Mitigating task randomness in graph few-shot learning

Shuzhen Rao

doi:10.1016/j.patrec.2025.11.022

Mitigating task randomness in graph few-shot learning

2025-11-20

0

PRE

AI

DOI：10.1016/j.patrec.2025.11.022

Original

Original request for help

Save

Abstract

En 中文

• We first identify the unstudied issue of excessive task randomness in FSNC. • A novel method called TRARM, is proposed to mitigate excessive task randomness. • Progressive grouping-based sampling reduces task randomness directly by stages. • Memory-based meta-update module is designed to accumulate cross-task knowledge. • TRARM achieves state-of-the-art results on 4 widely used datasets.

Mitigating task randomness in graph few-shot learning

Abstract

Journal

Organization

Cited Papers