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

QuAC

Benchmarks

Task NameDataset NameSOTA ResultTrend
Reading ComprehensionQuAC
F1 Score74.4
28
Conversational Question AnsweringQuAC
Accuracy53.51
9
Conversational Question AnsweringQuAC 3,000 3
Accuracy56.2
9
Conversational Question AnsweringQuAC-2 2,000
Accuracy58.05
9
Conversational Question AnsweringQuAC 1,000 1
Accuracy59.4
9
Conversational Question AnsweringQuAC
F1 Score67.7
9
Conversational Machine ComprehensionQuAC (test)
F1 Score80.8
8
Conversational RetrievalQuAC
Top-1 Recall56.8
7
Conversational Question AnsweringQuAC (test)
F1 Score66.1
7
Reading ComprehensionQuAC
Accuracy53.6
6
Reading ComprehensionQuAC (dev)
F1 Score44.3
6
Conversational RetrievalOR-QuAC (test)
NDCG@343.5
4
Conversational Question AnsweringQuAC (test dev)
F1 Score44.3
2
Dialogue Quality EvaluationQuAC
BF1 (qt, at)0.43
1
Showing 14 of 14 rows