Preference Evaluation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
559WR (%)
64
Jun 9, 2026
94.15Multi-choice One Preference p+
39
Apr 15, 2026
76Overall Preference Fraction
30
Feb 26, 2026
67.5Accuracy
29
Jun 23, 2026
86.2Helpfulness Score (BoN)
16
Mar 4, 2026
75Acc@0.5
15
Feb 26, 2026
83Acc@0.5
15
Feb 26, 2026
59Acc@0.5
15
Feb 26, 2026
63Acc@0.5
15
Feb 26, 2026
60Acc@0.5
15
Feb 26, 2026
59Acc@0.5
15
Feb 26, 2026
66Acc@0.5
15
Feb 26, 2026
81Acc@0.5
15
Feb 26, 2026
81Acc@0.5
15
Feb 26, 2026
8.25Avg Score
14
Feb 26, 2026
9.06Average Score
14
Feb 26, 2026
8.58Avg Score
14
Feb 26, 2026
80Win Rate
12
May 12, 2026
0.043Kuiper Statistic
8
Feb 26, 2026
52Win Count
8
Feb 26, 2026
60Win Rate
8
Feb 26, 2026
57Win Rate
8
Feb 26, 2026
41.7Win Rate
2
Feb 26, 2026
33.7Win Rate
2
Feb 26, 2026
43Win Rate
2
Feb 26, 2026