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

Short-form generation on TriviaQA

0.87AUROC

SeSE

0.65160.70830.7650.8217Nov 20, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.11
0.870.78
2025.11
0.860.87
2025.11
0.850.76
2025.11
0.850.75
2025.11
0.840.87
2025.11
0.830.74
2025.11
0.820.91
2025.11
0.820.73
2025.11
0.810.91
2025.11
0.810.91
2025.11
0.790.87
2025.11
0.780.73
2025.11
0.780.91
2025.11
0.780.72
2025.11
0.780.78
2025.11
0.780.84
2025.11
0.770.9
2025.11
0.760.72
2025.11
0.760.71
2025.11
0.760.9
2025.11
0.760.78
2025.11
0.760.83
2025.11
0.740.9
2025.11
0.730.83
2025.11
0.730.82
2025.11
0.720.68
2025.11
0.710.66
2025.11
0.680.66
2025.11
0.680.81
2025.11
0.660.66
2025.11
0.660.64
2025.11
0.660.88