Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Cognitive Poisoning Detection on Cognitive Poisoning Detection 5-fold evaluation (test)
Loading...
4.2
AMR
VISTA-Guard
3.36
9.03
14.7
20.37
May 17, 2026
AMR
BMR
RNR
ACNR
GUARDEDJOINT
Updated 15d ago
Evaluation Results
Method
Method
Links
AMR
BMR
RNR
ACNR
GUARDEDJOINT
VISTA-Guard
Backbone=Mistral-7B
2026.05
4.2
95.8
5
95
84.2
VISTA-Guard
Backbone=Llama-3.1-8B
2026.05
4.6
95.4
18.8
81.2
62.6
VISTA-Guard
Backbone=OLMo-2-7B
2026.05
6.6
93.4
15.1
84.9
64.2
VISTA-Guard
Backbone=Qwen3-8B
2026.05
7.6
92.4
11.6
88.4
67.6
VISTA-Guard
Backbone=Qwen2.5-7B
2026.05
9.6
90.4
10.6
89.4
64.9
GPT-5.4
Mode=Full (trajectory...
2026.05
22.2
77.8
80.4
19.6
0
GPT-5.4
Mode=Features only (10...
2026.05
25.2
74.8
80.9
19.1
0
Feedback
Search any
task
Search any
task