mlsanityData to 7 Oct 2026

Chargrid: Towards Understanding 2D Documents

arXiv 1809.08799 ↗First posted 24 Sep 2018cs.CL cs.CV cs.LGcited by 100

Not on the map. Why?

Abstract (arXiv)

We introduce a novel type of text representation that preserves the 2D layout of a document. This is achieved by encoding each document page as a two-dimensional grid of characters. Based on this representation, we present a generic document understanding pipeline for structured documents. This pipeline makes use of a fully convolutional encoder-decoder network that predicts a segmentation mask and bounding boxes. We demonstrate its capabilities on an information extraction task from invoices and show that it significantly outperforms approaches based on sequential text or document images.

Key facts. Cited by 100 papers in the map (data through 7 Oct 2026). First posted on arXiv 24 Sep 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 32; 2022: 17; 2023: 20; 2024: 13; 2025: 8; 2026: 10.

100 papers.

Sort:

By area:

  1. 2021DocFormer: End-to-End Transformer for Document Understandingcited by 254
  2. 2021VisualMRC: Machine Reading Comprehension on Document Imagescited by 150
  3. 2021SelfDoc: Self-Supervised Document Representation Learningcited by 117
  4. 2022LiLT: A Simple yet Effective Language-Independent Layout Transformer for Structured Document Understandingcited by 107
  5. 2021StructuralLM: Structural Pre-training for Form Understandingcited by 91
  6. 2021StrucTexT: Structured Text Understanding with Multi-Modal Transformerscited by 90
  7. 2021Kleister: Key Information Extraction Datasets Involving Long Documents with Complex Layoutscited by 82
  8. 2022XYLayoutLM: Towards Layout-Aware Multimodal Networks For Visually-Rich Document Understandingcited by 68
  9. 2022ERNIE-Layout: Layout Knowledge Enhanced Pre-training for Visually-rich Document Understandingcited by 67
  10. 2021Towards Robust Visual Information Extraction in Real World: New Dataset and Novel Solutioncited by 64
  11. 2022FormNet: Structural Encoding beyond Sequential Modeling in Form Document Information Extractioncited by 64
  12. 2022DAN: a Segmentation-free Document Attention Network for Handwritten Document Recognitioncited by 62
38 more
  1. 2021Document AI: Benchmarks, Models and Applicationscited by 61
  2. 2023DocILE Benchmark for Document Information Localization and Extractioncited by 58
  3. 2023Visual Information Extraction in the Wild: Practical Dataset and End-to-end Solutioncited by 54
  4. 2023Vision Grid Transformer for Document Layout Analysiscited by 45
  5. 2024Document Parsing Unveiled: Techniques, Challenges, and Prospects for Structured Information Extractioncited by 44
  6. 2021Spatial Dual-Modality Graph Reasoning for Key Information Extractioncited by 43
  7. 2021MMOCR: A Comprehensive Toolbox for Text Detection, Recognition and Understandingcited by 40
  8. 2024HRVDA: High-Resolution Visual Document Assistantcited by 37
  9. 2021ViBERTgrid: A Jointly Trained Multi-Modal 2D Document Representation for Key Information Extraction from Documentscited by 29
  10. 2023Layout and Task Aware Instruction Prompt for Zero-shot Document Image Question Answeringcited by 24
  11. 2021ROPE: Reading Order Equivariant Positional Encoding for Graph-based Document Information Extractioncited by 22
  12. 2026AgentOCR: Reimagining Agent History via Optical Self-Compressioncited by 22
  13. 2024Deep Learning based Visually Rich Document Content Understanding: A Surveycited by 21
  14. 2023FormNetV2: Multimodal Graph Contrastive Learning for Form Document Information Extractioncited by 20
  15. 2023DocParser: End-to-end OCR-free Information Extraction from Visually Rich Documentscited by 19
  16. 2021MatchVIE: Exploiting Match Relevancy between Entities for Visual Information Extractioncited by 18
  17. 2022Bi-VLDoc: Bidirectional Vision-Language Modeling for Visually-Rich Document Understandingcited by 18
  18. 2023DocTr: Document Transformer for Structured Information Extraction in Documentscited by 18
  19. 2023Attention Where It Matters: Rethinking Visual Document Understanding with Selective Region Concentrationcited by 18
  20. 2023LayoutMask: Enhance Text-Layout Interaction in Multi-modal Pre-training for Document Understandingcited by 17
  21. 2021Form2Seq : A Framework for Higher-Order Form Structure Extractioncited by 16
  22. 2021LAMPRET: Layout-Aware Multimodal PreTraining for Document Understandingcited by 15
  23. 2021Cost-effective End-to-end Information Extraction for Semi-structured Document Imagescited by 14
  24. 2026Seeing is Coding: On the Effectiveness of Vision Language Models in Code Understandingcited by 13
  25. 2024Visually Guided Generative Text-Layout Pre-training for Document Intelligencecited by 12
  26. 2022Business Document Information Extraction: Towards Practical Benchmarkscited by 11
  27. 2022ERNIE-mmLayout: Multi-grained MultiModal Transformer for Document Understandingcited by 9
  28. 2025Information Extraction from Visually Rich Documents using LLM-based Organization of Documents into Independent Textual Segmentscited by 9
  29. 2021Tag, Copy or Predict: A Unified Weakly-Supervised Learning Framework for Visual Information Extraction using Sequencescited by 8
  30. 2023GenKIE: Robust Generative Multimodal Document Key Information Extractioncited by 8
  31. 2025A Survey on MLLM-based Visually Rich Document Understanding: Methods, Challenges, and Emerging Trendscited by 8
  32. 2021End-to-End Hierarchical Relation Extraction for Generic Form Understandingcited by 7
  33. 2021Key Information Extraction From Documents: Evaluation And Generatorcited by 7
  34. 2021Skim-Attention: Learning to Focus via Document Layoutcited by 7
  35. 2022DocBed: A Multi-Stage OCR Solution for Documents with Complex Layoutscited by 7
  36. 2022QueryForm: A Simple Zero-shot Form Entity Query Frameworkcited by 7
  37. 2021Multi-Modal Association based Grouping for Form Structure Extractioncited by 6
  38. 2022GMN: Generative Multi-modal Network for Practical Document Information Extractioncited by 6

50 most cited of 100 All 100 (JSON)