Mathematical Reasoning on MATH500 (pass@1 accuracy)
82.2Pass@1 AccuracyAERO
Evaluation Results
| Method | Links | |
|---|---|---|
| AEROBackbone Model=Qwen3-8B-Base, Evolutionary Round=R52026.02 | 82.2 | |
| R-ZeroBackbone Model=Qwen3-8B-Base2026.02 | 82 | |
| AEROBackbone Model=Qwen3-8B-Base, Evolutionary Round=R42026.02 | 81.8 | |
| AEROBackbone Model=Qwen3-8B-Base, Evolutionary Round=R32026.02 | 79.8 | |
| AEROBackbone Model=Qwen3-8B-Base, Evolutionary Round=R12026.02 | 79.4 | |
| AEROBackbone Model=Qwen3-8B-Base, Evolutionary Round=R22026.02 | 79.2 | |
| Qwen3-8B-BaseBackbone Model=Qwen3-8B-Base2026.02 | 78 | |
| Absolute ZeroBackbone Model=Qwen3-8B-Base2026.02 | 76.6 | |
| Absolute ZeroBackbone Model=Qwen3-4B-Base2026.02 | 76.2 | |
| R-ZeroBackbone Model=Qwen3-4B-Base2026.02 | 74.8 | |
| AEROBackbone Model=Qwen3-4B-Base, Evolutionary Round=R52026.02 | 74.8 | |
| AEROBackbone Model=Qwen3-4B-Base, Evolutionary Round=R42026.02 | 74.4 | |
| AEROBackbone Model=Qwen3-4B-Base, Evolutionary Round=R32026.02 | 73.8 | |
| AEROBackbone Model=Qwen3-4B-Base, Evolutionary Round=R22026.02 | 72.6 | |
| AEROBackbone Model=Qwen3-4B-Base, Evolutionary Round=R12026.02 | 70.8 | |
| Qwen3-4B-BaseBackbone Model=Qwen3-4B-Base2026.02 | 68.2 |