REXEL: An End-to-end Model for Document-Level Relation Extraction and Entity Linking

TL;DR

REXEL is an efficient end-to-end model for document-level relation extraction and entity linking that outperforms existing methods in speed and accuracy, enabling scalable structured information extraction from unstructured text.

Contribution

The paper introduces REXEL, a unified model that performs multiple document-level IE tasks in a single pass, reducing error propagation and capturing long-range dependencies.

Findings

REXEL is 11 times faster than existing approaches.

REXEL surpasses baselines by over 6 F1 points on key tasks.

The model is effective across various joint IE subtasks.

Abstract

Extracting structured information from unstructured text is critical for many downstream NLP applications and is traditionally achieved by closed information extraction (cIE). However, existing approaches for cIE suffer from two limitations: (i) they are often pipelines which makes them prone to error propagation, and/or (ii) they are restricted to sentence level which prevents them from capturing long-range dependencies and results in expensive inference time. We address these limitations by proposing REXEL, a highly efficient and accurate model for the joint task of document level cIE (DocIE). REXEL performs mention detection, entity typing, entity disambiguation, coreference resolution and document-level relation classification in a single forward pass to yield facts fully linked to a reference knowledge graph. It is on average 11 times faster than competitive existing approaches in a similar setting and performs competitively both when optimised for any of the individual subtasks and a variety of combinations of different joint tasks, surpassing the baselines by an average of more than 6 F1 points. The combination of speed and accuracy makes REXEL an accurate cost-efficient system for extracting structured information at web-scale. We also release an extension of the DocRED dataset to enable benchmarking of future work on DocIE, which is available at https://github.com/amazon-science/e2e-docie.