Science Question Answering on GPQA main (test)
40.85Exact Match AccuracyMin-k
Evaluation Results
| Method | Links | |
|---|---|---|
| Min-kModel=LLaMA3-70B-Instruct, Temperature=5.02026.04 | 40.85 | |
| Min-kModel=LLaMA3-70B-Instruct, Temperature=10.02026.04 | 40.85 | |
| Min-kModel=LLaMA3-70B-Instruct, Temperature=3.02026.04 | 39.96 | |
| Top-nσModel=LLaMA3-70B-Instruct, Temperature=4.02026.04 | 39.95 | |
| Top-kModel=LLaMA3-70B-Instruct, Temperature=1.02026.04 | 39.73 | |
| Min-pModel=LLaMA3-70B-Instruct, Temperature=1.02026.04 | 39.73 | |
| Min-kModel=LLaMA3-70B-Instruct, Temperature=4.02026.04 | 39.39 | |
| Min-pModel=LLaMA3-70B-Instruct, Temperature=3.02026.04 | 39.06 | |
| Top-nσModel=LLaMA3-70B-Instruct, Temperature=5.02026.04 | 38.84 | |
| Top-nσModel=LLaMA3-70B-Instruct, Temperature=3.02026.04 | 38.62 | |
| Min-kModel=LLaMA3-70B-Instruct, Temperature=1.02026.04 | 38.62 | |
| Top-nσModel=LLaMA3-70B-Instruct, Temperature=2.02026.04 | 38.39 | |
| Top-nσModel=LLaMA3-70B-Instruct, Temperature=10.02026.04 | 38.39 | |
| Top-kModel=LLaMA3-70B-Instruct, Temperature=2.02026.04 | 38.17 | |
| Min-pModel=LLaMA3-70B-Instruct, Temperature=2.02026.04 | 37.95 | |
| Min-kModel=LLaMA3-70B-Instruct, Temperature=2.02026.04 | 37.95 | |
| Top-nσModel=LLaMA3-70B-Instruct, Temperature=1.02026.04 | 37.72 | |
| Top-pModel=LLaMA3-70B-Instruct, Temperature=2.02026.04 | 37.5 | |
| Top-pModel=LLaMA3-70B-Instruct, Temperature=1.02026.04 | 35.71 | |
| Top-kModel=LLaMA3-70B-Instruct, Temperature=3.02026.04 | 34.82 | |
| Top-pModel=LLaMA3-70B-Instruct, Temperature=3.02026.04 | 30.13 | |
| Min-pModel=LLaMA3-70B-Instruct, Temperature=4.02026.04 | 29.01 | |
| Top-kModel=Qwen3-30B-Instruct, Temperature=1.02026.04 | 26.56 | |
| Min-kModel=Qwen3-30B-Instruct, Temperature=2.02026.04 | 25.22 | |
| Min-kModel=Qwen3-30B-Instruct, Temperature=1.02026.04 | 25 | |
| Min-kModel=Qwen3-30B-Instruct, Temperature=5.02026.04 | 24.55 | |
| Min-kModel=Qwen3-30B-Instruct, Temperature=3.02026.04 | 24.33 | |
| Min-kModel=Qwen3-30B-Instruct, Temperature=4.02026.04 | 24.33 | |
| Min-kModel=Qwen3-30B-Instruct, Temperature=10.02026.04 | 24.33 | |
| Top-nσModel=Qwen3-30B-Instruct, Temperature=2.02026.04 | 23.88 | |
| Top-nσModel=Qwen3-30B-Instruct, Temperature=10.02026.04 | 23.88 | |
| Top-nσModel=Qwen3-30B-Instruct, Temperature=1.02026.04 | 23.66 | |
| Top-nσModel=Qwen3-30B-Instruct, Temperature=3.02026.04 | 23.43 | |
| Top-nσModel=Qwen3-30B-Instruct, Temperature=5.02026.04 | 23.43 | |
| Top-nσModel=Qwen3-30B-Instruct, Temperature=4.02026.04 | 23.21 | |
| Min-pModel=Qwen3-30B-Instruct, Temperature=1.02026.04 | 22.55 | |
| Top-pModel=Qwen3-30B-Instruct, Temperature=1.02026.04 | 22.54 | |
| Min-pModel=Qwen3-30B-Instruct, Temperature=2.02026.04 | 22.43 | |
| Top-kModel=Qwen3-30B-Instruct, Temperature=2.02026.04 | 21.87 | |
| Top-pModel=Qwen3-30B-Instruct, Temperature=2.02026.04 | 21.2 | |
| Top-kModel=LLaMA3-70B-Instruct, Temperature=4.02026.04 | 20.76 | |
| Min-pModel=Qwen3-30B-Instruct, Temperature=3.02026.04 | 20.11 | |
| Min-pModel=LLaMA3-70B-Instruct, Temperature=5.02026.04 | 17.19 | |
| Min-pModel=Qwen3-30B-Instruct, Temperature=4.02026.04 | 14.88 | |
| Top-kModel=Qwen3-30B-Instruct, Temperature=3.02026.04 | 14.73 | |
| Top-pModel=Qwen3-30B-Instruct, Temperature=3.02026.04 | 11.16 | |
| Top-pModel=LLaMA3-70B-Instruct, Temperature=4.02026.04 | 8.93 | |
| Min-pModel=Qwen3-30B-Instruct, Temperature=5.02026.04 | 8.56 | |
| Top-kModel=LLaMA3-70B-Instruct, Temperature=5.02026.04 | 8.26 | |
| Top-kModel=Qwen3-30B-Instruct, Temperature=4.02026.04 | 7.81 | |
| Top-pModel=LLaMA3-70B-Instruct, Temperature=5.02026.04 | 3.35 | |
| Top-kModel=Qwen3-30B-Instruct, Temperature=5.02026.04 | 3.34 | |
| Top-pModel=Qwen3-30B-Instruct, Temperature=4.02026.04 | 3.34 | |
| Min-pModel=Qwen3-30B-Instruct, Temperature=10.02026.04 | 0.51 | |
| Top-kModel=LLaMA3-70B-Instruct, Temperature=10.02026.04 | 0.22 | |
| Top-pModel=LLaMA3-70B-Instruct, Temperature=10.02026.04 | 0 | |
| Min-pModel=LLaMA3-70B-Instruct, Temperature=10.02026.04 | 0 | |
| Top-kModel=Qwen3-30B-Instruct, Temperature=10.02026.04 | 0 | |
| Top-pModel=Qwen3-30B-Instruct, Temperature=5.02026.04 | 0 | |
| Top-pModel=Qwen3-30B-Instruct, Temperature=10.02026.04 | 0 |