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

LLM Safety and Informativeness Evaluation on HarmfulQA

98.1Safety Rate

SafeMoE-XL

15.9437.2758.679.93May 30, 2026
Updated 1d ago

Evaluation Results

MethodLinks
2026.05
98.17.1
2026.05
89.17.17
2026.05
82.37.3
72.26.1
2026.05
69.96.9
2026.05
69.886.86
2026.05
48.27
2026.05
45.96.7
2026.05
44.64.4
2026.05
40.154.07
2026.05
40.14
2026.05
32.856.17
2026.05
32.86.2
2026.05
23.95.2
19.15.8