Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Harmful prompt detection on SimpST
Loading...
100
F1 Score
Ayub & Majumdar
90.952
93.301
95.65
97.999
Feb 22, 2025
F1 Score
Updated 1d ago
Evaluation Results
Method
Method
Links
F1 Score
Ayub & Majumdar
Backbone=OLMo2-7B-Inst...
2025.02
100
MLPM
Backbone=OLMo2-7B-Inst...
2025.02
100
MLPM
Backbone=Llama-8B-Inst...
2025.02
99.5
MLPM
Backbone=Mistral-7B-In...
2025.02
99.5
Abdelnabi et al.
Backbone=OLMo2-7B-Inst...
2025.02
99.5
LlamaGuard3
Methodology=Guard Model
2025.02
99.5
GraniteGuardian-3-1-8B
Methodology=Guard Model
2025.02
99.5
WildGuard
Methodology=Guard Model
2025.02
99.5
Abdelnabi et al.
Backbone=Llama-8B-Inst...
2025.02
98.99
Ayub & Majumdar
Backbone=Mistral-7B-In...
2025.02
98.99
Ayub & Majumdar
Backbone=Llama-8B-Inst...
2025.02
98.48
Abdelnabi et al.
Backbone=Mistral-7B-In...
2025.02
98.48
Abdelnabi et al.
Backbone=Qwen3-8B-Inst...
2025.02
97.96
Aegis-Guard-D
Methodology=Guard Model
2025.02
97.96
MLPM
Backbone=Qwen3-8B-Inst...
2025.02
96.91
Ayub & Majumdar
Backbone=Qwen3-8B-Inst...
2025.02
95.29
ShieldGemma-9B
Methodology=Guard Model
2025.02
91.3
Feedback
Search any
task
Search any
task