English

Graph Refinement for Coreference Resolution

Computation and Language 2022-04-01 v1 Machine Learning

Abstract

The state-of-the-art models for coreference resolution are based on independent mention pair-wise decisions. We propose a modelling approach that learns coreference at the document-level and takes global decisions. For this purpose, we model coreference links in a graph structure where the nodes are tokens in the text, and the edges represent the relationship between them. Our model predicts the graph in a non-autoregressive manner, then iteratively refines it based on previous predictions, allowing global dependencies between decisions. The experimental results show improvements over various baselines, reinforcing the hypothesis that document-level information improves conference resolution.

Keywords

Cite

@article{arxiv.2203.16574,
  title  = {Graph Refinement for Coreference Resolution},
  author = {Lesly Miculicich and James Henderson},
  journal= {arXiv preprint arXiv:2203.16574},
  year   = {2022}
}