Preference Alignment Evaluation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
3.7Reward (R)
30
Jun 4, 2026
92.12Average Score
18
Feb 26, 2026
0Win Rate
11
Feb 26, 2026
0.331BT Score (Mean)
10
Feb 26, 2026
83Win Rate (Llama Judge)
4
Apr 28, 2026