LightCAKE: A Lightweight Framework for Context-Aware Knowledge Graph Embedding

TL;DR

LightCAKE is a lightweight, flexible framework for context-aware knowledge graph embedding that improves efficiency and effectiveness without increasing model complexity.

Contribution

It introduces a novel, parameter-efficient method to explicitly model graph context and integrate it into embeddings, applicable to various KGE models.

Findings

Achieves superior results on benchmark datasets

Demonstrates high efficiency and effectiveness

Compatible with multiple KGE models

Abstract

Knowledge graph embedding (KGE) models learn to project symbolic entities and relations into a continuous vector space based on the observed triplets. However, existing KGE models cannot make a proper trade-off between the graph context and the model complexity, which makes them still far from satisfactory. In this paper, we propose a lightweight framework named LightCAKE for context-aware KGE. LightCAKE explicitly models the graph context without introducing redundant trainable parameters, and uses an iterative aggregation strategy to integrate the context information into the entity/relation embeddings. As a generic framework, it can be used with many simple KGE models to achieve excellent results. Finally, extensive experiments on public benchmarks demonstrate the efficiency and effectiveness of our framework.