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

MRQA

Benchmarks

Task NameDataset NameSOTA ResultTrend
Question AnsweringMRQA 2019 (dev)
SQuAD Score92.5
32
Question AnsweringMRQA Average across 6 domains
EM56.8
23
Extractive Question AnsweringMRQA
NewsQA Score73.6
19
Continual Model Refinement for Extractive Question AnsweringMRQA streams (val)
EFR97.49
16
Question AnsweringMRQA Out-of-domain
F1 Score47.72
12
Question AnsweringMRQA In-domain
F1 Score63.63
12
Showing 6 of 6 rows