Pairwise Preference Evaluation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
61Context 1 Preference Score
21
May 12, 2026
91HFS
15
Jun 25, 2026
95.5Win Rate
13
Feb 26, 2026
60.6Pairwise Win Rate
4
Feb 26, 2026
52.9Pairwise Win Rate
3
Feb 26, 2026