Document-Level Relation Extraction with Sentences Importance Estimation and Focusing

TL;DR

This paper introduces the SIEF framework for document-level relation extraction, emphasizing sentence importance to improve model robustness and accuracy by focusing on evidence sentences within documents.

Contribution

The paper proposes a novel SIEF framework that estimates sentence importance and applies focusing loss, enhancing robustness and performance across various DocRE models.

Findings

SIEF improves overall relation extraction accuracy.

SIEF increases model robustness to non-evidence sentences.

Effective across multiple base models.

Abstract

Document-level relation extraction (DocRE) aims to determine the relation between two entities from a document of multiple sentences. Recent studies typically represent the entire document by sequence- or graph-based models to predict the relations of all entity pairs. However, we find that such a model is not robust and exhibits bizarre behaviors: it predicts correctly when an entire test document is fed as input, but errs when non-evidence sentences are removed. To this end, we propose a Sentence Importance Estimation and Focusing (SIEF) framework for DocRE, where we design a sentence importance score and a sentence focusing loss, encouraging DocRE models to focus on evidence sentences. Experimental results on two domains show that our SIEF not only improves overall performance, but also makes DocRE models more robust. Moreover, SIEF is a general framework, shown to be effective when combined with a variety of base DocRE models.