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

Harmful Content Detection on WG, XS, AEGIS 2, ToxicChat, OAI Mod (test)

83.6F1 Score

Llama-3.1-8B GRPO

68.5272.43576.3580.265Jun 25, 2026
Updated 29d ago

Evaluation Results

MethodLinks
2026.06
83.6
2026.06
82.2
2026.06
82
2026.06
81.2
2026.06
80.9
2026.06
80.8
2026.06
80.7
2026.06
80.4
2026.06
80.4
2026.06
79.1
2026.06
70.9
2026.06
69.1