Summary

This repository serves as a comprehensive, curated collection of resources focused on graph-based agent memory. It targets researchers, engineers, and power users seeking to understand, implement, or contribute to the field of AI agents with sophisticated memory capabilities. The primary benefit is providing a structured and up-to-date overview of a rapidly evolving research area, consolidating key papers, benchmarks, and projects.

How It Works

The project functions as a living bibliography and knowledge hub, meticulously organizing research on graph-based agent memory. It employs a detailed taxonomy covering memory extraction, storage, retrieval, and evolution, categorizing numerous academic papers, benchmark datasets, and open-source projects. This structured approach allows users to efficiently navigate and identify relevant advancements, techniques, and tools within this specialized domain.

Quick Start & Requirements

This repository is a curated list of resources and does not require installation or execution. It serves as a reference guide to external papers, benchmarks, and open-source projects related to graph-based agent memory.

Highlighted Details

• Features a detailed taxonomy of graph-based agent memory, categorizing techniques by extraction, storage, retrieval, and evolution.
• Includes extensive links to academic papers, including a primary survey paper: "Graph-based Agent Memory: Taxonomy, Techniques, and Applications."
• Compiles a wide array of benchmarks and open-source projects relevant to agent memory research.
• Aims for continuous updates to incorporate the latest research and developments in the field.

Maintenance & Community

The repository is maintained by DEEP-PolyU and welcomes community contributions. Interested parties can reach out via email at chang.yang@connect.polyu.hk or qinggang.zhang@polyu.edu.hk, or by submitting issues and pull requests.

Licensing & Compatibility

The repository itself does not specify a license. Users are advised to consult the individual licenses of the linked papers, benchmarks, and open-source projects for their respective terms of use and compatibility.

Limitations & Caveats

As a curated list, the repository's comprehensiveness is dependent on ongoing community contributions and maintainer efforts. It acts as a pointer to external resources rather than a self-contained software tool, meaning users must engage with each linked project individually.