Math Reasoning on Olympiad (Accuracy)
57.2Accuracy (Olympiad)LUFFY
Evaluation Results
| Method | Links | |
|---|---|---|
| LUFFYBase Model=Qwen2.5-Math-7B2025.07 | 57.2 | |
| Prefix-RFTBase Model=Qwen2.5-Math-7B2025.07 | 55.7 | |
| ReLIFTBase Model=Qwen2.5-Math-7B2025.07 | 52.5 | |
| UFTBase Model=Qwen2.5-Math-7B2025.07 | 51.6 | |
| SFT+RFTBase Model=Qwen2.5-Math-7B2025.07 | 50.4 | |
| RFTBase Model=Qwen2.5-Math-7B2025.07 | 46.8 | |
| SFTBase Model=Qwen2.5-Math-7B2025.07 | 43.7 | |
| Oat-ZeroBase Model=Qwen2.5-Math-7B2025.07 | 43.4 | |
| RL w/ SFT LossBase Model=Qwen2.5-Math-7B2025.07 | 39.4 | |
| SimpleRL-ZeroBase Model=Qwen2.5-Math-7B2025.07 | 34.7 | |
| Qwen2.5-Math-7B2025.07 | 15.6 |