Revisiting Joint Modeling of Cross-document Entity and Event Coreference Resolution

Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 4179–4189 Florence, Italy, July 28 - August 2, 2019. c⃝2019 Association for Computational Linguistics 4179 Revisiting Joint Modeling of Cross-document Entity and Event Coreference Resolution Shany Barhom1, Vered Shwartz1, Alon Eirew2, Michael Bugert3, Nils Reimers3, and Ido Dagan1 1 Computer Science Department, Bar-Ilan University 2 Intel AI Lab, Israel 3 Ubiquitous Knowledge Processing Lab, Technische Universitat Darmstadt, Germany {shanyb21,vered1986}@gmail.com, alon.eirew@intel.com {bugert,reimers}@ukp.informatik.tu-darmstadt.de, dagan@cs.biu.ac.il Abstract Recognizing coreferring events and entities across multiple texts is crucial for many NLP applications. Despite the task’s importance, research focus was given mostly to within- document entity coreference, with rather lit- tle attention to the other variants. We pro- pose a neural architecture for cross-document coreference resolution. Inspired by Lee et al. (2012), we jointly model entity and event coreference. We represent an event (entity) mention using its lexical span, surrounding context, and relation to entity (event) mentions via predicate-arguments structures. Our model outperforms the previous state-of-the-art event coreference model on ECB+, while providing the first entity coreference results on this cor- pus. Our analysis confirms that all our rep- resentation elements, including the mention span itself, its context, and the relation to other mentions contribute to the model’s success. 1 Introduction Recognizing that various textual spans across mul- tiple texts refer to the same entity or event is an important NLP task. For example, consider the following news headlines: 1. 2018 Nobel prize for physics goes to Donna Strickland 2. Prof. Strickland is awarded the Nobel prize for physics Both sentences refer to the same entities (Donna Strickland and the Nobel prize for physics) and the same event (awarding the prize), using differ- ent words. In coreference resolution, the goal is to cluster expressions that refer to the same en- tity or event in a text, whether within a single document or across a document collection. Re- cently, there has been increasing interest in cross- text inferences, for example in question answering (Welbl et al., 2018; Yang et al., 2018; Khashabi et al., 2018; Postma et al., 2018). Such applica- tions would benefit from effective cross-document coreference resolution. Despite the importance of the task, the focus of most coreference resolution research has been on its within-document variant, and rather little on cross-document coreference (CDCR). The lat- ter is sometimes addressed partially using entity linking, which links mentions of an entity to its knowledge base entry. However, cross-document entity coreference is substantially broader than en- tity linking, addressing also mentions of common nouns and unfamiliar named entities. The commonly used dataset for CDCR is ECB+ (Cybulska and Vossen, 2014), which annotates within-document coreference as well. The annota- tions are denoted separately for entities and events, making it possible to solve one task while ignor- ing the other. Indeed, to the best of our knowledge, all previously published work on ECB+ addressed only event coreference. Cross-document entity coreference has been ad- dressed on EECB, a predecessor of the ECB+ dataset. Lee et al. (2012) proposed …

Content truncated to 500 words