Multimodal Document Representation is a document-level computational framework that integrates and encodes heterogeneous modalities—such as text, layout, image, and metadata—into a unified semantic representation to enable joint understanding and reasoning across diverse data types within a single document.
56
Publications
3.1K
Citations
258
Authors
72
Institutions
Publications per year
2017–2026
34
Parents
Publications
56
LayoutLMv2: Multi-modal Pre-training for Visually-rich Document Understanding
Yang Xu, Yiheng Xu, Tengchao Lv et al. · 2021 · 438 citations · Full text
Graph Convolution for Multimodal Information Extraction from Visually Rich Documents
Xiaojing Liu, Feiyu Gao, Qiong Zhang et al. · 2019 · 204 citations · Full text
1–5 of 56
Authors
258
- CZ
3
h-index
- GW
3
h-index
- YL
2
h-index
- TL
2
h-index
- SA
2
h-index
1–5 of 258
Institutions
72
University of Science and Technology of China
Hefei, China
3 h-index
4 publications
61 citations
University of California, Berkeley
Berkeley, United States
2 h-index
4 publications
166 citations
Waseda University
Tokyo, Japan
2 h-index
1 publications
18 citations
Alibaba Group (China)
Hangzhou, China
2 h-index
6 publications
284 citations
University of Delaware
Newark, United States
2 h-index
3 publications
170 citations
Harbin Institute of Technology
Harbin, China
2 h-index
5 publications
1.4K citations
Computer Vision Center
Cerdanyola del Vallès, Spain
2 h-index
4 publications
137 citations
Fudan University
Shanghai, China
2 h-index
4 publications
168 citations
1–8 of 72
Venues
- arXiv (Cornell University)
Repository · 7 publications · 227 citations
7
h-index
- Multimedia Tools and Applications
1 publications · 21 citations
1
h-index
- Computing in Science & Engineering
1 publications · 11 citations
1
h-index
- Discourse Studies
1 publications · 67 citations
1
h-index
- IEEE Transactions on Multimedia
1 publications · 33 citations
1
h-index
1–5