Multi-modal Knowledge Graphs: Evolution, Methods, and Opportunities
Abstract: Knowledge Graphs (KGs) are pivotal in advancing AI applications, and their extension into multi-modal dimensions (i.e., MMKGs) is opening new avenues for innovation. This survey systematically defines MMKGs, charts their construction progress, and analyzes existing MMKG-related tasks. We provide detailed task definitions, evaluation benchmarks, and insights into significant breakthroughs, while also discussing current challenges and highlighting emerging trends in the field.
Paper Type: Long
Research Area: NLP Applications
Research Area Keywords: multimodal applications, knowledge graphs, graph-based methods
Contribution Types: Surveys
Languages Studied: English
Submission Number: 631
Loading