Unsupervised Knowledge Graph Construction and Event-centric Knowledge Infusion for Scientific NLI

TL;DR

This paper introduces an unsupervised method to construct a scientific knowledge graph and an event-centric knowledge infusion technique to enhance natural language inference in scientific texts, achieving state-of-the-art results.

Contribution

It presents a novel unsupervised approach for building a scientific knowledge graph and an event-centric knowledge infusion method to improve NLI performance in scientific domains.

Findings

Achieves state-of-the-art NLI performance on scientific texts.

Demonstrates effectiveness and reliability of the unsupervised SKG.

Reduces reliance on labeled data and improves domain adaptation.

Abstract

With the advance of natural language inference (NLI), a rising demand for NLI is to handle scientific texts. Existing methods depend on pre-trained models (PTM) which lack domain-specific knowledge. To tackle this drawback, we introduce a scientific knowledge graph to generalize PTM to scientific domain. However, existing knowledge graph construction approaches suffer from some drawbacks, i.e., expensive labeled data, failure to apply in other domains, long inference time and difficulty extending to large corpora. Therefore, we propose an unsupervised knowledge graph construction method to build a scientific knowledge graph (SKG) without any labeled data. Moreover, to alleviate noise effect from SKG and complement knowledge in sentences better, we propose an event-centric knowledge infusion method to integrate external knowledge into each event that is a fine-grained semantic unit in sentences. Experimental results show that our method achieves state-of-the-art performance and the effectiveness and reliability of SKG.