Ensembling Graph Predictions for AMR Parsing

Open paper
Benchmark Model Rank Results
amr-parsing-on-bioGraphene Smatch#3Smatch: 62.8
amr-parsing-on-ldc2017t10Graphene Smatch (IBM)#2Smatch: 86.26
amr-parsing-on-ldc2017t10Graphene Support (IBM)#5Smatch: 85.85
amr-parsing-on-ldc2020t02Graphene Smatch (IBM)#2Smatch: 84.87
amr-parsing-on-ldc2020t02Graphene Support (IBM)#4Smatch: 84.41
amr-parsing-on-new3Graphene Smatch#2Smatch: 76.32
amr-parsing-on-the-little-princeGraphene Smatch#2Smatch: 79.52

Rank counts only results with a code link.