Coreference Resolution without Span Representations

TL;DR

This paper presents a lightweight, efficient end-to-end coreference resolution model that eliminates the need for span representations, maintaining competitive performance while reducing memory usage.

Contribution

The authors introduce a novel coreference model that removes span representations and heuristics, simplifying the architecture and improving efficiency.

Findings

Performs competitively with standard models

Reduces memory footprint significantly

Enables processing of longer documents

Abstract

The introduction of pretrained language models has reduced many complex task-specific NLP models to simple lightweight layers. An exception to this trend is coreference resolution, where a sophisticated task-specific model is appended to a pretrained transformer encoder. While highly effective, the model has a very large memory footprint -- primarily due to dynamically-constructed span and span-pair representations -- which hinders the processing of complete documents and the ability to train on multiple instances in a single batch. We introduce a lightweight end-to-end coreference model that removes the dependency on span representations, handcrafted features, and heuristics. Our model performs competitively with the current standard model, while being simpler and more efficient.