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

DuRecDial

Benchmarks

Task NameDataset NameSOTA ResultTrend
Dialogue Response GenerationDuRecDial 2.0
F145.84
14
Dialogue Response GenerationDuRecDial
F147.43
14
Dialogue GenerationDuRecDial OOD (test)
Coherence4.81
11
Proactive Dialogue GenerationDuRecDial OOD 2.0 (test)
Perplexity1.3
8
Proactive Dialogue GenerationDuRecDial ID 2.0 (test)
PPL1.28
8
Proactive Dialogue GenerationDuRecDial OOD 1.0 (test)
Perplexity (PPL)1.45
8
Proactive Dialogue GenerationDuRecDial ID 1.0 (test)
PPL1.46
8
Proactive Dialogue EvaluationDuRecDial OOD 2.0 (test)
Proactivity4.07
7
Proactive Dialogue EvaluationDuRecDial ID 2.0 (test)
Proactivity3.86
7
Proactive DialogueDuRecDial ID (test)
Proactivity4.21
7
ConversationDuRecDial
Dist-2 Score1.121
7
RecommendationDuRecDial
R@119.52
7
Dialogue Path PlanningDuRecDial 2.0
Dialog Action Accuracy97.68
6
Dialogue Path PlanningDuRecDial
Action Accuracy97.11
6
Target-guided proactive dialogue generationDuRecDial OOD 2.0 (test)
PPL6.46
5
Target-guided proactive dialogue generationDuRecDial ID 2.0 (test)
PPL3.93
5
Target-guided proactive dialogue generationDuRecDial OOD (test)
Perplexity4.17
5
Target-guided proactive dialogue generationDuRecDial ID (test)
Perplexity (PPL)3.31
5
Dialogue GenerationDuRecDial OOD 2.0 (test)
Proactivity3.03
4
Dialogue GenerationDuRecDial ID 2.0 (test)
Proactivity2.9
4
Dialogue GenerationDuRecDial ID (test)
Proficiency3.17
4
Showing 21 of 21 rows