A Cross-graph Tuning-free GNN Prompting Framework
This addresses the challenge of generalizing GNN prompting across graphs without parameter tuning, offering a plug-and-play solution for graph learning.
The paper tackled the problem of adapting GNNs across tasks and graphs without extensive retraining by introducing a cross-graph tuning-free prompting framework, achieving an average accuracy gain of 30.8% and up to 54% in few-shot prediction tasks.
GNN prompting aims to adapt models across tasks and graphs without requiring extensive retraining. However, most existing graph prompt methods still require task-specific parameter updates and face the issue of generalizing across graphs, limiting their performance and undermining the core promise of prompting. In this work, we introduce a Cross-graph Tuning-free Prompting Framework (CTP), which supports both homogeneous and heterogeneous graphs, can be directly deployed to unseen graphs without further parameter tuning, and thus enables a plug-and-play GNN inference engine. Extensive experiments on few-shot prediction tasks show that, compared to SOTAs, CTP achieves an average accuracy gain of 30.8% and a maximum gain of 54%, confirming its effectiveness and offering a new perspective on graph prompt learning.