This repository provides a comprehensive survey of research at the intersection of Knowledge Graphs (KGs) and Multi-Modal Learning (MML). It categorizes and lists papers focusing on KG-driven MML (KG4MM) and Multi-Modal KGs (MM4KG), serving as a valuable resource for researchers and practitioners in AI, NLP, and computer vision.

How It Works

The survey systematically organizes research into two primary streams: KG4MM, where KGs enhance multi-modal tasks, and MM4KG, which extends KG concepts to incorporate multiple modalities. It covers a wide spectrum of applications, including understanding, reasoning, classification, content generation, and retrieval tasks, as well as MMKG construction, fusion, inference, and reasoning.

Quick Start & Requirements

This is a survey repository, not a software package. It requires no installation or specific software dependencies to access the information.

Highlighted Details

• Comprehensive categorization of research into KG4MM and MM4KG.
• Extensive lists of papers with links to arXiv and GitHub repositories.
• Covers a broad range of tasks and applications within the KG-MML domain.
• Regularly updated with recent publications and accepted papers.

Maintenance & Community

The repository is maintained by zjukg and welcomes contributions through GitHub Issues or Pull Requests to add or update papers. Contact information for Zhuo Chen is provided.

Licensing & Compatibility

The repository itself does not specify a license. The content consists of links to research papers, which are subject to their respective licenses and publication terms.

Limitations & Caveats

As a survey, it relies on community contributions for completeness and accuracy. While regularly updated, there might be a slight delay in incorporating the very latest publications.