Skip to content

The benchmark and datasets of the ICML 2024 paper "VisionGraph: Leveraging Large Multimodal Models for Graph Theory Problems in Visual Context"

Notifications You must be signed in to change notification settings

HITsz-TMG/VisionGraph

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

19 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

VisionGraph: Leveraging Large Multimodal Models for Graph Theory Problems in Visual Context

✨ Overview

This repository contains the official implementation of our ICML 2024 paper, VisionGraph: Leveraging Large Multimodal Models for Graph Theory Problems in Visual Context.

VisionGraph, is a benchmark used to explore the capabilities of advanced LMMs in solving multimodal graph theory problems. It encompasses eight graph problem tasks, from connectivity to shortest path problems. To step forward in this direction, we introduce a Description-Program-Reasoning (DPR) chain to enhance the logical accuracy of reasoning processes through graphical structure description generation and algorithm-aware multi-step reasoning. All prompts, datasets, checkpoints and evaluation methods related to VisionGraph and DPR are available in this repo for easy access and utilization.

Graph Data for VisionGraph can be accessed here. Graph Understanding Data for Training can be accessed here.

If you have any question, please feel free to contact me via email at [email protected] or submit your issue in the repository.

🔥 News

[24.05.08] We have updated our paper: VisionGraph.

[24.05.11] We release the prompts, datasets, checkpoints and evaluation methods related to VisionGraph and DPR.

🚀 Illustration

Here, you can see the detailed introduction of VisionGraph and DPR.

VisionGraph

DPR

Citation

@article{li2024visiongraph,
    title={VisionGraph: Leveraging Large Multimodal Models for Graph Theory Problems in Visual Context}, 
    author={Yunxin Li and Baotian Hu and Haoyuan Shi and Wei Wang and Longyue Wang and Min Zhang},
    journal={arXiv preprint arXiv:2405.04950},
    year={2024},
}

About

The benchmark and datasets of the ICML 2024 paper "VisionGraph: Leveraging Large Multimodal Models for Graph Theory Problems in Visual Context"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •  

Languages