Instruction Following on PDD
95.4PDD Instruction Following ScoreQwen2.5-7B + GAPO
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen2.5-7B + GAPOPrompt=Naive Prompt2025.03 | 95.4 | |
| Qwen2.5-7B + PPOPrompt=Naive Prompt2025.03 | 89.4 | |
| Qwen2.5-7B + SFTPrompt=Naive Prompt2025.03 | 70.1 | |
| Qwen2.5-7B + KTOPrompt=Naive Prompt2025.03 | 64.5 | |
| Qwen2.5-7BPrompt=Plan-X-Solve2025.03 | 62.8 | |
| Qwen2.5-7BPrompt=Naive Prompt2025.03 | 61.4 | |
| Qwen2.5-7BPrompt=CoT2025.03 | 58.4 | |
| Qwen2.5-7B + ORPOPrompt=Naive Prompt2025.03 | 21.4 | |
| Qwen2.5-7B + DPOPrompt=Naive Prompt2025.03 | 12.5 | |
| Qwen2.5-7B + SimPOPrompt=Naive Prompt2025.03 | 5.3 |