zhaosuifeng/FinRAGBench-V

FinRAGBench-V: A Benchmark for Multimodal RAG with Visual Citation in the Financial Domain 🤗 Code 📄 Paper Overview FinRAGBench-V is a comprehensive benchmark for visual retrieval-augmented generation (RAG) in finance, addressing the challenge that most existing financial RAG research focuses predominantly on text while overlooking rich visual content in financial documents. By integrating multimodal data and providing visual citation, FinRAGBench-V ensures… See the full description on the dataset page: https://huggingface.co/datasets/zhaosuifeng/FinRAGBench-V.