List your company

Preference Alignment on 15,000 listwise rankings (test)

0.384BT ScoreHard Panel
-0.46464-0.24432-0.0240.19632Feb 4, 2026

Evaluation results

MethodLinks
Hard Panelbackbone=Llama-3.1-8B, bootstrap_n=10002026.020.3840.375
US-Repbackbone=Llama-3.1-8B, bootstrap_n=10002026.020.1020.113
Soft Panelbackbone=Llama-3.1-8B, bootstrap_n=10002026.020.0170.071
Full PRISMbackbone=Llama-3.1-8B, bootstrap_n=10002026.02-0.070.026
Basebackbone=Llama-3.1-8B, bootstrap_n=10002026.02-0.432-0.584