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

Multi-label Safety Categorization on HarmBench Responses

27.26Macro Accuracy

Opir-multitask-large

1.96728.533615.121.6664May 28, 2026
Updated 5d ago

Evaluation Results

MethodLinks
2026.05
27.26
2026.05
21.58
2026.05
20.09
2.94