Investigating Non-local Features for Neural Constituency Parsing

Open paper

Benchmark	Model	Rank	Results
constituency-parsing-on-penn-treebank	NFC + BERT-large	#9	F1 score: 95.92