2025.11 | 10.2 | 35.4 | 7.32 | |
| T2T-VICL (Seedream 4.0)Model=Seedream 4.0, Prompting Strategy=Ours2025.11 | 9.63 | 33.1 | 7.62 |