Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Phishing Detection on HTML-based PI (test)
Loading...
0.3
ASR
InjectDefuser
-3.076
19.712
42.5
65.288
Feb 5, 2026
ASR
Updated 4d ago
Evaluation Results
Method
Method
Links
ASR
InjectDefuser
Backbone=GPT-5, Model...
2026.02
0.3
Advanced
Backbone=GPT-5, Model...
2026.02
10.1
InjectDefuser
Backbone=Grok 4 Fast,...
2026.02
26.1
InjectDefuser
Backbone=Gemma 3 27B,...
2026.02
36.7
Standard
Backbone=GPT-5, Model...
2026.02
39.9
Advanced
Backbone=Gemma 3 27B,...
2026.02
55.4
InjectDefuser
Backbone=Llama 4 Maver...
2026.02
61.7
Standard
Backbone=Gemma 3 27B,...
2026.02
64.7
Standard
Backbone=Grok 4 Fast,...
2026.02
65.1
Advanced
Backbone=Llama 4 Maver...
2026.02
75.6
Advanced
Backbone=Grok 4 Fast,...
2026.02
76.2
Standard
Backbone=Llama 4 Maver...
2026.02
84.7
Feedback
Search any
task
Search any
task