Mathematical Reasoning on Olympiad (pass@1 accuracy %, Avg.)
47.1Pass@1 AccuracyrStar-Math-7B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| rStar-Math-7BCategory=Long CoT2025.10 | 47.1 | 43.6 | |
| Scaf-GRPOCategory=Long CoT2025.10 | 43.3 | 50.9 | |
| LUFFYCategory=Long CoT2025.10 | 41.7 | 46.6 | |
| OpenReasoner-ZeroCategory=Long CoT2025.10 | 41.6 | 45 | |
| Oat-ZeroCategory=Long CoT2025.10 | 41 | 46.5 | |
| SimpleRL-ZeroCategory=Long CoT2025.10 | 37.2 | 42.6 | |
| Eurus-2-7B-PRIMECategory=Long CoT2025.10 | 34.7 | 43.6 | |
| Qwen-2.5-7B-BaseCategory=Short CoT2025.10 | 34.4 | 31.3 | |
| Qwen-2.5-Math-7B-BaseCategory=Short CoT2025.10 | 18.2 | 27.5 |