The future of document indexing: GPT and Donut revolutionize table of content processing
Not on the map. Why?
Abstract (arXiv)
Industrial projects rely heavily on lengthy, complex specification documents, making tedious manual extraction of structured information a major bottleneck. This paper introduces an innovative approach to automate this process, leveraging the capabilities of two cutting-edge AI models: Donut, a model that extracts information directly from scanned documents without OCR, and OpenAI GPT-3.5 Turbo, a robust large language model. The proposed methodology is initiated by acquiring the table of contents (ToCs) from construction specification documents and subsequently structuring the ToCs text into JSON data. Remarkable accuracy is achieved, with Donut reaching 85% and GPT-3.5 Turbo reaching 89% in effectively organizing the ToCs. This landmark achievement represents a significant leap forward in document indexing, demonstrating the immense potential of AI to automate information extraction tasks across diverse document types, boosting efficiency and liberating critical resources in various industries.
Key facts. Cited by 1 paper in the map (data through 13 Oct 2026). Builds on 21 of its 32 references that are in the map. First posted on arXiv 12 Mar 2024.
Builds on
21 of 32 references.
References · 21
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,304
- 2023GPT-4 Technical Reportcited by 18,278
- 2019Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelismcited by 1,851
- 2023Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflectioncited by 1,815
- 2023A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questionscited by 1,594
- 2019LayoutLM: Pre-training of Text and Layout for Document Image Understandingcited by 551
- 2021OCR-free Document Understanding Transformercited by 402
- 2021ICDAR2019 Competition on Scanned Receipt OCR and Information Extractioncited by 254
- 2015Evaluation of Deep Convolutional Nets for Document Image Classification and Retrievalcited by 212
- 2023Large Language Models for Generative Information Extraction: A Surveycited by 184
- 2020PP-OCR: A Practical Ultra Lightweight OCR Systemcited by 148
- 2022PP-OCRv3: More Attempts for the Improvement of Ultra Lightweight OCR Systemcited by 106
9 more
- 2021BROS: A Pre-trained Language Model Focusing on Text and Layout for Better Key Information Extraction from Documentscited by 104
- 2021Kleister: Key Information Extraction Datasets Involving Long Documents with Complex Layoutscited by 82
- 2021ViBERTgrid: A Jointly Trained Multi-Modal 2D Document Representation for Key Information Extraction from Documentscited by 29
- 2023A Study on the Implementation of Generative AI Services Using an Enterprise Data-Based LLM Application Architecturecited by 28
- 2020Kleister: A novel task for Information Extraction involving Long Documents with Complex Layoutcited by 27
- 2023DocParser: End-to-end OCR-free Information Extraction from Visually Rich Documentscited by 19
- 2022PP-StructureV2: A Stronger Document Analysis Systemcited by 18
- 2023Improving Information Extraction on Business Documents with Specific Pre-Training Taskscited by 3
- 2022DeeperDive: The Unreasonable Effectiveness of Weak Supervision in Document Understanding A Case Study in Collaboration with UiPath Inccited by 2
Built on by
Citing papers per year: 2025: 1.
1 paper.