Multimodal Representation Learning
Multimodal Representation Learning is a subfield of representation learning that focuses on learning joint, interoperable representations from multiple distinct modalities—such as text, audio, image, and video—by modeling their semantic correspondences and complementary information to enable coherent understanding across heterogeneous data sources.
2K
Publications
164K
Citations
8K
Authors
1.1K
Institutions
Publications per year
2017–2026
1.7K
2K
Jiquan Ngiam, Aditya Khosla, Mingyu Kim et al. · 2011 · 2.3K citations
LXMERT: Learning Cross-Modality Encoder Representations from Transformers
Hao Tan, Mohit Bansal · 2019 · 2.3K citations · Full text
1–5 of 2K
8K
26
h-index
18
h-index
18
h-index
17
h-index
University of Electronic Science and Technology of China, China
16 publications · 1.4K citations
15
h-index
1–5 of 8K
1.1K

Chinese Academy of Sciences
Beijing, China
50 h-index
310 publications
25.6K citations
Tsinghua University
Beijing, China
41 h-index
146 publications
17.8K citations
University of Chinese Academy of Sciences
Beijing, China
41 h-index
193 publications
14.3K citations

Peking University
Beijing, China
36 h-index
120 publications
12.4K citations
National University of Singapore
Singapore, Singapore
35 h-index
113 publications
11.8K citations
Institute of Automation
Beijing, China
31 h-index
151 publications
14.6K citations

Zhejiang University
Hangzhou, China
31 h-index
156 publications
8.5K citations

Sun Yat-sen University
Guangzhou, China
29 h-index
103 publications
8.6K citations
1–8 of 1.1K
Repository · 343 publications · 24.6K citations
72
h-index
80 publications · 5.3K citations
43
h-index
Conference · 75 publications · 7.2K citations
43
h-index
TPAMI · 42 publications · 10.5K citations
32
h-index
39 publications · 3.1K citations
29
h-index
1–5