MindMap: Knowledge Graph Prompting Sparks Graph of Thoughts in Large Language Models

This is the official codebase of the MindMap :snowflake: framework for eliciting the graph-of-thoughts reasoning capability in LLMs, proposed in MindMap: Knowledge Graph Prompting Sparks Graph of Thoughts in Large Language Models.

This paper has been accepted by ACL'24.

Overview

We present MindMap, a plug-and-play prompting approach, which enables LLMs to comprehend graphical inputs to build their own mind map that supports evidence-grounded generation. Here is an overview of our model architecture:

Run MindMap

As the chatdoctor5k dataset for example. First, you need to create a Blank Sandbox on https://sandbox.neo4j.com/, click "connect via drivers", find your url and user password. Then replace the following parts in MindMap.py:

uri = "Your_url"
username = "Your_user"     
password = "Your_password"

Note that the data of CMCKG is too large, and it will take about two days to wait. We recommend clicking "extend your project" in neo4j sandbox. But don't worry, the EMCKG used by chatdoctor5k will be ready to build on your facility in no time. Then, don't forget to replace your openai_key in MindMap.py.

python MindMap.py

Citation

If you find this paper interesting, please consider cite it through

@inproceedings{wen2023mindmap,
  title={MindMap: Knowledge Graph Prompting Sparks Graph of Thoughts in Large Language Models},
  author={Wen, Yilin and Wang, Zifeng and Sun, Jimeng},
  booktitle={Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics},
  year={2024}
}