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

PersonalDialog

Benchmarks

Task NameDataset NameSOTA ResultTrend
Persona Consistency EvaluationPersonalDialog KvPI (test)
Delta P2.8
7
Dialogue GenerationPersonalDialog Biased (test)
PPL19.5
7
Dialogue GenerationPersonalDialog Random (test)
PPL18.5
7
Showing 3 of 3 rows