Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Long-form Generation on PopQA (test)

0.8588AUROC

SeSE

0.506240.597770.68930.78083Nov 20, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.11
0.85880.8119
2025.11
0.8480.8055
2025.11
0.84680.8224
2025.11
0.79090.8094
2025.11
0.78910.7791
2025.11
0.72470.7709
2025.11
0.67530.73
2025.11
0.64760.7307
2025.11
0.64610.7006
2025.11
0.6210.7331
2025.11
0.58230.685
2025.11
0.51980.6394