Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Trojan Attack (Target action: 'arithmetic') on HalfCheetah
Loading...
100
Attack Success Rate (ASR)
IMC
-4
23
50
77
Jun 15, 2025
Attack Success Rate (ASR)
BTP
CP
Updated 5d ago
Evaluation Results
Method
Method
Links
Attack Success Rate (ASR)
BTP
CP
IMC
Model=DT
2025.06
100
94.2
97
TrojanTO
Model=DT
2025.06
100
98.1
99
TrojanTO
Model=DC
2025.06
100
100
100
TrojanTO
Model=GDT
2025.06
90.7
100
94.6
IMC
Model=DC
2025.06
37
94
46.4
Baffle
Model=DT
2025.06
2
92
3.9
IMC
Model=GDT
2025.06
1
100
2
Baffle
Model=GDT
2025.06
0
100
0
Baffle
Model=DC
2025.06
0
92.4
0
Feedback
Search any
task
Search any
task