arXiv 2510.077362025-10-09cs.CL

多语言知识图谱补全：通过高效的多语言知识共享

Multilingual Knowledge Graph Completion via Efficient Multilingual Knowledge Sharing

Faculty of Information Engineering and Automation, Kunming University of Science and Technology（昆明理工大学信息工程与自动化学院）；Yunnan Key Laboratory of Artificial Intelligence（云南人工智能重点实验室）；The Key Laboratory of Cognition and Decision Intelligence for Complex Systems, Institute of Automation, Chinese Academy of Sciences（复杂系统认知与决策智能重点实验室，中国科学院自动化研究所）；School of Artificial Intelligence, University of Chinese Academy of Science（中国科学院大学人工智能学院）

Cunli Mao, Xiaofei Gao, Ran Song, Shizhu He, Shengxiang Gao, Kang Liu, Zhengtao Yu

备注：: EMNLP 2025, Findings, Long Paper

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出KL-GMoE与IER框架，通过多语言共享知识提升多语言知识图谱补全性能，在Hits@1/3/10上分别提升5.47%、3.27%和1.01%。

摘要：

基于大型语言模型（LLMs）的多语言知识图谱补全（MKGC）旨在利用LLMs的多语言理解能力预测缺失的事实，从而提高多语言知识图谱（KGs）的完整性。然而，现有的MKGC研究未能充分利用LLMs的多语言能力，并忽略了跨语言知识的可共享性。在本文中，我们提出了一种新颖的MKGC框架，通过两个组件利用多语言共享知识显著提升性能：知识级分组混合专家（KL-GMoE）和迭代实体重排序（IER）。KL-GMoE高效地建模共享知识，而IER则显著增强其利用率。为评估我们的框架，我们构建了一个包含5种语言的mKG数据集，并与现有的最先进（SOTA）MKGC方法进行了全面的对比实验。实验结果表明，与SOTA MKGC方法相比，我们的框架在Hits@1、Hits@3和Hits@10指标上分别实现了5.47%、3.27%和1.01%的提升。进一步的实验分析揭示了在未见语言和不平衡语言设置下知识共享的特性。我们已在https://github.com/gaoxiaofei07/KL-GMoE上发布了我们的数据集和代码。

英文摘要：

Large language models (LLMs) based Multilingual Knowledge Graph Completion (MKGC) aim to predict missing facts by leveraging LLMs' multilingual understanding capabilities, improving the completeness of multilingual knowledge graphs (KGs). However, existing MKGC research underutilizes the multilingual capabilities of LLMs and ignores the shareability of cross-lingual knowledge. In this paper, we propose a novel MKGC framework that leverages multilingual shared knowledge to significantly enhance performance through two components: Knowledge-level Grouped Mixture of Experts (KL-GMoE) and Iterative Entity Reranking (IER). KL-GMoE efficiently models shared knowledge, while IER significantly enhances its utilization. To evaluate our framework, we constructed a mKG dataset containing 5 languages and conducted comprehensive comparative experiments with existing state-of-the-art (SOTA) MKGC method. The experimental results demonstrate that our framework achieves improvements of 5.47%, 3.27%, and 1.01% in the Hits@1, Hits@3, and Hits@10 metrics, respectively, compared with SOTA MKGC method. Further experimental analysis revealed the properties of knowledge sharing in settings of unseen and unbalanced languages. We have released the dataset and code for our work on https://github.com/gaoxiaofei07/KL-GMoE.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。