MMKG: Multi-Modal Knowledge Graphs

By Ye Liu, Hui Li, Alberto García-Durán, Mathias Niepert, Daniel Oñoro-Rubio, David S. Rosenblum. Extended Semantic Web Conference, 2019

335 citations recorded by the source corpus.

We present MMKG, a collection of three knowledge graphs that contain both numerical features and (links to) images for all entities as well as entity alignments between pairs of KGs. Therefore, multi-relational link prediction and entity matching communities can benefit from this resource. We believe this data set has the potential to facilitate the development of novel multi-modal learning approaches for knowledge graphs.We validate the utility ofMMKG in the sameAs link prediction task with an extensive set of experiments. These experiments show that the task at hand benefits from learning of multiple feature types.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/d593a5830a7e7d84443473c3912b59165056d45a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Zhixu Li
2. Linhao Luo
3. Penglei Sun
4. Xindong Wu
5. Hao Peng
6. Lingfeng Zhong
7. Xueyao Jiang
8. Yanghua Xiao

In Computer Vision and Pattern Recognition

1. Zhixu Li
2. Penglei Sun
3. Xindong Wu
4. Hao Peng
5. Lingfeng Zhong
6. Xueyao Jiang
7. Yanghua Xiao
8. Qian Li

Citations in the corpus

• Multi-Modal Knowledge Graph Construction and Application: A Survey (2022, background): “Another line of work [24] elaborates a Product of Experts (PoE) model to answer queries such as (h?, sameAs, t) or (h, sameAs, t?) where h and t are from different KGs.”
• Unifying Large Language Models and Knowledge Graphs: A Roadmap (2023, background): “For example, IMGpedia [84], MMKG [85], and Richpedia [86] incorporate both the text and image information into the knowledge graphs.”
• A Survey of Knowledge Graph Reasoning on Graph Types: Static, Dynamic, and Multi-Modal (2022, neutral): “Besides,KGRmodelscanalsohelpwithAnti-frauddetection,whichisanimportanttaskinthefinancefield.”
• A Comprehensive Survey on Automatic Knowledge Graph Construction (2023, background): “MMKG [353] is a model for completing multi-modal tuples that reasons over image information, while Dost et al.”
• MG2-RAG: Multi-Granularity Graph for Multimodal Retrieval-Augmented Generation (2026, unknown): “However, most systems rely on flat vector similarity, which struggles with structured multi-hop reasoning across modalities [36,49].”
• mKG-RAG: Leveraging Multimodal Knowledge Graphs in Retrieval-Augmented Generation for Knowledge-intensive VQA (2025, unknown): “A naive strategy is to directly supply G with the corresponding images [40].”