Zhujiu- knowledge世界知识类关系知识评测数据集

本数据集是Zhujiu-Knowledge评测平台中面向世界知识类关系知识的评测数据集，旨在评估大型语言模型（LLMs）对实体之间复杂联系的理解和推理能力。本数据集不是检验模型对单一属性的记忆，而是通过开放式问题深入探究模型对实体间联系的深层关系的掌握程度。数据以开放式问答的形式组织，核心内容来源于大规模知识库Wikidata。This dataset is a world knowledge relational knowledge evaluation dataset from the Zhujiu-Knowledge Evaluation Platform. It aims to evaluate the understanding and reasoning capabilities of Large Language Models (LLMs) regarding the complex connections between entities. Unlike merely testing the model's memorization of single attributes, this dataset deeply explores the model's mastery of the deep relational connections between entities through open-ended questions. The dataset is structured in the form of open-ended question answering, with its core content sourced from the large-scale knowledge base Wikidata.

搜集汇总

背景与挑战

背景概述

该数据集是Zhujiu-Knowledge评测平台的一部分，专门针对世界知识类关系知识设计，用于评估大型语言模型对实体间复杂联系的理解与推理能力。数据以开放式问答形式呈现，核心内容源自大规模知识库Wikidata。