A survey on knowledge-enhanced multimodal learning

TL;DR

This survey reviews how integrating knowledge graphs with vision-language models enhances understanding, explainability, and performance in multimodal learning tasks involving images and text.

Contribution

It unifies vision-language representation learning with knowledge graph integration, providing a taxonomy and comprehensive analysis of knowledge-enhanced VL models.

Findings

Knowledge graphs fill gaps in commonsense and factual understanding.

Knowledge integration improves explainability and fairness.

VL models benefit from explicit knowledge sources for better performance.

Abstract

Multimodal learning has been a field of increasing interest, aiming to combine various modalities in a single joint representation. Especially in the area of visiolinguistic (VL) learning multiple models and techniques have been developed, targeting a variety of tasks that involve images and text. VL models have reached unprecedented performances by extending the idea of Transformers, so that both modalities can learn from each other. Massive pre-training procedures enable VL models to acquire a certain level of real-world understanding, although many gaps can be identified: the limited comprehension of commonsense, factual, temporal and other everyday knowledge aspects questions the extendability of VL tasks. Knowledge graphs and other knowledge sources can fill those gaps by explicitly providing missing information, unlocking novel capabilities of VL models. In the same time, knowledge graphs enhance explainability, fairness and validity of decision making, issues of outermost importance for such complex implementations. The current survey aims to unify the fields of VL representation learning and knowledge graphs, and provides a taxonomy and analysis of knowledge-enhanced VL models.