illuin-conteb/esg-reports

ConTEB ESG报告数据集是ConTEB（上下文感知文本嵌入基准）的一部分，旨在评估上下文嵌入模型的能力。该数据集聚焦于工业ESG报告，尤其是来自快餐行业的报告。数据集由30个原始文档、3702个文本块和36个查询组成，提供了一个针对上下文化嵌入的专注基准。数据集结构包括两个部分：documents和queries，其中documents包含块信息，queries包含查询信息。The ConTEB ESG Reports dataset is part of the ConTEB (Context-aware Text Embedding Benchmark), designed to evaluate the capabilities of contextual embedding models. It focuses on the theme of Industrial ESG Reports, particularly from the fast-food industry. The dataset consists of 30 original documents, 3702 chunks, and 36 queries, providing a focused benchmark for contextualized embeddings. The dataset structure includes two parts: documents and queries, with documents containing chunk information and queries containing query information.