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

Faithfulness Evaluation on WikiHow (test)

39.06SummaC

GPT-5

30.63632.82335.0137.197Dec 3, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.12
39.0674.44
2025.12
38.4389.11
2025.12
35.156.23
2025.12
34.7392.01
2025.12
34.4792.83
2025.12
34.3285.87
2025.12
34.0694.17
2025.12
34.0694.17
2025.12
32.7691.91
2025.12
32.5275.7
2025.12
30.9786.16
2025.12
30.9687.23