Hybrid Rule-Neural Coreference Resolution System based on Actor-Critic Learning

TL;DR

This paper introduces a hybrid rule-neural coreference resolution system utilizing actor-critic learning, combining heuristic rules and neural models to improve coreference clustering performance, achieving state-of-the-art results on the CoNLL-2012 dataset.

Contribution

It presents a novel end-to-end hybrid system that jointly performs mention detection and resolution using actor-critic learning with BERT representations.

Findings

Achieves state-of-the-art performance on CoNLL-2012 English Test Set.

Effectively combines heuristic rules with neural models for coreference resolution.

Demonstrates the benefits of joint training for mention detection and resolution.

Abstract

A coreference resolution system is to cluster all mentions that refer to the same entity in a given context. All coreference resolution systems need to tackle two main tasks: one task is to detect all of the potential mentions, and the other is to learn the linking of an antecedent for each possible mention. In this paper, we propose a hybrid rule-neural coreference resolution system based on actor-critic learning, such that it can achieve better coreference performance by leveraging the advantages from both the heuristic rules and a neural conference model. This end-to-end system can also perform both mention detection and resolution by leveraging a joint training algorithm. We experiment on the BERT model to generate input span representations. Our model with the BERT span representation achieves the state-of-the-art performance among the models on the CoNLL-2012 Shared Task English Test Set.