List your company

Mathematical Reasoning on GSM8K (Answer Accuracy, Mode Accuracy, & Generation Count)

97.2Answer AccuracyL = 2 (known prior)
94.18494.96795.7596.533Feb 5, 2026

Evaluation results

MethodLinks
L = 2 (known prior)Model=Qwen-2.5-72B, 1 - delta=0.9752026.0297.299.82.19
L = 2 (known prior)Model=Qwen-2.5-72B, 1 - delta=0.992026.0297.299.82.51
L = 3 (known prior)Model=Qwen-2.5-72B, 1 - delta=0.9752026.0297.299.82.15
L = 3 (known prior)Model=Qwen-2.5-72B, 1 - delta=0.992026.0297.299.82.44
L = 2 (uncertain prior)Model=Qwen-2.5-72B, 1 - delta=0.992026.0297.21003.37
L = 3 (uncertain prior)Model=Qwen-2.5-72B, 1 - delta=0.992026.0297.21002.95
ASCModel=Qwen-2.5-72B, 1 - delta=0.952026.0297.21004.58
ASCModel=Qwen-2.5-72B, 1 - delta=0.9752026.0297.21005.61
ASCModel=Qwen-2.5-72B, 1 - delta=0.992026.0297.21006.91
L = 2 (known prior)Model=LLaMA-3.1-405B, 1 - delta=0.992026.0297.21002.31
L = 3 (known prior)Model=LLaMA-3.1-405B, 1 - delta=0.992026.0297.21002.25
L = 2 (uncertain prior)Model=LLaMA-3.1-405B, 1 - delta=0.9752026.0297.299.72.49
L = 2 (uncertain prior)Model=LLaMA-3.1-405B, 1 - delta=0.992026.0297.299.72.7
L = 3 (uncertain prior)Model=LLaMA-3.1-405B, 1 - delta=0.9752026.0297.299.72.42
L = 3 (uncertain prior)Model=LLaMA-3.1-405B, 1 - delta=0.992026.0297.299.72.64
ASCModel=LLaMA-3.1-405B, 1 - delta=0.952026.0297.21004.5
ASCModel=LLaMA-3.1-405B, 1 - delta=0.9752026.0297.21005.78
ASCModel=LLaMA-3.1-405B, 1 - delta=0.992026.0297.21007.02
L = 2 (known prior)Model=Qwen-2.5-72B, 1 - delta=0.952026.0297.199.82.03
L = 3 (known prior)Model=Qwen-2.5-72B, 1 - delta=0.952026.0297.199.72
L = 3 (known prior)Model=LLaMA-3.1-405B, 1 - delta=0.9752026.0297.199.92.04
L = 2 (known prior)Model=LLaMA-3.1-405B, 1 - delta=0.952026.029799.81.95
L = 2 (known prior)Model=LLaMA-3.1-405B, 1 - delta=0.9752026.029799.82.09
L = 3 (known prior)Model=LLaMA-3.1-405B, 1 - delta=0.952026.029799.81.88
ASCModel=GPT-4o mini, 1 - delta=0.952026.0296.899.65.33
ASCModel=GPT-4o mini, 1 - delta=0.9752026.0296.899.66.64
L = 3 (known prior)Model=GPT-4o mini, 1 - delta=0.9752026.0296.799.53.73
L = 2 (known prior)Model=GPT-4o mini, 1 - delta=0.952026.0296.699.43.53
L = 2 (known prior)Model=GPT-4o mini, 1 - delta=0.9752026.0296.699.43.81
L = 3 (known prior)Model=GPT-4o mini, 1 - delta=0.952026.0296.699.43.42
L = 2 (uncertain prior)Model=GPT-4o mini, 1 - delta=0.9752026.0296.699.13.98
L = 2 (uncertain prior)Model=GPT-4o mini, 1 - delta=0.992026.0296.6994.41
L = 3 (uncertain prior)Model=GPT-4o mini, 1 - delta=0.9752026.0296.6993.49
L = 3 (uncertain prior)Model=GPT-4o mini, 1 - delta=0.992026.0296.6993.8
L = 2 (known prior)Model=GPT-4o mini, 1 - delta=0.992026.0296.499.74.04
L = 3 (known prior)Model=GPT-4o mini, 1 - delta=0.992026.0296.499.73.94
ASCModel=GPT-4o mini, 1 - delta=0.992026.0296.41008.27
L = 2 (uncertain prior)Model=Qwen-2.5-72B, 1 - delta=0.952026.0296.398.21
L = 2 (uncertain prior)Model=Qwen-2.5-72B, 1 - delta=0.9752026.0296.398.21
L = 3 (uncertain prior)Model=Qwen-2.5-72B, 1 - delta=0.952026.0296.398.21
L = 3 (uncertain prior)Model=Qwen-2.5-72B, 1 - delta=0.9752026.0296.398.21
L = 2 (uncertain prior)Model=LLaMA-3.1-405B, 1 - delta=0.952026.0296.297.71
L = 3 (uncertain prior)Model=LLaMA-3.1-405B, 1 - delta=0.952026.0296.297.71
L = 2 (uncertain prior)Model=GPT-4o mini, 1 - delta=0.952026.0294.396.51
L = 3 (uncertain prior)Model=GPT-4o mini, 1 - delta=0.952026.0294.396.51