Coreferential Reasoning Learning for Language Representation
Open paper
Benchmark
Model
Rank
Results
relation-extraction-on-docred
CorefRoBERTa-large
#23
F1: 60.25
Ign F1: 57.90
relation-extraction-on-docred
CorefBERT-large
#32
F1: 58.83
Ign F1: 56.40
relation-extraction-on-docred
CorefBERT-base
#37
F1: 56.96
Ign F1: 54.54
Rank counts only results with a code link.