Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Preference Alignment on PRISM 1.0 (test)
Loading...
2.393
Borda Average
Hard Panel
1.71076
1.88788
2.065
2.24212
Feb 4, 2026
Borda Average
Copeland Score
Updated 4d ago
Evaluation Results
Method
Method
Links
Borda Average
Copeland Score
Hard Panel
Backbone=Llama-3.2-3B,...
2026.02
2.393
11,797.238
US-Rep
Backbone=Llama-3.2-3B,...
2026.02
2.151
4,537.12
Soft Panel
Backbone=Llama-3.2-3B,...
2026.02
1.882
-3,542.388
Full PRISM
Backbone=Llama-3.2-3B,...
2026.02
1.837
-4,892.736
Base
Backbone=Llama-3.2-3B,...
2026.02
1.737
-7,899.234
Feedback
Search any
task
Search any
task