Research map / Image generation / Autoregressive and latent generation
HybriDLA: Hybrid Generation for Document Layout Analysis
Yufan Chen, Omar Moured, Ruiping Liu, Junwei Zheng, Kunyu Peng, Jiaming Zhang, Rainer Stiefelhagen
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 autoregressive query expansion (AQE) (ablation) | ResNet-50 DINO DETR · DocLayNet | mAP: 75.3 -> 76.8 | Improved |
| add AQE to vanilla DETR (ablation) | ResNet-50 DETR · DocLayNet | mAP: 74.2 -> 74.4 | No clear effect |
| add AQE to deformable DETR (ablation) | ResNet-50 · DocLayNet | mAP: 75.1 -> 76.3 | Improved |
| add feature fusion encoder (FFE) instead of deformable encoder, with DR (ablation) | HybriDLA Swin-L · DocLayNet | mAP: 78.1 -> 80.4 | Improved |
| add FFE instead of deformable encoder (ablation) | HybriDLA InternImage · DocLayNet | mAP: 82.4 -> 83.5 | Improved |
| add FFE instead of deformable encoder (ablation) | HybriDLA ResNet-50 · DocLayNet | mAP: 76.2 -> 74.4 | Hurt |
| add diffusion-based refinement (DR) (ablation) | HybriDLA Swin-L · DocLayNet | mAP: 79.1 -> 80.4 | Improved |
| add DR (ablation) | HybriDLA ResNet-50 · DocLayNet | mAP: 74.4 -> 74.4 | No clear effect |
| add DR (ablation) | HybriDLA InternImage · DocLayNet | mAP: 81.3 -> 83.5 | Improved |
| add DR with one decoder layer (ablation) | HybriDLA ResNet-50 · DocLayNet | mAP: 74.20 -> 72.80 | Hurt |
| increase decoder layers from 1 to 6 (no DR) (ablation) | HybriDLA Swin-L · DocLayNet | mAP: 76.80 -> 79.10 | Improved |
| reduce AQE queries from 300 to 30 (ablation) | HybriDLA InternImage · DocLayNet | mAP: 83.5 -> 80.7 | Hurt |
Closest papers by meaning
- M2Doc: A Multi-Modal Fusion Approach for Document Layout Analysis (AAAI 2024 · similarity 0.89)
- A Simple yet Effective Layout Token in Large Language Models for Document Understanding (CVPR 2025 · similarity 0.86)
- RoDLA: Benchmarking the Robustness of Document Layout Analysis Models (CVPR 2024 · similarity 0.86)
- LayoutLLM: Layout Instruction Tuning with Large Language Models for Document Understanding (CVPR 2024 · similarity 0.85)
- Graph-based Document Structure Analysis (ICLR 2025 · similarity 0.85)
- Relation-Rich Visual Document Generator for Visual Information Extraction (CVPR 2025 · similarity 0.85)
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.
