Entity and Evidence Guided Relation Extraction for DocRED

Open paper
Benchmark Model Rank Results
relation-extraction-on-docredE2GRE-BERT-base–F1: 58.72Ign F1: 55.22
relation-extraction-on-docredE2GRE-RoBERTa-large–F1: 62.50Ign F1: 60.30

Rank counts only results with a code link.