Text Generation on XSum
25.76F1 ScorePrompt-R1
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Prompt-R12025.11 | 25.76 | 63.02 | |
| TextGradCategory=APO (GPT-4o-mini)2025.11 | 24.88 | 28.45 | |
| BaselineBackbone=GPT-4o-mini2025.11 | 24.35 | 60.56 | |
| GRPOBackbone=Qwen3-4B2025.11 | 21.98 | 52.09 | |
| SFTBackbone=Qwen3-4B2025.11 | 21.13 | 55.65 | |
| OPROCategory=APO (GPT-4o-mini)2025.11 | 17.87 | 29.47 | |
| BaselineBackbone=Qwen3-4B2025.11 | 16.75 | 31.73 | |
| CoT ReasoningBackbone=GPT-4o-mini2025.11 | 8.92 | 53.94 | |
| CoT ReasoningBackbone=Qwen3-4B2025.11 | 8.33 | 6.81 | |
| GEPACategory=APO (GPT-4o-mini)2025.11 | 0.21 | 0.37 |