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

P4G

Benchmarks

Task NameDataset NameSOTA ResultTrend
Dialogue PlanningP4G (test)
Average Turn (AT)3.12
19
Charity PersuasionP4G User Simulation
Success Rate (SR)78
16
Proactive dialogueP4G
SR96.67
10
Proactive dialogueP4G+
Success Rate (SR)59.17
9
Failure prediction stoppingP4G
Max Accuracy77.5
6
Strategy PredictionP4G
Macro F10.1496
6
Dialogue Response GenerationP4G (test)
Accuracy89
3
Interactive Persuasive DialogueP4G (interactive)
Competence4.21
2
Showing 8 of 8 rows