Tokenization, Fusion, and Augmentation: Towards Fine-grained Multi-modal Entity Representation
This work addresses the inherent incompleteness in knowledge graphs for researchers and practitioners in AI, offering an incremental improvement by enhancing fine-grained semantic details in multi-modal entity representations.
The paper tackles the problem of coarse handling of multi-modal entity information in multi-modal knowledge graph completion by introducing the MyGO framework, which tokenizes, fuses, and augments fine-grained representations, resulting in superior performance that surpasses 19 latest models on standard benchmarks.
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