Word-Level Coreference Resolution

TL;DR

This paper introduces a word-level approach to coreference resolution that reduces computational complexity and outperforms span-based models, while maintaining competitive accuracy on benchmark datasets.

Contribution

The authors propose a novel word-level coreference resolution model that simplifies computations and eliminates the need for pruning, outperforming span-based models like SpanBERT.

Findings

Reduces complexity from O(n^4) to O(n^2)

Outperforms SpanBERT on coreference resolution

Maintains competitive accuracy on OntoNotes

Abstract

Recent coreference resolution models rely heavily on span representations to find coreference links between word spans. As the number of spans is $O(n^2)$ in the length of text and the number of potential links is $O(n^4)$, various pruning techniques are necessary to make this approach computationally feasible. We propose instead to consider coreference links between individual words rather than word spans and then reconstruct the word spans. This reduces the complexity of the coreference model to $O(n^2)$ and allows it to consider all potential mentions without pruning any of them out. We also demonstrate that, with these changes, SpanBERT for coreference resolution will be significantly outperformed by RoBERTa. While being highly efficient, our model performs competitively with recent coreference resolution systems on the OntoNotes benchmark.