yale-nlp/M3SciQA
收藏Hugging Face2025-01-13 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/yale-nlp/M3SciQA
下载链接
链接失效反馈官方服务：
资源简介：
M3SciQA是一个多模态、多文档的科学问答基准数据集，设计用于更全面地评估基础模型。该数据集包括1452个由专家标注的问题，跨越70个自然语言处理论文集群，每个集群包括一篇主要论文及其所有引用的文档，反映了通过多模态和多文档数据进行单篇论文理解的工作流程。
M3SciQA is a Multi-Modal, Multi-document Scientific Question Answering benchmark designed for a more comprehensive evaluation of foundation models. The dataset consists of 1,452 expert-annotated questions spanning 70 natural language processing (NLP) paper clusters, with each cluster representing a primary paper along with all its cited documents, reflecting the workflow of comprehending a single paper through multi-modal and multi-document data.
提供机构：
yale-nlp
