吉林大学学报（理学版）2026，Vol.64Issue(2)：291-300,10.DOI:10.13413/j.cnki.jdxblxb.2024525
基于大语言模型与适配器驱动的知识图谱补全算法
Knowledge Graph Completion Algorithm Based on Large Language Models and Adapter Driver
摘要
Abstract
Aiming at the problems that the knowledge graph completion method based on Transformer as the backbone network,included parameter redundancy in feed-forward networks,difficulties in identifying tail entities under commonsense scenarios,and embedding biases in contrastive learning,we proposed an adapter-enhanced knowledge graph completion algorithm that integrated large language models with multi-positive sample contrastive learning.The algorithm reduced redundant features by introducing multi-head adapters in feed-forward network,and utilized large language models to enhance commonsense reasoning ability.At the same time,it corrected embedding biases through multi-positive sample contrastive learning.Experimental results show that,compared to the current state-of-the-art models,the algorithm improves MRR by 5.4％and 9.2％on WN18RR and FB15k-237 datasets,respectively,and by 3.6％and 6.7％in transductive and inductive settings on the more complex Wikidata5M dataset,respectively,and demonstrates superior generalization ability under low-resource and complex scenarios.关键词
知识图谱补全/知识图谱/大语言模型/对比学习/适配器学习Key words
knowledge graph completion/knowledge graph/large language model/contrastive learning/adapter learning分类
信息技术与安全科学引用本文复制引用
姜昀奇,韩晓同,田原..基于大语言模型与适配器驱动的知识图谱补全算法[J].吉林大学学报（理学版）,2026,64(2):291-300,10.基金项目
国家重点研发计划项目(批准号:2023YFF0905400)、国家自然科学基金"叶企孙"科学基金(批准号:U2341229)和吉林省发展和改革委员会基金(批准号:2024C003). （批准号:2023YFF0905400）
