Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Risk Detection on EMBHAZARD (Held-out Set)
Loading...
85
Potential Risk Score
Human
60.04
66.52
73
79.48
May 29, 2026
Potential Risk Score
Risk Type Score
Hazard Score
Updated 2d ago
Evaluation Results
Method
Method
Links
Potential Risk Score
Risk Type Score
Hazard Score
Human
2026.05
85
92.8
58.8
EMBGUARD-4B
Parameters=4B
2026.05
77
60.7
53.3
Qwen-3-VL-235B-a22b
Parameters=235B
2026.05
70.3
71.7
46.2
GPT-5.1
2026.05
67.3
67.2
61.9
EMBGUARD-2B
Parameters=2B
2026.05
61
39.2
22.7
Feedback
Search any
task
Search any
task