A survey on knowledge-enhanced multimodal learning

arXiv:2211.12328v3 Announce Type: replace
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 …

abstract arxiv cs.ai cs.lg images knowledge multimodal multimodal learning multiple representation survey targeting tasks text type