Capability Self-Assessment on Math
88.6M-F1OLMo2-7B
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| OLMo2-7BTraining Strategy=RLVR2026.05 | 88.6 | 26.6 | 101.9 | |
| OLMo2-13BTraining Strategy=RLVR2026.05 | 88.6 | 22.73 | 100.9 | |
| OLMo2-7BTraining Strategy=SFTSelf2026.05 | 84.2 | 19.07 | 96.6 | |
| OLMo2-7BTraining Strategy=SFTTeacher2026.05 | 83.2 | 18.26 | 92.7 | |
| AverageTraining Strategy=RLVR2026.05 | 80.9 | 17.8 | 100.4 | |
| Qwen3-1.7BTraining Strategy=RLVR2026.05 | 80.4 | 17.76 | 98.8 | |
| Qwen3-4BTraining Strategy=RLVR2026.05 | 80.1 | 17.17 | 100.7 | |
| OLMo2-13BTraining Strategy=SFTTeacher2026.05 | 78.9 | 15.02 | 72.2 | |
| OLMo2-13BTraining Strategy=SFTSelf2026.05 | 77.9 | 16.06 | 49.7 | |
| Qwen3-8BTraining Strategy=RLVR2026.05 | 77.2 | 14.13 | 101.8 | |
| Qwen3-0.6BTraining Strategy=RLVR2026.05 | 76.1 | 15.26 | 98 | |
| Qwen3-8BTraining Strategy=SFTLabel2026.05 | 76 | 11.1 | 89.2 | |
| Qwen3-14BTraining Strategy=RLVR2026.05 | 75.3 | 10.94 | 100.8 | |
| OLMo2-7BTraining Strategy=Vanilla2026.05 | 73.6 | 10.87 | 100 | |
| Qwen3-14BTraining Strategy=SFTLabel2026.05 | 72.7 | 9.31 | 84.3 | |
| AverageTraining Strategy=SFTTeacher2026.05 | 70.4 | 9.4 | 88.5 | |
| Qwen3-0.6BTraining Strategy=SFTTeacher2026.05 | 70.3 | 9.75 | 76.7 | |
| AverageTraining Strategy=SFTSelf2026.05 | 70.3 | 9.85 | 85.8 | |
| Qwen3-1.7BTraining Strategy=SFTTeacher2026.05 | 68.9 | 7.86 | 73.1 | |
| Qwen3-1.7BTraining Strategy=SFTSelf2026.05 | 67.7 | 7.67 | 92 | |
| Qwen3-8BTraining Strategy=SFTTeacher2026.05 | 67.7 | 6.35 | 102.3 | |
| Qwen3-14BTraining Strategy=SFTSelf2026.05 | 67.2 | 7.05 | 98.2 | |
| OLMo2-13BTraining Strategy=Vanilla2026.05 | 66.8 | 7.24 | 100 | |
| Qwen3-8BTraining Strategy=SFTSelf2026.05 | 65.6 | 6.08 | 98.7 | |
| Qwen3-4BTraining Strategy=SFTSelf2026.05 | 64.8 | 6.14 | 95 | |
| Qwen3-0.6BTraining Strategy=SFTSelf2026.05 | 64.7 | 6.87 | 70.2 | |
| Qwen3-4BTraining Strategy=SFTTeacher2026.05 | 64.2 | 5.15 | 96.4 | |
| Qwen3-14BTraining Strategy=SFTTeacher2026.05 | 59.8 | 3.43 | 106.3 | |
| AverageTraining Strategy=Vanilla2026.05 | 55.2 | 5.98 | 100 | |
| Qwen3-1.7BTraining Strategy=Vanilla2026.05 | 53.2 | 5.15 | 100 | |
| Qwen3-4BTraining Strategy=Vanilla2026.05 | 50.7 | — | 100 | |
| Qwen3-14BTraining Strategy=Vanilla2026.05 | 50.6 | — | 100 | |
| Qwen3-8BTraining Strategy=Vanilla2026.05 | 48.8 | — | 100 | |
| AverageTraining Strategy=SFTLabel2026.05 | 47.6 | 5.26 | 30.9 | |
| Qwen3-0.6BTraining Strategy=SFTLabel2026.05 | 45.6 | 2.95 | 13.8 | |
| Qwen3-0.6BTraining Strategy=Vanilla2026.05 | 42.6 | 0.66 | 100 | |
| Qwen3-4BTraining Strategy=SFTLabel2026.05 | 40.6 | 5.51 | 13.3 | |
| Qwen3-1.7BTraining Strategy=SFTLabel2026.05 | 40.4 | 3.41 | 10.4 | |
| OLMo2-7BTraining Strategy=SFTLabel2026.05 | 29.6 | 3.62 | 2.1 | |
| OLMo2-13BTraining Strategy=SFTLabel2026.05 | 28.4 | 0.91 | 3.1 |