Research map / Vision-language and multimodal / Human and embodied modeling
Learning to Extract Semantic Structure From Documents Using Multimodal Fully Convolutional Neural Networks
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 a text embedding map using OCR-extracted text (ablation) | Model5 · DSSE-200 | mean IoU: 73.0 -> 73.3 | Helped |
| Add a text embedding map using OCR-extracted text (ablation) | Model5 · Synthetic dataset | mean IoU: 79.6 -> 82.2 | Helped |
| Use real rather than OCR-extracted text in the embedding map (ablation) | Model5 · Synthetic dataset | mean IoU: 82.2 -> 86.0 | Helped |
| Add skip connections (ablation) | Bilinear-upsampled encoder-decoder · DSSE-200 | mean IoU: 61.4 -> 65.4 | Helped |
| Replace bilinear upsampling with unpooling (ablation) | Model 2 · DSSE-200 | mean IoU: 65.4% vs. 71.2% | Helped |
| Use dilation d=8 instead of d=1 (ablation) | Model 3 · DSSE-200 | mean IoU: 71.2 -> 69.5 | Hurt |
| Use a multirate dilated convolution block (ablation) | Model 3 · DSSE-200 | mean IoU: 71.2 -> 73.0 | Helped |
| Add reconstruction loss (ablation) | Model5 with classification objective · DSSE-200 | mean IoU: 73.3 -> 73.9 | Helped |
| Add consistency loss (ablation) | Model5 with classification objective · DSSE-200 | mean IoU: 73.3 -> 75.4 | Helped |
| Add both reconstruction and consistency losses (ablation) | Model5 with classification objective · DSSE-200 | mean IoU: 73.3 -> 75.9 | Helped |
Closest papers by meaning
- Relation-Rich Visual Document Generator for Visual Information Extraction (CVPR 2025 · similarity 0.85)
- Towards Real-World Document Parsing via Realistic Scene Synthesis and Document-Aware Training (CVPR 2026 · similarity 0.85)
- Image-Text Co-Decomposition for Text-Supervised Semantic Segmentation (CVPR 2024 · similarity 0.84)
- DocLayLLM: An Efficient Multi-modal Extension of Large Language Models for Text-rich Document Understanding (CVPR 2025 · similarity 0.83)
- DocSAM: Unified Document Image Segmentation via Query Decomposition and Heterogeneous Mixed Learning (CVPR 2025 · similarity 0.83)
- DocMamba: Efficient Document Pre-training with State Space Model (AAAI 2025 · similarity 0.83)
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.