arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2409.15902cs.CL
收藏

Konstruktor:一个用于简单知识图谱问答的强基线

Konstruktor: A Strong Baseline for Simple Knowledge Graph Question Answering

Skolkovo Institute of Science and Technology(斯科尔科沃科学技术学院);Artificial Intelligence Research Institute(人工智能研究院);Nazarbayev University(纳扎尔巴耶夫大学)

Maria Lysyuk, Mikhail Salnikov, Pavel Braslavski, Alexander Panchenko

AI总结:

本文提出Konstruktor方法,通过结合语言模型与知识图谱,将简单知识图谱问答分解为实体抽取链接、关系预测和图谱查询三步,在四个数据集上取得强基线效果。

摘要:

尽管“谁是《灰姑娘》的作者?”这类简单问题是最受欢迎的提问类型之一,但它们仍未被完全解决。令人惊讶的是,即使是最强大的现代大型语言模型在处理此类问题时也容易出错,尤其是在处理罕见实体时。同时,由于答案可能距离问题实体仅一跳,因此可以尝试开发一种使用结构化知识图谱来回答此类问题的方法。在本文中,我们介绍了 Konstruktor——一种高效且稳健的方法,它将该问题分解为三个步骤:(i) 实体抽取与实体链接,(ii) 关系预测,以及 (iii) 查询知识图谱。我们的方法结合了语言模型和知识图谱,利用了前者的强大能力和后者的可解释性。我们实验了两种命名实体识别与实体链接方法以及几种关系检测技术。我们表明,对于工作流中最具挑战性的关系检测步骤,关系分类/生成与排序的组合优于其他方法。我们报告了 Konstruktor 在四个数据集上的优异结果。

英文摘要:

While being one of the most popular question types, simple questions such as "Who is the author of Cinderella?", are still not completely solved. Surprisingly, even the most powerful modern Large Language Models are prone to errors when dealing with such questions, especially when dealing with rare entities. At the same time, as an answer may be one hop away from the question entity, one can try to develop a method that uses structured knowledge graphs (KGs) to answer such questions. In this paper, we introduce Konstruktor - an efficient and robust approach that breaks down the problem into three steps: (i) entity extraction and entity linking, (ii) relation prediction, and (iii) querying the knowledge graph. Our approach integrates language models and knowledge graphs, exploiting the power of the former and the interpretability of the latter. We experiment with two named entity recognition and entity linking methods and several relation detection techniques. We show that for relation detection, the most challenging step of the workflow, a combination of relation classification/generation and ranking outperforms other methods. We report Konstruktor's strong results on four datasets.

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。

↑