# Advanced Layout Analysis Models for Docling

arXiv: https://arxiv.org/abs/2509.11720 · HTML: https://mlsanity.com/paper/2509.11720/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 16 references that are in the map.
- First posted on arXiv 15 Sep 2025.

## Builds on (most cited first, 11 of 11)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [End-to-End Object Detection with Transformers](https://mlsanity.com/paper/2005.12872/) (2020)
- [You Only Look Once: Unified, Real-Time Object Detection](https://mlsanity.com/paper/1506.02640/) (2015)
- [TableFormer: Table Structure Understanding with Transformers](https://mlsanity.com/paper/2203.01017/) (2022)
- [RT-DETRv2: Improved Baseline with Bag-of-Freebies for Real-Time Detection Transformer](https://mlsanity.com/paper/2407.17140/) (2024)
- [MonkeyOCR: Document Parsing with a Structure-Recognition-Relation Triplet Paradigm](https://mlsanity.com/paper/2506.05218/) (2025)
- [Dolphin: Document Image Parsing via Heterogeneous Anchor Prompting](https://mlsanity.com/paper/2505.14059/) (2025)
- [D-FINE: Redefine Regression Task in DETRs as Fine-grained Distribution Refinement](https://mlsanity.com/paper/2410.13842/) (2024)
- [SmolDocling: An ultra-compact vision-language model for end-to-end multi-modal document conversion](https://mlsanity.com/paper/2503.11576/) (2025)
- [Optimized Table Tokenization for Table Structure Recognition](https://mlsanity.com/paper/2305.03393/) (2023)
- [MolGrapher: Graph-based Visual Recognition of Chemical Structures](https://mlsanity.com/paper/2308.12234/) (2023)

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

- [Argumentation for Explainable and Globally Contestable Decision Support with LLMs](https://mlsanity.com/paper/2603.14643/) (2026)
- [The Character Error Vector: Decomposable errors for page-level OCR evaluation](https://mlsanity.com/paper/2604.06160/) (2026)
- [The COTe score: A decomposable framework for evaluating Document Layout Analysis models](https://mlsanity.com/paper/2603.12718/) (2026)
- [AutoForest: Automatically Generating Forest Plots from Biomedical Studies with End-to-End Evidence Extraction and Synthesis](https://mlsanity.com/paper/2606.02403/) (2026)
- [Improving Layout Representation Learning Across Inconsistently Annotated Datasets via Agentic Harmonization](https://mlsanity.com/paper/2604.11042/) (2026)
- [Structured Layout Priors for Robust Out-of-Distribution Visual Document Understanding](https://mlsanity.com/paper/2605.19866/) (2026)

## Abstract (arXiv)

This technical report documents the development of novel Layout Analysis models integrated into the Docling document-conversion pipeline. We trained several state-of-the-art object detectors based on the RT-DETR, RT-DETRv2 and DFINE architectures on a heterogeneous corpus of 150,000 documents (both openly available and proprietary). Post-processing steps were applied to the raw detections to make them more applicable to the document conversion task. We evaluated the effectiveness of the layout analysis on various document benchmarks using different methodologies while also measuring the runtime performance across different environments (CPU, Nvidia and Apple GPUs). We introduce five new document layout models achieving 20.6% - 23.9% mAP improvement over Docling's previous baseline, with comparable or better runtime. Our best model, "heron-101", attains 78% mAP with 28 ms/image inference time on a single NVIDIA A100 GPU. Extensive quantitative and qualitative experiments establish best practices for training, evaluating, and deploying document-layout detectors, providing actionable guidance for the document conversion community. All trained checkpoints, code, and documentation are released under a permissive license on HuggingFace.