Large Language Models in Document Intelligence: A Comprehensive Survey, Recent Advances, Challenges, and Future Trends

2025/09/16 by Wenjun Ke, Yifan Zheng, Youlan Li +5
Computer Science · #Handwritten Text Recognition Techniques #Natural Language Processing Techniques #Topic Modeling

paper · doi:10.1145/3768156

openalex publication_date 2025/09/16 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/17

Abstract

The rapid proliferation of documents has made document intelligence increasingly critical across various industries. In recent years, Large Language Models (LLMs) have dramatically transformed the field of document intelligence, allowing for more advanced and accurate document processing solutions. Despite these advancements, most existing surveys have failed to focus on these breakthroughs, instead concentrating on traditional methods and earlier machine learning techniques. This survey seeks to fill that gap by offering an in-depth analysis of approximately 300 papers published between 2021 and mid-2025, thus providing a comprehensive overview of the impact of LLMs in document intelligence. The key topics explored include Retrieval-Augmented Generation (RAG), long-context processing, and fine-tuning LLMs for document comprehension. Furthermore, the survey highlights essential datasets, practical applications, current challenges, and future research directions, offering critical insights for both researchers and industry practitioners looking to advance the field.

Citations

• TabTransformer: Tabular Data Modeling Using Contextual Embeddings
• Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
• FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness
• Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
• TRUST: An Accurate and End-to-End Table structure Recognizer Using Splitting-based Transformers
• Scaling Instruction-Finetuned Language Models
• Think-in-Memory: Recalling and Post-thinking Enable LLMs with Long-Term Memory
• HeGTa: Leveraging Heterogeneous Graph-enhanced Large Language Models for Few-shot Complex Table Understanding
• Focus Anywhere for Fine-grained Multi-page Document Understanding
• Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling

Cited by

• Gold Panning: Turning Positional Bias into Signal for Multi-Document LLM Reasoning

Related

• Document Intelligence in the Era of Large Language Models: A Survey
• A Survey on RAG Meeting LLMs: Towards Retrieval-Augmented Large Language Models
• Scaling Beyond Context: A Survey of Multimodal Retrieval-Augmented Generation for Document Understanding
• A Survey on Retrieval-Augmented Text Generation for Large Language Models
• Report on the 1st Workshop on Large Language Model for Evaluation in Information Retrieval (LLM4Eval 2024) at SIGIR 2024