Improving few-shot relation classification with multi-scale hierarchical prototype learning

doi:10.1016/j.neunet.2025.108124

Improving few-shot relation classification with multi-scale hierarchical prototype learning

2025-09-18

0

PRE

AI

DOI：10.1016/j.neunet.2025.108124

Original

Original request for help

Save

Abstract

En 中文

• Propose a multi-scale hierarchical prototype learning method for few-shot relation classification. • Integrate inter-set, inter-class, and intra-class interaction for better prototype representation. • Incorporate relational descriptions to emulate human cognition and enhance semantic understanding. • Alleviate data sparsity and improve classification by reducing distribution differences.

Improving few-shot relation classification with multi-scale hierarchical prototype learning

Abstract

Journal

Organization

Cited Papers