Mathematical Reasoning on GSM8K (Exact Match %)
93.63Exact Match Accuracy (GSM8K)Min-k
Evaluation Results
| Method | Links | |
|---|---|---|
| Min-kModel=Qwen3-4B-Instruct, Temperature=1.02026.04 | 93.63 | |
| Top-pModel=Qwen3-4B-Instruct, Temperature=2.02026.04 | 93.56 | |
| Min-kModel=Qwen3-4B-Instruct, Temperature=5.02026.04 | 93.56 | |
| Top-nσModel=Qwen3-4B-Instruct, Temperature=1.02026.04 | 93.4 | |
| Top-kModel=Qwen3-4B-Instruct, Temperature=1.02026.04 | 93.33 | |
| Min-pModel=Qwen3-4B-Instruct, Temperature=1.02026.04 | 93.18 | |
| Top-nσModel=Qwen3-4B-Instruct, Temperature=10.02026.04 | 93.18 | |
| Min-kModel=Qwen3-4B-Instruct, Temperature=4.02026.04 | 93.18 | |
| Min-kModel=Qwen3-4B-Instruct, Temperature=2.02026.04 | 93.1 | |
| Min-kModel=Qwen3-4B-Instruct, Temperature=10.02026.04 | 93.1 | |
| Top-nσModel=Qwen3-4B-Instruct, Temperature=2.02026.04 | 93.03 | |
| Top-nσModel=Qwen3-4B-Instruct, Temperature=3.02026.04 | 92.87 | |
| Top-nσModel=Qwen3-4B-Instruct, Temperature=4.02026.04 | 92.87 | |
| Top-pModel=Qwen3-4B-Instruct, Temperature=1.02026.04 | 92.8 | |
| Min-pModel=Qwen3-4B-Instruct, Temperature=2.02026.04 | 92.8 | |
| Top-kModel=Qwen3-4B-Instruct, Temperature=2.02026.04 | 92.72 | |
| Top-nσModel=Qwen3-4B-Instruct, Temperature=5.02026.04 | 92.72 | |
| Min-kModel=Qwen3-4B-Instruct, Temperature=3.02026.04 | 92.65 | |
| Top-kModel=Qwen3-4B-Instruct, Temperature=3.02026.04 | 92.49 | |
| Top-pModel=Qwen3-4B-Instruct, Temperature=3.02026.04 | 91.74 | |
| Min-pModel=Qwen3-4B-Instruct, Temperature=3.02026.04 | 91.74 | |
| Min-pModel=Qwen3-4B-Instruct, Temperature=4.02026.04 | 89.92 | |
| Top-kModel=Qwen3-4B-Instruct, Temperature=4.02026.04 | 89.61 | |
| Top-pModel=Qwen3-4B-Instruct, Temperature=4.02026.04 | 88.76 | |
| Min-pModel=Qwen3-4B-Instruct, Temperature=5.02026.04 | 85.14 | |
| Top-kModel=Qwen3-4B-Instruct, Temperature=5.02026.04 | 79.08 | |
| Min-kModel=LLaMA3-8B-Instruct, Temperature=1.02026.04 | 77.39 | |
| Min-kModel=LLaMA3-8B-Instruct, Temperature=2.02026.04 | 76.65 | |
| Min-kModel=LLaMA3-8B-Instruct, Temperature=5.02026.04 | 76.48 | |
| Top-pModel=LLaMA3-8B-Instruct, Temperature=1.02026.04 | 76.27 | |
| Min-kModel=LLaMA3-8B-Instruct, Temperature=4.02026.04 | 76.15 | |
| Min-kModel=LLaMA3-8B-Instruct, Temperature=3.02026.04 | 76.02 | |
| Top-kModel=LLaMA3-8B-Instruct, Temperature=1.02026.04 | 75.44 | |
| Top-nσModel=LLaMA3-8B-Instruct, Temperature=1.02026.04 | 75.44 | |
| Min-pModel=LLaMA3-8B-Instruct, Temperature=1.02026.04 | 75.36 | |
| Min-kModel=LLaMA3-8B-Instruct, Temperature=10.02026.04 | 74.79 | |
| Top-nσModel=LLaMA3-8B-Instruct, Temperature=5.02026.04 | 74.07 | |
| Top-nσModel=LLaMA3-8B-Instruct, Temperature=10.02026.04 | 73.77 | |
| Top-nσModel=LLaMA3-8B-Instruct, Temperature=2.02026.04 | 73.69 | |
| Top-nσModel=LLaMA3-8B-Instruct, Temperature=4.02026.04 | 73.09 | |
| Top-nσModel=LLaMA3-8B-Instruct, Temperature=3.02026.04 | 72.55 | |
| Min-pModel=LLaMA3-8B-Instruct, Temperature=2.02026.04 | 62.7 | |
| Top-pModel=Qwen3-4B-Instruct, Temperature=5.02026.04 | 56.25 | |
| Top-kModel=LLaMA3-8B-Instruct, Temperature=2.02026.04 | 54.13 | |
| Top-pModel=LLaMA3-8B-Instruct, Temperature=2.02026.04 | 46.63 | |
| Min-pModel=LLaMA3-8B-Instruct, Temperature=3.02026.04 | 28.81 | |
| Top-kModel=LLaMA3-8B-Instruct, Temperature=3.02026.04 | 9.86 | |
| Min-pModel=LLaMA3-8B-Instruct, Temperature=4.02026.04 | 3.94 | |
| Top-pModel=LLaMA3-8B-Instruct, Temperature=3.02026.04 | 2.5 | |
| Top-kModel=LLaMA3-8B-Instruct, Temperature=4.02026.04 | 1.06 | |
| Min-pModel=Qwen3-4B-Instruct, Temperature=10.02026.04 | 0.45 | |
| Min-pModel=LLaMA3-8B-Instruct, Temperature=5.02026.04 | 0.38 | |
| Top-pModel=LLaMA3-8B-Instruct, Temperature=4.02026.04 | 0.15 | |
| Top-kModel=LLaMA3-8B-Instruct, Temperature=5.02026.04 | 0.08 | |
| Top-kModel=LLaMA3-8B-Instruct, Temperature=10.02026.04 | 0 | |
| Top-pModel=LLaMA3-8B-Instruct, Temperature=5.02026.04 | 0 | |
| Top-pModel=LLaMA3-8B-Instruct, Temperature=10.02026.04 | 0 | |
| Min-pModel=LLaMA3-8B-Instruct, Temperature=10.02026.04 | 0 | |
| Top-kModel=Qwen3-4B-Instruct, Temperature=10.02026.04 | 0 | |
| Top-pModel=Qwen3-4B-Instruct, Temperature=10.02026.04 | 0 |