Sequential Cross-Document Coreference Resolution

TL;DR

This paper introduces a sequential model for cross-document coreference resolution that incrementally builds entity and event clusters, demonstrating competitive results and highlighting the effectiveness of sequential and higher-order inference methods.

Contribution

It extends sequential coreference models to cross-document scenarios, achieving competitive performance and providing insights through extensive ablation studies.

Findings

The model achieves competitive results for entity and event coreference.

Sequential and higher-order inference methods are effective in cross-document coreference.

Ablation studies reveal the importance of various inputs and representations.

Abstract

Relating entities and events in text is a key component of natural language understanding. Cross-document coreference resolution, in particular, is important for the growing interest in multi-document analysis tasks. In this work we propose a new model that extends the efficient sequential prediction paradigm for coreference resolution to cross-document settings and achieves competitive results for both entity and event coreference while provides strong evidence of the efficacy of both sequential models and higher-order inference in cross-document settings. Our model incrementally composes mentions into cluster representations and predicts links between a mention and the already constructed clusters, approximating a higher-order model. In addition, we conduct extensive ablation studies that provide new insights into the importance of various inputs and representation types in coreference.