FUNSD

The 'FUNSD' dataset is a benchmark used to evaluate Document Visual Question Answering (DocVQA) models by providing annotated documents that require understanding of textual semantics, spatial layout, and visual features.

Papers using FUNSD (5)

ROAP: A Reading-order And Attention-prior Pipeline For Optimizing Layout Transformers In Key Information Extraction2026OTCR: Optimal Transmission, Compression And Representation For Multimodal Information Extraction2025MGA-VQA: Secure And Interpretable Graph-augmented Visual Question Answering With Memory-guided Protection Against Unauthorized Knowledge Use2025LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding2020 · 59 citesDocFormerv2: Local Features for Document Understanding2023 · 3 cites