Leveraging Large Language Models for Node Generation in Few-Shot Learning on Text-Attributed Graphs (AAAI 2025)

LLM4NG

Environment Settings

python==3.8.0
torch==1.12.0
numpy==1.24.3
scikit_learn==1.1.1
torch-cluster==1.6.0
torch-geometric==2.3.1
torch-scatter==2.1.0
torch-sparse==0.6.16
torch-spline-conv==1.2.1

Usage

You can use the following commend to run edge predicitor;

python edge.py --dataset cora --model_type Edge

You can use the following commend to get node classifcation result; If you don't want to use the adjacency matrix you can set the parameter lam to -1.

python LLM4NG.py --dataset cora --model_type Node

Dataset file

Since the ogbn-arxiv dataset exceeds the size limit in github, it has been uploaded to Baidu Cloud:

url: https://pan.baidu.com/s/15nAPDlPYDJeJzY5Mgq2amQ

pwd: piw9

Google Drive: https://drive.google.com/drive/folders/1msUpRr2JfK80kp9fdGMG5PFlIPwi4Re-?usp=drive_link