通过KG-BERT集成知识图谱以增强下一代语言模型：扩展Claude、Mistral IA和GPT-4
计算与语言
2025-12-12 v1
摘要
像Claude、Mistral IA和GPT-4这样的大语言模型（LLM）在自然语言处理方面表现出色，但缺乏结构化知识，导致事实性错误。我们通过KG-BERT集成知识图谱（KG）来增强其 grounding 与推理能力。实验表明，在问答和实体链接等知识密集型任务上取得显著提升。该方法提高了事实可靠性，使下一代语言模型更具情境感知能力。
引用
@article{arxiv.2512.10440,
title = {Enhancing Next-Generation Language Models with Knowledge Graphs: Extending Claude, Mistral IA, and GPT-4 via KG-BERT},
author = {Nour El Houda Ben Chaabene and Hamza Hammami},
journal= {arXiv preprint arXiv:2512.10440},
year = {2025}
}
备注
This paper was accepted and scheduled for inclusion in the ICALT 2025 proceedings but was ultimately not published due to absence from the conference presentation. It appears in the official program booklet. Conference: 2025 IEEE International Conference on Advanced Learning Technologies (ICALT)