Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

CBT

Benchmarks

Task NameDataset NameSOTA ResultTrend
Machine ComprehensionCBT-CN (test)
Accuracy83.7
56
Machine ComprehensionCBT NE (test)
Accuracy81.6
56
Machine ComprehensionCBT-CN (val)
Accuracy85.7
37
Machine ComprehensionCBT (test)
Named Entities73.2
12
CBT Conversation GenerationCBT conversation evaluation dataset
Semantic Coherence1.94
10
Information ExchangeCBT
F1 Score72.2
10
Zero-shot Language ModelingCBT (test)
Accuracy84.2
4
Showing 7 of 7 rows