TA-KAND: Advancing Few-shot Knowledge Graph Completion with Diffusion

This research explores a novel approach to few-shot knowledge graph completion using a two-stage attention mechanism and a U-KAN based diffusion model. The application of diffusion models to knowledge graph completion is a promising area with potential for improving the accuracy of inferring relationships from sparse data.

• •Proposes a novel approach to few-shot knowledge graph completion.
• •Employs a two-stage attention mechanism for enhanced performance.
• •Utilizes a U-KAN based diffusion model for relation inference.