# Problem Solved? Information Extraction Design Space for Layout-Rich Documents using LLMs

arXiv: https://arxiv.org/abs/2502.18179 · HTML: https://mlsanity.com/paper/2502.18179/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 37 of its 65 references that are in the map.
- First posted on arXiv 25 Feb 2025.

## Builds on (most cited first, 30 of 37)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [Least-to-Most Prompting Enables Complex Reasoning in Large Language Models](https://mlsanity.com/paper/2205.10625/) (2022)
- [LayoutLM: Pre-training of Text and Layout for Document Image Understanding](https://mlsanity.com/paper/1912.13318/) (2019)
- [LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking](https://mlsanity.com/paper/2204.08387/) (2022)
- [LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding](https://mlsanity.com/paper/2012.14740/) (2020)
- [FUNSD: A Dataset for Form Understanding in Noisy Scanned Documents](https://mlsanity.com/paper/1905.13538/) (2019)
- [Scene Text Visual Question Answering](https://mlsanity.com/paper/1905.13648/) (2019)
- [DocFormer: End-to-End Transformer for Document Understanding](https://mlsanity.com/paper/2106.11539/) (2021)
- [Unifying Vision, Text, and Layout for Universal Document Processing](https://mlsanity.com/paper/2212.02623/) (2022)
- [Document Understanding Dataset and Evaluation (DUDE)](https://mlsanity.com/paper/2305.08455/) (2023)
- [Going Full-TILT Boogie on Document Understanding with Text-Image-Layout Transformer](https://mlsanity.com/paper/2102.09550/) (2021)
- [DocLLM: A layout-aware generative language model for multimodal document understanding](https://mlsanity.com/paper/2401.00908/) (2023)
- [LayoutLLM: Layout Instruction Tuning with Large Language Models for Document Understanding](https://mlsanity.com/paper/2404.05225/) (2024)
- [Graph Convolution for Multimodal Information Extraction from Visually Rich Documents](https://mlsanity.com/paper/1903.11279/) (2019)
- [Kleister: Key Information Extraction Datasets Involving Long Documents with Complex Layouts](https://mlsanity.com/paper/2105.05796/) (2021)
- [ERNIE-Layout: Layout Knowledge Enhanced Pre-training for Visually-rich Document Understanding](https://mlsanity.com/paper/2210.06155/) (2022)
- [FormNet: Structural Encoding beyond Sequential Modeling in Form Document Information Extraction](https://mlsanity.com/paper/2203.08411/) (2022)
- [Document AI: Benchmarks, Models and Applications](https://mlsanity.com/paper/2111.08609/) (2021)
- [LMDX: Language Model-based Document Information Extraction and Localization](https://mlsanity.com/paper/2309.10952/) (2023)
- [VRDU: A Benchmark for Visually-rich Document Understanding](https://mlsanity.com/paper/2211.15421/) (2022)
- [Schema-Driven Information Extraction from Heterogeneous Tables](https://mlsanity.com/paper/2305.14336/) (2023)
- [A Zero-shot and Few-shot Study of Instruction-Finetuned Large Language Models Applied to Clinical and Biomedical Tasks](https://mlsanity.com/paper/2307.12114/) (2023)
- [Multimodal Pre-training Based on Graph Attention Network for Document Understanding](https://mlsanity.com/paper/2203.13530/) (2022)
- [LayoutLLM: Large Language Model Instruction Tuning for Visually Rich Document Understanding](https://mlsanity.com/paper/2403.14252/) (2024)
- [Reading Order Matters: Information Extraction from Visually-rich Documents by Token Path Prediction](https://mlsanity.com/paper/2310.11016/) (2023)

## Built on by (most cited first, 6 of 6)

- [Watt Counts: Energy-Aware Benchmark for Sustainable LLM Inference on Heterogeneous GPU Architectures](https://mlsanity.com/paper/2604.09048/) (2026)
- [Clean & Clear: Feasibility of Safe LLM Clinical Guidance](https://mlsanity.com/paper/2503.20953/) (2025)
- [AgenticIE: An Adaptive Agent for Information Extraction from Complex Regulatory Documents](https://mlsanity.com/paper/2509.11773/) (2025)
- [TextMineX: Data, Evaluation Framework and Ontology-guided LLM Pipeline for Humanitarian Mine Action](https://mlsanity.com/paper/2509.15098/) (2025)
- [LLM-Based Examination of Eligibility Criteria from Securities Prospectuses at the German Central Bank](https://mlsanity.com/paper/2606.27316/) (2026)
- [The Right Information Extraction Pipeline Depends on the Document: Accuracy-Energy Trade-offs for Small, Local Models](https://mlsanity.com/paper/2609.31341/) (2026)

## Abstract (arXiv)

This paper defines and explores the design space for information extraction (IE) from layout-rich documents using large language models (LLMs). The three core challenges of layout-aware IE with LLMs are 1) data structuring, 2) model engagement, and 3) output refinement. Our study investigates the sub-problems and methods within these core challenges, such as input representation, chunking, prompting, selection of LLMs, and multimodal models. It examines the effect of different design choices through LayIE-LLM, a new, open-source, layout-aware IE test suite, benchmarking against traditional, fine-tuned IE models. The results on two IE datasets show that LLMs require adjustment of the IE pipeline to achieve competitive performance: the optimized configuration found with LayIE-LLM achieves 13.3--37.5 F1 points more than a general-practice baseline configuration using the same LLM. To find a well-working configuration, we develop a one-factor-at-a-time (OFAT) method that achieves near-optimal results. Our method is only 0.8--1.8 points lower than the best full factorial exploration with a fraction (2.8%) of the required computation. Overall, we demonstrate that, if well-configured, general-purpose LLMs match the performance of specialized models, providing a cost-effective, finetuning-free alternative. Our test-suite is available at https://github.com/gayecolakoglu/LayIE-LLM.