List your company

Mathematical Reasoning on GSM8K MATH-500 (test)

92.3Accuracy (GSM8K)Qwen3
50.59661.42372.2583.077Mar 6, 2026

Evaluation results

MethodLinks
Qwen3Sparsity=Full, Model=Qwen32026.0392.380.286.3
REAPSparsity=25%, Allocation Protocol=UNI2026.039178.484.7
FreqSparsity=25%, Allocation Protocol=ESAP2026.0390.773.482.1
EANSparsity=25%, Allocation Protocol=UNI2026.0390.274.882.5
EANSparsity=25%, Allocation Protocol=ESAP2026.0390.175.282.7
SEERSparsity=25%, Allocation Protocol=ESAP2026.0389.771.680.6
SEERSparsity=25%, Allocation Protocol=UNI2026.0389.163.676.4
FreqSparsity=25%, Allocation Protocol=UNI2026.0389.165.877.4
REAPSparsity=25%, Allocation Protocol=ESAP2026.0388.677.883.2
REAPSparsity=50%, Allocation Protocol=UNI2026.0387.279.883.5
REAPSparsity=50%, Allocation Protocol=ESAP2026.0386.779.283
EANSparsity=50%, Allocation Protocol=ESAP2026.0384.649.467
EANSparsity=50%, Allocation Protocol=UNI2026.0383.345.664.4
ERNIESparsity=Full2026.0382.97880.4
ESAPModel=ERNIE, Sparsity=25%, Criterion=EAN2026.0382.377.279.7
REAPModel=ERNIE, Sparsity=25%, Allocation=UNI2026.0382.17377.5
EANModel=ERNIE, Sparsity=25%, Allocation=UNI2026.0381.574.878.1
ESAPModel=ERNIE, Sparsity=25%, Criterion=Freq2026.0381.571.676.5
ESAPModel=ERNIE, Sparsity=25%, Criterion=REAP2026.0381.475.278.3
FreqModel=ERNIE, Sparsity=25%, Allocation=UNI2026.038169.275.1
SEERModel=ERNIE, Sparsity=25%, Allocation=UNI2026.0380.473.677
ESAPModel=ERNIE, Sparsity=25%, Criterion=SEER2026.0380.472.876.6
EANModel=ERNIE, Sparsity=50%, Allocation=UNI2026.0374.854.264.5
ESAPModel=ERNIE, Sparsity=50%, Criterion=EAN2026.0374.455.865.1
ESAPModel=ERNIE, Sparsity=50%, Criterion=REAP2026.0371.857.864.8
FreqSparsity=50%, Allocation Protocol=ESAP2026.0369.721.445.5
REAPModel=ERNIE, Sparsity=50%, Allocation=UNI2026.0369.559.864.6
OLMoESparsity=Full2026.0368.222.245.2
ESAPModel=ERNIE, Sparsity=50%, Criterion=SEER2026.036450.857.4
ESAPModel=OLMoE, Sparsity=25%, Criterion=REAP2026.0363.621.642.6
ESAPModel=ERNIE, Sparsity=50%, Criterion=Freq2026.0363.146.854.9
SEERSparsity=50%, Allocation Protocol=UNI2026.0360.514.437.4
ESAPModel=OLMoE, Sparsity=25%, Criterion=SEER2026.0360.124.842.4
ESAPModel=OLMoE, Sparsity=25%, Criterion=Freq2026.0359.620.840.2
REAPModel=OLMoE, Sparsity=25%, Allocation=UNI2026.0359.62039.8
FreqSparsity=50%, Allocation Protocol=UNI2026.0359.212.836
FreqModel=OLMoE, Sparsity=25%, Allocation=UNI2026.0359.12240.5
EANModel=OLMoE, Sparsity=25%, Allocation=UNI2026.0358.51938.7
SEERModel=OLMoE, Sparsity=25%, Allocation=UNI2026.0357.720.439
ESAPModel=OLMoE, Sparsity=25%, Criterion=EAN2026.0357.623.240.4
SEERModel=ERNIE, Sparsity=50%, Allocation=UNI2026.0355.541.848.7
SEERSparsity=50%, Allocation Protocol=ESAP2026.03552238.5
FreqModel=ERNIE, Sparsity=50%, Allocation=UNI2026.0352.227.239.7