Pseudo-Label Calibration Semi-supervised Multi-Modal Entity Alignment

TL;DR

This paper introduces PCMEA, a semi-supervised method for multi-modal entity alignment that enhances entity representations by filtering noise and leveraging both labeled and unlabeled data, achieving state-of-the-art results.

Contribution

The paper proposes a novel semi-supervised framework with pseudo-label calibration and mutual information maximization for improved multi-modal entity alignment.

Findings

Achieves state-of-the-art performance on two datasets.

Effectively filters modal-specific noise.

Utilizes both labeled and unlabeled data for better alignment.

Abstract

Multi-modal entity alignment (MMEA) aims to identify equivalent entities between two multi-modal knowledge graphs for integration. Unfortunately, prior arts have attempted to improve the interaction and fusion of multi-modal information, which have overlooked the influence of modal-specific noise and the usage of labeled and unlabeled data in semi-supervised settings. In this work, we introduce a Pseudo-label Calibration Multi-modal Entity Alignment (PCMEA) in a semi-supervised way. Specifically, in order to generate holistic entity representations, we first devise various embedding modules and attention mechanisms to extract visual, structural, relational, and attribute features. Different from the prior direct fusion methods, we next propose to exploit mutual information maximization to filter the modal-specific noise and to augment modal-invariant commonality. Then, we combine pseudo-label calibration with momentum-based contrastive learning to make full use of the labeled and unlabeled data, which improves the quality of pseudo-label and pulls aligned entities closer. Finally, extensive experiments on two MMEA datasets demonstrate the effectiveness of our PCMEA, which yields state-of-the-art performance.