Awesome-Graph-Prompt

A collection of AWESOME things about performing Pre-training and Prompt on Graphs.

Recently, the workflow of "pre-training and fine-tuning" has been proved less effective and efficient when applied to diverse graph downstream tasks. Inspired by the prompt learning in natural language processing (NLP), the "pre-training and prompting" workflow has emerged as a promising solution. This repo aims to provide a curated list of research papers that explore the prompting on graphs.

Table of Contents

• Awesome-Graph-Prompt
  • Table of Contents
  • Survey
  • GNN Prompting Papers
  • Application Papers
    • Recommender Systems
    • Text Attributed Graphs
    • Question Answering
    • Fake News Detection
    • Fraud Detection
    • OOD Detection
  • Ohter Resources
  • Contributing

Survey

• A Survey of Graph Prompting Methods: Techniques, Applications, and Challenges (May 2023, arXiv) [Paper]

GNN Prompting Papers

• One for All: Towards Training One Graph Model for All Classification Tasks (September 2023, arXiv) [Paper] [Code]
• Deep Prompt Tuning for Graph Transformers (September 2023, arXiv) [Paper)
• Graph Neural Prompting with Large Language Models (September 2023, arXiv) [Paper]
• Universal Prompt Tuning for Graph Neural Networks (NeurIPS'2023) [Paper]
• Virtual Node Tuning for Few-shot Node Classification (KDD'2023) [Paper]
• All in One: Multi-Task Prompting for Graph Neural Networks (KDD'2023 Best Paper Award 🌟) [Paper] [Code]
• PRODIGY: Enabling In-context Learning Over Graphs (May 2023, arXiv) [Paper]
• GraphPrompt: Unifying Pre-Training and Downstream Tasks for Graph Neural Networks (WWW'2023) [Paper] [Code]
• SGL-PT: A Strong Graph Learner with Graph Prompt Tuning (Feb 2023, arXiv) [Paper]
• GPPT: Graph Pre-training and Prompt Tuning to Generalize Graph Neural Networks (KDD'2022) [Paper] [Code]

Application Papers

Recommender Systems

• An Empirical Study Towards Prompt-Tuning for Graph Contrastive Pre-Training in Recommendations (NeurIPS'2023)
• Contrastive Graph Prompt-tuning for Cross-domain Recommendation (August 2023, arXiv) [Paper]

Text Attributed Graphs

• Prompt-based Node Feature Extractor for Few-shot Learning on Text-Attributed Graphs (September 2023, arXiv) [Paper]
• Prompt-Based Zero- and Few-Shot Node Classification: A Multimodal Approach (July 2023, arXiv) [Paper]
• Prompt Tuning on Graph-augmented Low-resource Text Classification (July 2023, arXiv) [Paper] [Code]
• Augmenting Low-Resource Text Classification with Graph-Grounded Pre-training and Prompting (SIGIR'2023) [Paper] [Code]

Question Answering

• Knowledge Graph Prompting for Multi-Document Question Answering (August 2023, arXiv) [Paper] [Code]

Fake News Detection

• Prompt-and-Align: Prompt-Based Social Alignment for Few-Shot Fake News Detection (CIKM'2023) [Paper] [Code]

Fraud Detection

• Voucher Abuse Detection with Prompt-based Fine-tuning on Graph Neural Networks (CIKM'2023) [Paper]

OOD Detection

• A Data-centric Framework to Endow Graph Neural Networks with Out-Of-Distribution Detection Ability (KDD'2023) [Paper] [Code]

Other Resources

• A Chinese Blog on Graph Prompting (including GPPT, GraphPrompt, All in One, etc) [Link]

Contributing

👍 Contributions to this repository are welcome!

If you have come across relevant resources, feel free to open an issue or submit a pull request.

- paper_name (***journal***) [[Paper](link)] [[Code](link)]