Tokenization, Fusion, and Augmentation: Towards Fine-grained Multi-modal Entity Representation

TL;DR

This paper introduces MyGO, a novel framework that tokenizes, fuses, and augments fine-grained multi-modal entity representations to improve knowledge graph completion, outperforming recent models by leveraging detailed semantic interactions.

Contribution

MyGO is the first to tokenize multi-modal entity information for fine-grained representation learning in MMKGC, significantly enhancing performance through contrastive augmentation.

Findings

Outperforms 19 recent models on standard benchmarks.

Utilizes tokenization for detailed multi-modal feature extraction.

Employs contrastive learning to emphasize entity specificity.

Abstract

Multi-modal knowledge graph completion (MMKGC) aims to discover unobserved knowledge from given knowledge graphs, collaboratively leveraging structural information from the triples and multi-modal information of the entities to overcome the inherent incompleteness. Existing MMKGC methods usually extract multi-modal features with pre-trained models, resulting in coarse handling of multi-modal entity information, overlooking the nuanced, fine-grained semantic details and their complex interactions. To tackle this shortfall, we introduce a novel framework MyGO to tokenize, fuse, and augment the fine-grained multi-modal representations of entities and enhance the MMKGC performance. Motivated by the tokenization technology, MyGO tokenizes multi-modal entity information as fine-grained discrete tokens and learns entity representations with a cross-modal entity encoder. To further augment the multi-modal representations, MyGO incorporates fine-grained contrastive learning to highlight the specificity of the entity representations. Experiments on standard MMKGC benchmarks reveal that our method surpasses 19 of the latest models, underlining its superior performance. Code and data can be found in https://github.com/zjukg/MyGO