DiVA-DocRE: A Discriminative and Voice-Aware Paradigm for Document-Level Relation Extraction

TL;DR

DiVA-DocRE introduces a streamlined, voice-aware approach for document-level relation extraction, significantly improving accuracy by focusing on discriminative relation identification and voice distinctions, outperforming previous methods.

Contribution

The paper presents a novel discriminative, voice-aware paradigm for document-level relation extraction that simplifies the process and enhances performance over existing approaches.

Findings

Achieved state-of-the-art results on Re-DocRED and DocRED datasets.

Effectively distinguishes active and passive voice in relation triplet extraction.

Streamlined two-step process improves efficiency and accuracy.

Abstract

The remarkable capabilities of Large Language Models (LLMs) in text comprehension and generation have revolutionized Information Extraction (IE). One such advancement is in Document-level Relation Triplet Extraction (DocRTE), a critical task in information systems that aims to extract entities and their semantic relationships from documents. However, existing methods are primarily designed for Sentence level Relation Triplet Extraction (SentRTE), which typically handles a limited set of relations and triplet facts within a single sentence. Additionally, some approaches treat relations as candidate choices integrated into prompt templates, resulting in inefficient processing and suboptimal performance when determining the relation elements in triplets. To address these limitations, we introduce a Discriminative and Voice Aware Paradigm DiVA. DiVA involves only two steps: performing document-level relation extraction (DocRE) and then identifying the subject object entities based on the relation. No additional processing is required simply input the document to directly obtain the triplets. This streamlined process more accurately reflects real-world scenarios for triplet extraction. Our innovation lies in transforming DocRE into a discriminative task, where the model pays attention to each relation and to the often overlooked issue of active vs. passive voice within the triplet. Our experiments on the Re-DocRED and DocRED datasets demonstrate state-of-the-art results for the DocRTE task.