Skip to content

Add a Paper GITA #3

Description

@WEIYanbin1999

Hi, we'd like to add a related a paper to the paper list, GITA: Graph to Image-Text Integration for Vision-Language Graph Reasoning, which was accepted by NeurIPS 2024.
Paper.
GITA is the first work to explore and establish the promising vision-language question answering on graph-related reasoning area. It systematically enable VLMs for general language-based graph reasoning tasks.
In this paper, we provide new pre-trained VLM model weights for graph reasoning:
Model: GITA-7B/13B, the model weights are in both Github repo and Model weight huggingface.
We also proposed the first dataset GVLQA for vision-language graph reasoning, they are VQA image-text-query-answer pairs for graph reasoning. GVLQA Datasets.
Wish your research smoothly. Looking forward to your reply!

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions