List your company

Alphabet-size estimation on Pooled QA Datasets (SQuAD, CoQA, NQ-Open, TriviaQA, HotpotQA)

1.14MAESHADE
1.01841.83922.663.4808Apr 21, 2026

Evaluation results

MethodLinks
SHADEsubsample size (n)=50, oracle size (N)=1002026.041.141.86
Hybrid (GT+EigV)subsample size (n)=50, oracle size (N)=1002026.041.382.23
Hybrid-GGTsubsample size (n)=50, oracle size (N)=1002026.041.382.21
Generalized GT (GGT)subsample size (n)=50, oracle size (N)=1002026.041.422.3
Good-Turing (GT)subsample size (n)=50, oracle size (N)=1002026.041.432.32
Plugin (kobs)subsample size (n)=50, oracle size (N)=1002026.041.542.62
SHADEsubsample size (n)=25, oracle size (N)=1002026.042.053.36
Hybrid-GGTsubsample size (n)=25, oracle size (N)=1002026.042.273.6
Hybrid (GT+EigV)subsample size (n)=25, oracle size (N)=1002026.042.283.63
U-EigV (Laplacian)subsample size (n)=50, oracle size (N)=1002026.042.473.93
Generalized GT (GGT)subsample size (n)=25, oracle size (N)=1002026.042.483.81
Good-Turing (GT)subsample size (n)=25, oracle size (N)=1002026.042.53.85
Plugin (kobs)subsample size (n)=25, oracle size (N)=1002026.042.664.24
SHADEsubsample size (n)=8, oracle size (N)=1002026.042.834.37
SHADEsubsample size (n)=5, oracle size (N)=1002026.042.864.82
U-EigV (Laplacian)subsample size (n)=25, oracle size (N)=1002026.042.884.59
SHADEsubsample size (n)=10, oracle size (N)=1002026.043.014.25
Hybrid (GT+EigV)subsample size (n)=10, oracle size (N)=1002026.043.064.7
Hybrid-GGTsubsample size (n)=10, oracle size (N)=1002026.043.064.76
Hybrid-GGTsubsample size (n)=8, oracle size (N)=1002026.043.25.03
Hybrid (GT+EigV)subsample size (n)=8, oracle size (N)=1002026.043.224.99
Good-Turing (GT)subsample size (n)=10, oracle size (N)=1002026.043.34.86
Generalized GT (GGT)subsample size (n)=10, oracle size (N)=1002026.043.34.96
Hybrid-GGTsubsample size (n)=5, oracle size (N)=1002026.043.45.42
Generalized GT (GGT)subsample size (n)=8, oracle size (N)=1002026.043.435.22
Good-Turing (GT)subsample size (n)=8, oracle size (N)=1002026.043.445.12
U-EigV (Laplacian)subsample size (n)=10, oracle size (N)=1002026.043.515.49
Hybrid (GT+EigV)subsample size (n)=5, oracle size (N)=1002026.043.525.52
Generalized GT (GGT)subsample size (n)=5, oracle size (N)=1002026.043.65.57
Plugin (kobs)subsample size (n)=10, oracle size (N)=1002026.043.655.6
Good-Turing (GT)subsample size (n)=5, oracle size (N)=1002026.043.665.55
U-EigV (Laplacian)subsample size (n)=8, oracle size (N)=1002026.043.665.73
Plugin (kobs)subsample size (n)=8, oracle size (N)=1002026.043.835.87
U-EigV (Laplacian)subsample size (n)=5, oracle size (N)=1002026.043.986.16
Plugin (kobs)subsample size (n)=5, oracle size (N)=1002026.044.186.34