Mathematics on AMC (Accuracy)
69.88AccuracyCOSE
Evaluation Results
| Method | Links | |
|---|---|---|
| COSEBackbone=Llama-3.2-3B-Instruct2026.05 | 69.88 | |
| COSEBackbone=Qwen3-4B2026.05 | 68.8 | |
| MAEBackbone=Qwen3-4B2026.05 | 65.6 | |
| R-ZeroBackbone=Qwen3-4B2026.05 | 64.9 | |
| AZRBackbone=Qwen3-4B2026.05 | 64.8 | |
| BaseBackbone=Qwen3-4B2026.05 | 62.6 | |
| MAEBackbone=Llama-3.2-3B-Instruct2026.05 | 40.6 | |
| R-ZeroBackbone=Llama-3.2-3B-Instruct2026.05 | 38.9 | |
| AZRBackbone=Llama-3.2-3B-Instruct2026.05 | 36.8 | |
| BaseBackbone=Llama-3.2-3B-Instruct2026.05 | 27.7 |