List your company

Correctness Assessment on UltraFeedback Property Constraints Satisfaction (test)

0.07Worst-case Size Distortion (Helpfulness & Instruction-following)naive (1)
-0.56363.71327.9912.2668Oct 29, 2025

Evaluation results

MethodLinks
naive (1)Score=naive (1)2025.100.070.09
e-score (2)Score=e-score (2)2025.100.80.97
e-score (combined)Score=e-score (combined)2025.100.941.01
e-score (1)Score=e-score (1)2025.1011
e-score (3)Score=e-score (3)2025.1011.05
naive (3)Score=naive (3)2025.101.84.82
naive (2)Score=naive (2)2025.102.255.42
p-scoreScore=p-score2025.109.67.55
p-score (randomized)Score=p-score (randomized)2025.1015.9114.92