milIE: Modular & Iterative Multilingual Open Information Extraction

TL;DR

milIE is a modular, iterative neural OpenIE system that improves extraction accuracy by sequentially extracting triple slots, integrating rule-based methods, and demonstrating superior multilingual performance, including new datasets for Arabic and Galician.

Contribution

The paper introduces milIE, a novel iterative and modular neural OpenIE system that enhances extraction performance and supports integration with rule-based methods across multiple languages.

Findings

milIE outperforms state-of-the-art systems on multiple languages

The iterative approach improves extraction accuracy

First OpenIE datasets for Arabic and Galician are provided

Abstract

Open Information Extraction (OpenIE) is the task of extracting (subject, predicate, object) triples from natural language sentences. Current OpenIE systems extract all triple slots independently. In contrast, we explore the hypothesis that it may be beneficial to extract triple slots iteratively: first extract easy slots, followed by the difficult ones by conditioning on the easy slots, and therefore achieve a better overall extraction. Based on this hypothesis, we propose a neural OpenIE system, milIE, that operates in an iterative fashion. Due to the iterative nature, the system is also modular -- it is possible to seamlessly integrate rule based extraction systems with a neural end-to-end system, thereby allowing rule based systems to supply extraction slots which milIE can leverage for extracting the remaining slots. We confirm our hypothesis empirically: milIE outperforms SOTA systems on multiple languages ranging from Chinese to Arabic. Additionally, we are the first to provide an OpenIE test dataset for Arabic and Galician.