Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Backdoor Attack on Wiki topic evaluation requests (test)
Loading...
93.2
ASR
BadDLM
-3.728
21.436
46.6
71.764
May 10, 2026
ASR
Updated 22d ago
Evaluation Results
Method
Method
Links
ASR
BadDLM
Method=BadDLM (Ours),...
2026.05
93.2
RL-based
Method=RL-based, Base...
2026.05
67.5
VPI
Method=VPI, Base Model...
2026.05
54.1
SFT-based
Method=SFT-based, Base...
2026.05
48.9
Benign (No Attack)
Method=Benign (No Atta...
2026.05
0
Feedback
Search any
task
Search any
task