{"url":"/dataset/funsd","name":"FUNSD","full_name":"Form Understanding in Noisy Scanned Documents","description_markdown":"Form Understanding in Noisy Scanned Documents (FUNSD) comprises 199 real, fully annotated, scanned forms. The documents are noisy and vary widely in appearance, making form understanding (FoUn) a challenging task. The proposed dataset can be used for various tasks, including text detection, optical character recognition, spatial layout analysis, and entity labeling/linking.\r\n\r\nSource: [FUNSD: A Dataset for Form Understanding in Noisy Scanned Documents](/paper/190513538)\r\n\r\nImage source: [https://guillaumejaume.github.io/FUNSD/](https://guillaumejaume.github.io/FUNSD/)","description_withheld":null,"homepage":"https://guillaumejaume.github.io/FUNSD/","introduced_date":null,"introduced_date_note":null,"introduced_by":{"paper":"/paper/190513538","title":"FUNSD: A Dataset for Form Understanding in Noisy Scanned Documents","first_author":"Guillaume Jaume","url":null},"license":{"name":"Custom","url":"https://guillaumejaume.github.io/FUNSD/work/"},"modalities":[{"name":"Images","url":"/datasets/modality/images"},{"name":"Texts","url":"/datasets/modality/texts"}],"tasks":[{"name":"Named Entity Recognition (NER)","url":"/task/named-entity-recognition-ner","datasets_with_task":"/datasets/task/named-entity-recognition-ner"},{"name":"Relation Extraction","url":"/task/relation-extraction","datasets_with_task":"/datasets/task/relation-extraction"},{"name":"Optical Character Recognition (OCR)","url":"/task/optical-character-recognition","datasets_with_task":"/datasets/task/optical-character-recognition"},{"name":"Entity Linking","url":"/task/entity-linking","datasets_with_task":"/datasets/task/entity-linking"},{"name":"Table Detection","url":"/task/table-detection","datasets_with_task":"/datasets/task/table-detection"},{"name":"Semantic entity labeling","url":"/task/semantic-entity-labeling","datasets_with_task":"/datasets/task/semantic-entity-labeling"}],"languages":[],"variants":["FUNSD"],"data_loaders":[{"repo":"https://github.com/huggingface/datasets","url":"https://huggingface.co/datasets/nielsr/FUNSD_layoutlmv2","frameworks":["tf","pytorch","jax"]},{"repo":"https://github.com/mindee/doctr","url":"https://mindee.github.io/doctr/latest/datasets.html#doctr.datasets.FUNSD","frameworks":["tf","pytorch"]}],"num_papers_in_archive":179,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/semantic-entity-labeling-on-funsd","task":"Semantic entity labeling","dataset_variant":"FUNSD","rows":15,"metrics":["F1"],"first_row_in_archive_order":{"model":"LayoutMask (large)","paper":"/paper/layoutmask-enhance-text-layout-interaction-in","metrics":{"F1":"93.20"},"code_links":[]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/relation-extraction-on-funsd","task":"Relation Extraction","dataset_variant":"FUNSD","rows":9,"metrics":["F1"],"first_row_in_archive_order":{"model":"LayoutLMv3 large EM + BBO + RSF","paper":"/paper/a-layoutlmv3-based-model-for-enhanced","metrics":{"F1":"90.81"},"code_links":[]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/entity-linking-on-funsd","task":"Entity Linking","dataset_variant":"FUNSD","rows":7,"metrics":["F1"],"first_row_in_archive_order":{"model":"GeoLayoutLM","paper":"/paper/geolayoutlm-geometric-pre-training-for-visual","metrics":{"F1":"89.45"},"code_links":[{"title":"alibabaresearch/advancedliteratemachinery","url":"https://github.com/alibabaresearch/advancedliteratemachinery"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/modeling-layout-reading-order-as-ordering","title":"Modeling Layout Reading Order as Ordering Relations for Visually-rich Document Understanding","date":"2024-09-29","rows_on_this_dataset":3,"code_links":1,"syntology":null},{"paper":"/paper/a-layoutlmv3-based-model-for-enhanced","title":"A LayoutLMv3-Based Model for Enhanced Relation Extraction in Visually-Rich Documents","date":"2024-04-16","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/reading-order-matters-information-extraction","title":"Reading