Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

FrameNet

Benchmarks

Task NameDataset NameSOTA ResultTrend
Argument ClusteringFrameNet 1.7 (three-fold cross-validation)
PU90.1
10
Semantic-conditioned sentence generationFrameNet After Filtering 1.7 (test)
PPL97
8
Semantic-conditioned sentence generationFrameNet Before Filtering 1.7 (test)
FE Fid.0.979
8
Frame-SRLFrameNet Augmented LUs (test)
F1 Score0.682
3
Frame-SRLFrameNet All LUs (test)
F1 Score68.4
3
Semantic Role LabelingFrameNet 1.7 (test)
Accuracy49.81
2
Semantic Role ClassificationFrameNet 1.7 (test)
Accuracy59.35
2
Showing 7 of 7 rows