Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Information flow management in multi-party professional scenarios on ConfAIde Tier 4 (test)
Loading...
95
MS-E Score
base
43
56.5
70
83.5
Feb 14, 2026
MS-E Score
AI-E Score
Updated 3mo ago
Evaluation Results
Method
Method
Links
MS-E Score
AI-E Score
base
Backbone=Mistral-7B, Z...
2026.02
95
95
base
Backbone=Llama-3B, Zer...
2026.02
85
100
PrivAct (Refiner)
Backbone=Mistral-7B, Z...
2026.02
80
90
base
Backbone=Llama-8B, Zer...
2026.02
75
95
PrivAct (Refiner)
Backbone=Llama-3B, Zer...
2026.02
75
100
PrivAct (Verifier)
Backbone=Llama-3B, Zer...
2026.02
75
100
PrivAct (Refiner)
Backbone=Llama-8B, Zer...
2026.02
70
100
PrivAct (Verifier)
Backbone=Llama-8B, Zer...
2026.02
70
88
PrivAct (Verifier)
Backbone=Mistral-7B, Z...
2026.02
68
100
base
Backbone=Qwen-4B, Zero...
2026.02
55
80
PrivAct (Verifier)
Backbone=Qwen-4B, Zero...
2026.02
54.5
85
PrivAct (Refiner)
Backbone=Qwen-4B, Zero...
2026.02
45
80
Feedback
Search any
task
Search any
task