Prompt Optimization on DSG-1K
0.91DSGScoreCRAFT
Evaluation Results
| Method | Links | |
|---|---|---|
| CRAFTJudge=gpt-5-nano-2025-08-072025.12 | 0.91 | |
| MaestroJudge=Gemini 2.0 Flash2025.12 | 0.882 | |
| PromptistJudge=Gemini 2.0 Flash2025.12 | 0.849 | |
| OPT2IJudge=Gemini 2.0 Flash2025.12 | 0.838 | |
| RewriteJudge=Gemini 2.0 Flash2025.12 | 0.815 | |
| LM-BBOJudge=Gemini 2.0 Flash2025.12 | 0.806 | |
| OriginalJudge=Gemini 2.0 Flash2025.12 | 0.772 |