Revisiting the Negative Data of Distantly Supervised Relation Extraction

Open paper
Benchmark Model Rank Results
relation-extraction-on-nyt10-hrlReRe#1F1: 73.95
relation-extraction-on-nyt10-hrlReRe (exact)#2F1: 73.4
relation-extraction-on-nyt10-hrlTPLinker Wang et al. (2020)*#4F1: 72.45
relation-extraction-on-nyt10-hrlTPLinker Wang et al. (2020)*(exact)#6F1: 71.93
relation-extraction-on-nyt10-hrlCasRel (exact)#8F1: 70.11
relation-extraction-on-nyt10-hrlHRL Takanobu et al. (2019)#10F1: 64.4
relation-extraction-on-nyt11-hrlRERE#1F1: 56.23
relation-extraction-on-nyt11-hrlReRe (exact)#3F1: 55.47
relation-extraction-on-nyt11-hrlHRL#7F1: 53.8
relation-extraction-on-nyt21ReRe#1F1: 59.62
relation-extraction-on-nyt21ReRe (exact)#2F1: 58.88
relation-extraction-on-nyt21TPLinker(exact)#3F1: 57.33
relation-extraction-on-nyt21CasRel (exact)#4F1: 54.78

Rank counts only results with a code link.