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

Toxicity Defense on PGD-Toxic

1.53ASR (PGD-Toxic)

ASTRA

-1.102416.666334.43552.2037Mar 28, 2026
Updated 19d ago

Evaluation Results

MethodLinks
2026.03
1.53
2026.03
2.37
2026.03
3.38
2026.03
4.6
2026.03
5.54
2026.03
11.43
2026.03
12.25
2026.03
12.78
2026.03
14.13
2026.03
15.12
2026.03
55.98
2026.03
57.13
2026.03
60.38
2026.03
67.34