CorefDRE: Document-level Relation Extraction with coreference resolution

TL;DR

CorefDRE introduces a novel approach for document-level relation extraction by leveraging coreference resolution and a noise suppression mechanism to better handle pronouns across multiple sentences, improving extraction accuracy.

Contribution

The paper proposes a coreference-aware graph inference network that effectively models mention-pronoun relations for improved document-level relation extraction.

Findings

Outperforms state-of-the-art on DocRED, DialogRE, and MPDD datasets.

Effectively models mention-pronoun coreference for relation extraction.

Noise suppression reduces errors caused by ambiguous pronouns.

Abstract

Document-level relation extraction is to extract relation facts from a document consisting of multiple sentences, in which pronoun crossed sentences are a ubiquitous phenomenon against a single sentence. However, most of the previous works focus more on mentions coreference resolution except for pronouns, and rarely pay attention to mention-pronoun coreference and capturing the relations. To represent multi-sentence features by pronouns, we imitate the reading process of humans by leveraging coreference information when dynamically constructing a heterogeneous graph to enhance semantic information. Since the pronoun is notoriously ambiguous in the graph, a mention-pronoun coreference resolution is introduced to calculate the affinity between pronouns and corresponding mentions, and the noise suppression mechanism is proposed to reduce the noise caused by pronouns. Experiments on the public dataset, DocRED, DialogRE and MPDD, show that Coref-aware Doc-level Relation Extraction based on Graph Inference Network outperforms the state-of-the-art.