Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Privacy and Utility Evaluation on AgentLeak Sequential, 3 agents
Loading...
84.5
Privacy Score
Full-System LCGuard
24.7
40.225
55.75
71.275
May 21, 2026
Privacy Score
Task Performance
Helpfulness
Leakage
ASR
Updated 12d ago
Evaluation Results
Method
Method
Links
Privacy Score
Task Performance
Helpfulness
Leakage
ASR
Full-System LCGuard
Backbone=Gemma-9B, Top...
2026.05
84.5
65
72
15.5
23.5
ADAPT
Backbone=Gemma-9B, Top...
2026.05
83
35
28.5
17
40
Per-Agent LCGuard
Backbone=Gemma-9B, Top...
2026.05
78.5
61
66
21.5
27.5
PrivAct
Backbone=Gemma-9B, Top...
2026.05
58.2
56
45.5
41.8
77
Vanilla KV Sharing (LatentMAS)
Backbone=Gemma-9B, Top...
2026.05
27
66
78.5
73
89.5
Feedback
Search any
task
Search any
task