IBMEA: Exploring Variational Information Bottleneck for Multi-modal Entity Alignment

TL;DR

This paper introduces IBMEA, a novel approach using variational information bottleneck to improve multi-modal entity alignment by filtering irrelevant information and effectively integrating modal-specific representations, leading to superior performance.

Contribution

The paper proposes a variational information bottleneck framework with modal-specific encoders and regularizers for enhanced multi-modal entity alignment, addressing redundancy and noise issues.

Findings

Outperforms previous state-of-the-art methods on multiple datasets.

Shows robustness in low-resource and high-noise scenarios.

Effectively suppresses irrelevant information in multi-modal representations.

Abstract

Multi-modal entity alignment (MMEA) aims to identify equivalent entities between multi-modal knowledge graphs (MMKGs), where the entities can be associated with related images. Most existing studies integrate multi-modal information heavily relying on the automatically-learned fusion module, rarely suppressing the redundant information for MMEA explicitly. To this end, we explore variational information bottleneck for multi-modal entity alignment (IBMEA), which emphasizes the alignment-relevant information and suppresses the alignment-irrelevant information in generating entity representations. Specifically, we devise multi-modal variational encoders to generate modal-specific entity representations as probability distributions. Then, we propose four modal-specific information bottleneck regularizers, limiting the misleading clues in refining modal-specific entity representations. Finally, we propose a modal-hybrid information contrastive regularizer to integrate all the refined modal-specific representations, enhancing the entity similarity between MMKGs to achieve MMEA. We conduct extensive experiments on two cross-KG and three bilingual MMEA datasets. Experimental results demonstrate that our model consistently outperforms previous state-of-the-art methods, and also shows promising and robust performance in low-resource and high-noise data scenarios.