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

Language detoxification on Bot-Adversarial Dialogue (BAD) 1.0 (test)

0.084Toxicity Probability

CLICK

0.069240.168870.26850.36813Jun 6, 2023
Updated 1mo ago

Evaluation Results

MethodLinks
2023.06
0.0846.480.490.56
2023.06
0.1056.480.490.56
2023.06
0.1056.870.50.57
2023.06
0.1647.930.260.29
2023.06
0.1877.10.140.15
2023.06
0.27.370.520.59
2023.06
0.3038.920.40.44
2023.06
0.4379.060.420.49
2023.06
0.455.230.40.46
2023.06
0.4536.320.420.49