Editing Language Model-based Knowledge Graph Embeddings

TL;DR

This paper introduces a novel task for editing language model-based knowledge graph embeddings to enable quick, data-efficient updates without retraining, and proposes a new baseline model called KGEditor that outperforms existing methods.

Contribution

The paper defines the task of editing language model-based KG embeddings and presents a new dataset and a strong baseline model for effective fact editing.

Findings

KGEditor effectively updates specific facts without affecting overall performance.

Existing models show limited ability to handle the proposed editing task.

New datasets demonstrate the challenge and importance of knowledge editing in KG embeddings.

Abstract

Recently decades have witnessed the empirical success of framing Knowledge Graph (KG) embeddings via language models. However, language model-based KG embeddings are usually deployed as static artifacts, making them difficult to modify post-deployment without re-training after deployment. To address this issue, we propose a new task of editing language model-based KG embeddings in this paper. This task is designed to facilitate rapid, data-efficient updates to KG embeddings without compromising the performance of other aspects. We build four new datasets: E-FB15k237, A-FB15k237, E-WN18RR, and A-WN18RR, and evaluate several knowledge editing baselines demonstrating the limited ability of previous models to handle the proposed challenging task. We further propose a simple yet strong baseline dubbed KGEditor, which utilizes additional parametric layers of the hypernetwork to edit/add facts. Our comprehensive experimental results reveal that KGEditor excels in updating specific facts without impacting the overall performance, even when faced with limited training resources. Code and datasets are available in https://github.com/zjunlp/PromptKG/tree/main/deltaKG.