Structure-aware and parameter-efficient framework for editing LM-based knowledge graph embeddings

Mingjie Wang

doi:10.1016/j.knosys.2026.116976

Structure-aware and parameter-efficient framework for editing LM-based knowledge graph embeddings

2026-09-05

0

PRE

AI

DOI：10.1016/j.knosys.2026.116976

Original

Original request for help

Save

Abstract

En 中文

• A unified framework integrates the PLM, editing branch, and hypernetwork. • Three modules enable precise, controllable, and efficient knowledge updates. • Our method improves editing reliability with much lower computational cost.

Keywords:

LM-based KGE

Knowledge editing

Parameter-efficient learning

Structure-aware and parameter-efficient framework for editing LM-based knowledge graph embeddings

Abstract

Journal

Organization