Research map / Architectures and efficiency / RNNs and normalization
RT-DocLayout: Real-Time End-to-End Document Layout Analysis with Reading Order in the Wild
Cheng Cui, Tingquan Gao, Xueqing Wang, Changda Zhou, Hongen Liu, Ting Sun, Yubo Zhang, Zelun Zhang, Jiaxuan Liu, Manhui Lin, Yue Zhang, Suyin Liang and 2 more
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add mask prediction to bbox supervision (ablation) | RT-DocLayout (XYCut order) · OmniDocBench v1.5 | Overall: 93.81 -> 94.33 | Improved |
| replace XYCut with coupled reading order learning (ablation) | RT-DocLayout bbox+mask · OmniDocBench v1.5 | Reading Order distance: 0.074 -> 0.041 | Improved |
| replace XYCut with coupled reading order learning (ablation) | RT-DocLayout bbox+mask · Real5-OmniDocBench (Warp) | Reading Order distance: 0.721 -> 0.092 | Improved |
| use decoupled separately trained reading order instead of XYCut (ablation) | RT-DocLayout bbox+mask · OmniDocBench v1.5 | Reading Order distance: 0.074 -> 0.189 | Hurt |
| make reading order jointly optimized (coupled) rather than decoupled (ablation) | RT-DocLayout bbox+mask · OmniDocBench v1.5 | Reading Order distance: 0.189 -> 0.041 | Improved |
| couple reading order with detection instead of decoupled (ablation) | RT-DocLayout bbox+mask · OmniDocBench v1.5 | Overall: 93.87 -> 94.39 | Improved |
| add physical spatial aware data augmentation (ablation) | RT-DocLayout · Real5-OmniDocBench (Warp) | Overall: 86.68 -> 91.25 | Improved |
| add physical spatial aware data augmentation (ablation) | RT-DocLayout · Real5-OmniDocBench (Skew) | Overall: 77.55 -> 91.66 | Improved |
| add physical spatial aware data augmentation (ablation) | RT-DocLayout · Real5-OmniDocBench (Skew) | Reading Order distance: 0.146 -> 0.061 | Improved |
| add physical spatial aware data augmentation (ablation) | RT-DocLayout · Real5-OmniDocBench (Warp) | Reading Order distance: 0.092 -> 0.063 | Improved |
| add physical spatial aware data augmentation (ablation) | RT-DocLayout · OmniDocBench v1.5 | Overall: 94.39 -> 94.50 | No clear effect |
| add mask prediction to bbox supervision (ablation) | RT-DocLayout (XYCut order) · Real5-OmniDocBench (Skew) | Overall: 77.18 -> 75.63 | Hurt |
Closest papers by meaning
- Towards Real-World Document Parsing via Realistic Scene Synthesis and Document-Aware Training (CVPR 2026 · similarity 0.88)
- Graph-based Document Structure Analysis (ICLR 2025 · similarity 0.87)
- A Simple yet Effective Layout Token in Large Language Models for Document Understanding (CVPR 2025 · similarity 0.87)
- M2Doc: A Multi-Modal Fusion Approach for Document Layout Analysis (AAAI 2024 · similarity 0.87)
- Read, Parse, Describe: Unified Document Parsing with Visual Element Description Generation (NeurIPS 2026 · similarity 0.87)
- LayoutLLM: Layout Instruction Tuning with Large Language Models for Document Understanding (CVPR 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
