Mathematical Reasoning on DAPO-17k
26.7AIME 2024 ScorePCL
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| PCLModel=Qwen3-8B Base2026.05 | 26.7 | 10 | 76.6 | 28.3 | 35.4 | 17.3 | 46.9 | |
| No Curr.Model=Qwen3-8B Base2026.05 | 23.3 | 10 | 78.2 | 27.9 | 34.9 | 32.6 | — | |
| METISModel=Qwen3-8B Base2026.05 | 20 | 16.7 | 78.2 | 27.9 | 35.7 | 15.7 | 51.8 | |
| ADCLModel=Qwen3-8B Base2026.05 | 16.7 | 16.7 | 74.6 | 26 | 33.5 | 36.8 | 12.9 | |
| PCLModel=Qwen3-4B Base2026.05 | 16.7 | 6.7 | 65 | 18.9 | 26.8 | 11.8 | 12.6 | |
| METISModel=Qwen3-4B Base2026.05 | 16.7 | 10 | 64.8 | 20.2 | 27.9 | 8.4 | 37.8 | |
| SECModel=Llama-3.1 8B-Instruct2026.05 | 16.7 | 6.7 | 50.4 | 20.2 | 23.5 | 3.1 | 51.6 | |
| ADCLModel=Llama-3.1 8B-Instruct2026.05 | 16.7 | 3.3 | 49.6 | 19.1 | 22.2 | 25.6 | 300 | |
| PCLModel=Llama-3.1 8B-Instruct2026.05 | 16.7 | 6.7 | 48 | 21.3 | 23.2 | 17.1 | 167.2 | |
| METISModel=Llama-3.1 8B-Instruct2026.05 | 16.7 | 10 | 50.2 | 17.3 | 23.6 | 5.2 | 18.8 | |
| SECModel=Qwen3-8B Base2026.05 | 13.3 | 16.7 | 78.2 | 27.9 | 34 | 23 | 29.4 | |
| No Curr.Model=Qwen3-4B Base2026.05 | 13.3 | 6.7 | 64.8 | 20.9 | 26.4 | 13.5 | — | |
| ADCLModel=Qwen3-4B Base2026.05 | 13.3 | 6.7 | 65.6 | 17.6 | 25.8 | 18.6 | 37.8 | |
| No Curr.Model=Llama-3.1 8B-Instruct2026.05 | 13.3 | 3.3 | 50 | 18.8 | 21.4 | 6.4 | — | |
| SECModel=Qwen3-4B Base2026.05 | 10 | 3.3 | 64.4 | 17.3 | 23.8 | 7.8 | 42.2 |