Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Cloze-style Question Answering on ReCoRD (dev)
Loading...
91
EM
DeBERTa_large
42.224
54.887
67.55
80.213
Jun 5, 2020
Jun 24, 2020
Jul 14, 2020
Aug 3, 2020
Aug 23, 2020
Sep 12, 2020
Oct 2, 2020
EM
F1
Updated 4d ago
Evaluation Results
Method
Method
Links
EM
F1
DeBERTa_large
Model Size=Large
2020.06
91
91.4
LUKE
Ensemble=false
2020.10
90.8
91.4
RoBERTa_large
Model Size=Large
2020.06
90
90.6
RoBERTa
Ensemble=false
2020.10
89
89.5
XLNet+Verifier
Ensemble=false
2020.10
80.6
82.1
DocQA+ELMo
Ensemble=false
2020.10
44.1
45.4
Feedback
Search any
task
Search any
task