ibm-research/REAL-MM-RAG_FinSlides

REAL-MM-RAG-Bench是一个面向现实世界的多模态检索基准，旨在在可靠、具有挑战性和现实的环境中对检索模型进行评估。该数据集包含多种模态的文档，如文本、表格和图形，以测试模型根据自然语言查询检索相关信息的能力。

REAL-MM-RAG-Bench is a real-world multi-modal retrieval benchmark designed to evaluate retrieval models in reliable, challenging, and realistic settings. The dataset includes multi-modal documents such as text, tables, and figures to test the models ability to retrieve relevant information based on natural language queries.