List your company

Mathematical Reasoning on SVAMP (Accuracy by Context Size)

0.8933Accuracy (Context Size 128)EvoToken (10k FT)
0.6576360.7188180.780.841182Jan 12, 2026

Evaluation results

MethodLinks
EvoToken (10k FT)NFE / Gen Len=1/22026.010.89330.92330.89670.9044
EvoToken (10k FT)NFE / Gen Len=12026.010.890.89670.90.8956
BaselineNFE / Gen Len=12026.010.88330.84670.860.8633
FT-Base (10k FT)NFE / Gen Len=12026.010.87670.89670.89670.89
BaselineNFE / Gen Len=1/22026.010.860.86670.840.8556
FT-Base (10k FT)NFE / Gen Len=1/22026.010.850.870.89330.8711
BaselineNFE / Gen Len=1/42026.010.810.77330.750.7778
EvoToken (10k FT)NFE / Gen Len=1/42026.010.78330.83330.81330.81
FT-Base (10k FT)NFE / Gen Len=1/42026.010.66670.75330.720.7133