Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

PyDPO

Benchmarks

Task NameDataset NameSOTA ResultTrend
Preference AlignmentPyDPO (test)
Accuracy94.32
11
Direct Preference OptimizationPyDPO
Accuracy91.47
11
Showing 2 of 2 rows