Fine-tuning BERT in a span-ranking coreference model raises OntoNotes F1 by 3.9 and GAP F1 by 11.5 points, with a qualitative analysis of persistent long-document and conversational errors.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
BERT for Coreference Resolution: Baselines and Analysis
Fine-tuning BERT in a span-ranking coreference model raises OntoNotes F1 by 3.9 and GAP F1 by 11.5 points, with a qualitative analysis of persistent long-document and conversational errors.