Graph Prompting for Graph Learning Models: Recent Advances and Future Directions

Abstract

Graph learning models have demonstrated great prowess in learning expressive representations from large-scale graph data in a wide variety of real-world scenarios. As a prevalent strategy for training powerful graph learning models, the "pre-training, adaptation" scheme first pre-trains graph learning models on unlabeled graph data in a self-supervised manner and then adapts them to specific downstream tasks. During the adaptation phase, graph prompting emerges as a promising approach that learns trainable prompts while keeping the pre-trained graph learning models unchanged. This tutorial will cover recent advancements in graph prompting including:

• Representative graph pre-training methods that serve as the foundation step of graph prompting
• Mainstream techniques in graph prompting and elaborate on how they design learnable prompts for graph prompting
• Real-world applications of graph prompting from different domains
• Open challenges in existing studies with promising future directions in this field

Presenters

Xingbo Fu

University of Virginia

Zehong Wang

University of Notre Dame

Jundong Li

University of Virginia

Organizers

• Xingbo Fu, PhD student, University of Virginia
• Zehong Wang, PhD student, University of Notre Dame
• Zihan Chen, PhD student, University of Virginia
• Jiazheng Li, PhD student, University of Connecticut
• Yaochen Zhu, PhD student, University of Virginia
• Zhenyu Lei, PhD student, University of Virginia
• Cong Shen, Associate Professor, University of Virginia
• Yanfang Ye, Galassi Family Collegiate Professor, University of Notre Dame
• Chuxu Zhang, Associate Professor, University of Connecticut
• Jundong Li, Associate Professor, University of Virginia