# Attend, Copy, Parse -- End-to-end information extraction from documents

arXiv: https://arxiv.org/abs/1812.07248 · HTML: https://mlsanity.com/paper/1812.07248/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 Dec 2018.

## Built on by (most cited first, 25 of 25)

- [StrucTexT: Structured Text Understanding with Multi-Modal Transformers](https://mlsanity.com/paper/2108.02923/) (2021)
- [Kleister: Key Information Extraction Datasets Involving Long Documents with Complex Layouts](https://mlsanity.com/paper/2105.05796/) (2021)
- [DocILE Benchmark for Document Information Localization and Extraction](https://mlsanity.com/paper/2302.05658/) (2023)
- [ViBERTgrid: A Jointly Trained Multi-Modal 2D Document Representation for Key Information Extraction from Documents](https://mlsanity.com/paper/2105.11672/) (2021)
- [Deep Learning based Visually Rich Document Content Understanding: A Survey](https://mlsanity.com/paper/2408.01287/) (2024)
- [Visual FUDGE: Form Understanding via Dynamic Graph Editing](https://mlsanity.com/paper/2105.08194/) (2021)
- [Business Document Information Extraction: Towards Practical Benchmarks](https://mlsanity.com/paper/2206.11229/) (2022)
- [End-to-End Hierarchical Relation Extraction for Generic Form Understanding](https://mlsanity.com/paper/2106.00980/) (2021)
- [Doc2Dict: Information Extraction as Text Generation](https://mlsanity.com/paper/2105.07510/) (2021)
- [TRIE++: Towards End-to-End Information Extraction from Visually Rich Documents](https://mlsanity.com/paper/2207.06744/) (2022)
- [Deep Learning based Key Information Extraction from Business Documents: Systematic Literature Review](https://mlsanity.com/paper/2408.06345/) (2024)
- [Value Retrieval with Arbitrary Queries for Form-like Documents](https://mlsanity.com/paper/2112.07820/) (2021)
- [Information Extraction from Scanned Invoice Images using Text Analysis and Layout Features](https://mlsanity.com/paper/2208.04011/) (2022)
- [SmartFlow: Robotic Process Automation using LLMs](https://mlsanity.com/paper/2405.12842/) (2024)
- [Leveraging Large Language Models for Semantic Query Processing in a Scholarly Knowledge Graph](https://mlsanity.com/paper/2405.15374/) (2024)
- [End-to-End Information Extraction by Character-Level Embedding and Multi-Stage Attentional U-Net](https://mlsanity.com/paper/2106.00952/) (2021)
- [Field Extraction from Forms with Unlabeled Data](https://mlsanity.com/paper/2110.04282/) (2021)
- [One-shot Key Information Extraction from Document with Deep Partial Graph Matching](https://mlsanity.com/paper/2109.13967/) (2021)
- [Knowing Where and What: Unified Word Block Pretraining for Document Understanding](https://mlsanity.com/paper/2207.13979/) (2022)
- [End-to-End Approach for Recognition of Historical Digit Strings](https://mlsanity.com/paper/2104.13666/) (2021)
- [Robustness Evaluation of Transformer-based Form Field Extractors via Form Attacks](https://mlsanity.com/paper/2110.04413/) (2021)
- [DocILE 2023 Teaser: Document Information Localization and Extraction](https://mlsanity.com/paper/2301.12394/) (2023)
- [End-to-End Document Classification and Key Information Extraction using Assignment Optimization](https://mlsanity.com/paper/2306.00750/) (2023)
- [AutoFormBench: Benchmark Dataset for Automating Form Understanding](https://mlsanity.com/paper/2603.29832/) (2026)
- [Pharos-ESG: A Framework for Multimodal Parsing, Contextual Narration, and Hierarchical Labeling of ESG Report](https://mlsanity.com/paper/2511.16417/) (2025)

## Abstract (arXiv)

Document information extraction tasks performed by humans create data consisting of a PDF or document image input, and extracted string outputs. This end-to-end data is naturally consumed and produced when performing the task because it is valuable in and of itself. It is naturally available, at no additional cost. Unfortunately, state-of-the-art word classification methods for information extraction cannot use this data, instead requiring word-level labels which are expensive to create and consequently not available for many real life tasks. In this paper we propose the Attend, Copy, Parse architecture, a deep neural network model that can be trained directly on end-to-end data, bypassing the need for word-level labels. We evaluate the proposed architecture on a large diverse set of invoices, and outperform a state-of-the-art production system based on word classification. We believe our proposed architecture can be used on many real life information extraction tasks where word classification cannot be used due to a lack of the required word-level labels.