This repository serves as a comprehensive collection of papers, code, and datasets related to knowledge graph (KG) reasoning. It aims to provide a structured overview of the field, catering to researchers and practitioners interested in various KG reasoning tasks, including static, temporal, and multi-modal KGs. The project offers a curated list of influential works and datasets, facilitating exploration and benchmarking of KG reasoning techniques.

How It Works

The repository categorizes KG reasoning approaches into distinct types: Static, Temporal, and Multi-Modal. Within each category, it further breaks down methods by underlying techniques such as translational models, tensor decomposition, neural networks (CNNs, GNNs, Transformers), and path-based or rule-based methods. Datasets are also meticulously listed with their statistics, enabling users to select appropriate benchmarks for their research.

Quick Start & Requirements

This repository is a curated list and does not have direct installation or execution commands. It links to external resources for papers and code.

Highlighted Details

• Comprehensive categorization of KG reasoning methods across static, temporal, and multi-modal KGs.
• Extensive lists of relevant datasets with detailed statistics (entities, relations, triplets).
• Includes links to papers and code for a wide array of KG reasoning models, from foundational to recent advancements.
• Features a dedicated section for survey papers, offering high-level overviews of specific sub-fields.

Maintenance & Community

The repository is maintained by LIANGKE23. Contact information (liangke200694@126.com, liangke200694@gmail.com) is provided for issues and contributions. Users are encouraged to star the repository and submit new papers or codes.

Licensing & Compatibility

The repository itself is a collection of links and does not impose a specific license. However, the underlying code and datasets linked within will have their own respective licenses, which users must adhere to.

Limitations & Caveats

This repository is a curated list and does not provide executable code or integrated tools. Users must navigate to external links for papers and code implementations, and manage their own environments and dependencies. The breadth of coverage means some niche or very recent advancements might not yet be included.