Narrowing the Modeling Gap: A Cluster-Ranking Approach to Coreference Resolution

TL;DR

This paper introduces a cluster-ranking approach to coreference resolution that combines mention-ranking and entity-mention models, improving performance by lexicalization and joint anaphoricity modeling, outperforming previous methods on ACE datasets.

Contribution

The paper proposes a novel cluster-ranking model that integrates strengths of existing models and enhances it with lexicalization and joint anaphoricity features for better coreference resolution.

Findings

Cluster-ranking models outperform mention-pair and entity-mention models.

Lexicalization improves coreference resolution accuracy.

Joint modeling of anaphoricity and coreference enhances performance.

Abstract

Traditional learning-based coreference resolvers operate by training the mention-pair model for determining whether two mentions are coreferent or not. Though conceptually simple and easy to understand, the mention-pair model is linguistically rather unappealing and lags far behind the heuristic-based coreference models proposed in the pre-statistical NLP era in terms of sophistication. Two independent lines of recent research have attempted to improve the mention-pair model, one by acquiring the mention-ranking model to rank preceding mentions for a given anaphor, and the other by training the entity-mention model to determine whether a preceding cluster is coreferent with a given mention. We propose a cluster-ranking approach to coreference resolution, which combines the strengths of the mention-ranking model and the entity-mention model, and is therefore theoretically more appealing than both of these models. In addition, we seek to improve cluster rankers via two extensions: (1) lexicalization and (2) incorporating knowledge of anaphoricity by jointly modeling anaphoricity determination and coreference resolution. Experimental results on the ACE data sets demonstrate the superior performance of cluster rankers to competing approaches as well as the effectiveness of our two extensions.