GP2F: Cross-Domain Graph Prompting with Adaptive Fusion of Pre-trained Graph Neural Networks
Researchers have introduced GP2F, a novel method for cross-domain graph prompting that aims to improve the adaptation of pre-trained graph neural networks to new tasks. The method is based on theoretical analysis showing that combining a frozen branch of pre-trained knowledge with a lightweight, adapted branch for task-specific learning yields better results than using either alone. GP2F employs adaptive fusion through contrastive and topology-consistent losses, demonstrating superior performance on cross-domain few-shot node and graph classification tasks. AI
IMPACT Introduces a new technique for adapting graph neural networks to different domains, potentially improving performance in real-world applications.