MACO: A Modality Adversarial and Contrastive Framework for Modality-missing Multi-modal Knowledge Graph Completion

TL;DR

This paper introduces MACO, a novel framework that addresses missing modality issues in multi-modal knowledge graph completion by using adversarial training and contrastive learning to generate and incorporate missing modality features, improving model performance.

Contribution

The paper presents a new adversarial and contrastive framework, MACO, that effectively handles missing modalities in MMKGC, enhancing existing models' capabilities.

Findings

Achieves state-of-the-art results on benchmark datasets.

Demonstrates robustness across various MMKGC models.

Provides a versatile framework adaptable to different models.

Abstract

Recent years have seen significant advancements in multi-modal knowledge graph completion (MMKGC). MMKGC enhances knowledge graph completion (KGC) by integrating multi-modal entity information, thereby facilitating the discovery of unobserved triples in the large-scale knowledge graphs (KGs). Nevertheless, existing methods emphasize the design of elegant KGC models to facilitate modality interaction, neglecting the real-life problem of missing modalities in KGs. The missing modality information impedes modal interaction, consequently undermining the model's performance. In this paper, we propose a modality adversarial and contrastive framework (MACO) to solve the modality-missing problem in MMKGC. MACO trains a generator and discriminator adversarially to generate missing modality features that can be incorporated into the MMKGC model. Meanwhile, we design a cross-modal contrastive loss to improve the performance of the generator. Experiments on public benchmarks with further explorations demonstrate that MACO could achieve state-of-the-art results and serve as a versatile framework to bolster various MMKGC models. Our code and benchmark data are available at https://github.com/zjukg/MACO.