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

MultiSpanQA

Benchmarks

Task NameDataset NameSOTA ResultTrend
Question AnsweringMultiSpanQA (test)
Exact F171.4
11
Multi-answer Machine Reading ComprehensionMultiSpanQA (val)
EM F1 Score0.7193
8
Instance-level EvaluationMultiSpanQA
AUC-ROC72.9
7
Closed book Question AnsweringMultiSpanQA (test)
F1 Score48
5
Showing 4 of 4 rows