List your company

Reasoning on In-distribution reasoning (test)

29.01GPQA ScoreAutoSelection
14.158818.014421.8725.7256May 13, 2026

Evaluation results

MethodLinks
AutoSelectionModel=Qwen2.5-1.5B, Group=Recipe2026.0529.0154.583055.3342.23
VarentropyModel=Qwen2.5-3B, Group=Top-k2026.0528.1249.235.2161.2743.45
AutoSelectionModel=Llama3.2-1B, Group=Recipe2026.0526.7811.526.0834.5519.73
N-gramModel=Llama3.2-1B, Group=Top-k2026.0524.7710.763.2634.5518.34
N-gramModel=Qwen2.5-1.5B, Group=Top-k2026.0524.151.6319.1357.3838.06
AOModel=Qwen2.5-1.5B, Group=Top-k2026.0523.6649.8133.3655.3340.54
Full dataModel=Qwen2.5-3B, Group=Full2026.0523.6664.8231.956145.36
MONAModel=Qwen2.5-3B, Group=Top-k2026.0523.6670.5828.6961.8346.19
SemDedupModel=Qwen2.5-3B, Group=Dedup2026.0523.6660.1929.2361.8843.74
RandomModel=Qwen2.5-1.5B, Group=Recipe2026.0523.4352.2327.3956.8339.97
IFDModel=Qwen2.5-1.5B, Group=Top-k2026.0523.2123.8824.4556.2231.94
AOModel=Qwen2.5-3B, Group=Top-k2026.0523.2162.6935.8660.1145.47
AutoSelectionModel=Qwen2.5-3B, Group=Recipe2026.0522.9972.7836.8463.7249.08
RandomModel=Qwen2.5-3B, Group=Top-k2026.0521.8764.5931.7361.9445.03
AOModel=Llama3.2-1B, Group=Top-k2026.0521.439.559.0433.3318.34
RandomModel=Qwen2.5-3B, Group=Recipe2026.0521.4364.4430.226244.52
IFDModel=Qwen2.5-3B, Group=Top-k2026.0521.4221.2221.8460.6631.29
N-gramModel=Qwen2.5-3B, Group=Top-k2026.0521.4259.6621.6361.1140.96
Full dataModel=Qwen2.5-1.5B, Group=Full2026.0521.255.2622.9355.9438.83
RandomModel=Qwen2.5-1.5B, Group=Top-k2026.0521.252.1627.8256.3839.39
MONAModel=Qwen2.5-1.5B, Group=Top-k2026.0521.254.3524.8955.7739.05
VarentropyModel=Qwen2.5-1.5B, Group=Top-k2026.0520.3138.5125.7655.535.02
RandomModel=Llama3.2-1B, Group=Top-k2026.0519.4114.18.4733.3318.83
MONAModel=Llama3.2-1B, Group=Top-k2026.0518.9710.389.8933.6118.21
SemDedupModel=Llama3.2-1B, Group=Dedup2026.0517.416.360.7626.7712.83
Full dataModel=Llama3.2-1B, Group=Full2026.0517.1916.64.7833.0617.91
VarentropyModel=Llama3.2-1B, Group=Top-k2026.0517.183.568.6931.0515.12
IFDModel=Llama3.2-1B, Group=Top-k2026.0516.0710.084.3429.3314.96
RandomModel=Llama3.2-1B, Group=Recipe2026.0515.6213.499.4532.3317.72
SemDedupModel=Qwen2.5-1.5B, Group=Dedup2026.0514.7340.420.4356.8833.11