Multi-Task Identification of Entities, Relations, and Coreference for Scientific Knowledge Graph Construction

TL;DR

This paper presents a multi-task framework for extracting entities, relations, and coreference information from scientific articles to build comprehensive scientific knowledge graphs, improving accuracy without domain-specific features.

Contribution

It introduces SciERC dataset and SciIE framework, enabling joint extraction of multiple scientific information types with shared representations, reducing errors and enhancing knowledge graph construction.

Findings

Multi-task model outperforms previous methods in scientific info extraction.

Shared span representations improve extraction accuracy.

Framework supports effective scientific knowledge graph construction.

Abstract

We introduce a multi-task setup of identifying and classifying entities, relations, and coreference clusters in scientific articles. We create SciERC, a dataset that includes annotations for all three tasks and develop a unified framework called Scientific Information Extractor (SciIE) for with shared span representations. The multi-task setup reduces cascading errors between tasks and leverages cross-sentence relations through coreference links. Experiments show that our multi-task model outperforms previous models in scientific information extraction without using any domain-specific features. We further show that the framework supports construction of a scientific knowledge graph, which we use to analyze information in scientific literature.