Order Matters: Information Extraction from Visually-rich Documents by Token Path Prediction","date":"2023-10-17","rows_on_this_dataset":3,"code_links":2,"syntology":null},{"paper":"/paper/doctr-document-transformer-for-structured","title":"DocTr: Document Transformer for Structured Information Extraction in Documents","date":"2023-07-16","rows_on_this_dataset":2,"code_links":0,"syntology":null},{"paper":"/paper/layoutmask-enhance-text-layout-interaction-in","title":"LayoutMask: Enhance Text-Layout Interaction in Multi-modal Pre-training for Document Understanding","date":"2023-05-30","rows_on_this_dataset":2,"code_links":0,"syntology":null},{"paper":"/paper/geolayoutlm-geometric-pre-training-for-visual","title":"GeoLayoutLM: Geometric Pre-training for Visual Information Extraction","date":"2023-04-21","rows_on_this_dataset":4,"code_links":1,"syntology":null},{"paper":"/paper/structextv2-masked-visual-textual-prediction","title":"StrucTexTv2: Masked Visual-Textual Prediction for Document Image Pre-training","date":"2023-03-01","rows_on_this_dataset":2,"code_links":1,"syntology":null},{"paper":"/paper/dgcn-based-solution-for-entity-linking-on","title":"DGCN Based Solution for Entity Linking on Visual Rich Document","date":"2022-11-16","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/ernie-layout-layout-knowledge-enhanced-pre","title":"ERNIE-Layout: Layout Knowledge Enhanced Pre-training for Visually-rich Document Understanding","date":"2022-10-12","rows_on_this_dataset":1,"code_links":2,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":7,"samples_ran":2,"samples_unverified":5,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/xdoc-unified-pre-training-for-cross-format","title":"XDoc: Unified Pre-training for Cross-Format Document Understanding","date":"2022-10-06","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/doc2graph-a-task-agnostic-document","title":"Doc2Graph: a Task Agnostic Document Understanding Framework based on Graph Neural Networks","date":"2022-08-23","rows_on_this_dataset":2,"code_links":1,"syntology":null},{"paper":"/paper/layoutlmv3-pre-training-for-document-ai-with","title":"LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking","date":"2022-04-18","rows_on_this_dataset":2,"code_links":4,"syntology":null},{"paper":"/paper/lilt-a-simple-yet-effective-language","title":"LiLT: A Simple yet Effective Language-Independent Layout Transformer for Structured Document Understanding","date":"2022-02-28","rows_on_this_dataset":1,"code_links":5,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":3,"samples_ran":2,"samples_unverified":1,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/entity-relation-extraction-as-dependency","title":"Entity Relation Extraction as Dependency Parsing in Visually Rich Documents","date":"2021-10-19","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/bros-a-layout-aware-pre-trained-language","title":"BROS: A Pre-trained Language Model Focusing on Text and Layout for Better Key Information Extraction from Documents","date":"2021-08-10","rows_on_this_dataset":1,"code_links":2,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":13,"samples_ran":2,"samples_unverified":11,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/layoutlmv2-multi-modal-pre-training-for","title":"LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding","date":"2020-12-29","rows_on_this_dataset":3,"code_links":9,"syntology":null},{"paper":"/paper/layoutlm-pre-training-of-text-and-layout-for","title":"LayoutLM: Pre-training of Text and Layout for Document Image Understanding","date":"2019-12-31","rows_on_this_dataset":1,"code_links":19,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":3,"samples_ran":2,"samples_unverified":1,"pointer_only_for_licence":1,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":4,"samples_harvested":26,"samples_ran":8,"samples_unverified":18,"pointer_only_for_licence":1,"papers_with_no_sample_that_ran":0,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}