End-to-End Neural Discourse Deixis Resolution in Dialogue

TL;DR

This paper introduces an end-to-end neural model for discourse deixis resolution in dialogue, adapting span-based coreference techniques to improve performance on multiple datasets, achieving state-of-the-art results.

Contribution

It extends Lee et al.'s coreference model specifically for discourse deixis in dialogue, incorporating task-specific features for better accuracy.

Findings

Achieved state-of-the-art results on four datasets

Extended span-based coreference model for discourse deixis

Demonstrated effectiveness of task-specific adaptations

Abstract

We adapt Lee et al.'s (2018) span-based entity coreference model to the task of end-to-end discourse deixis resolution in dialogue, specifically by proposing extensions to their model that exploit task-specific characteristics. The resulting model, dd-utt, achieves state-of-the-art results on the four datasets in the CODI-CRAC 2021 shared task.