Contextualized Embeddings
Contextualized Embeddings is a type of distributed representation in natural language processing that, unlike static embeddings, generates word or token vectors dynamically based on the surrounding context within a sentence or document, thereby capturing nuanced semantic and syntactic meanings specific to each usage instance.
397
Publications
24.8K
Citations
1.1K
Authors
396
Institutions
Publications per year
2017–2026
372
Parents
397
Deep Contextualized Word Representations
Matthew E. Peters, Mark E Neumann, Mohit Iyyer et al. · 2018 · 1.8K citations · Full text
Dependency-Based Word Embeddings
Omer Levy, Yoav Goldberg · 2014 · 1.1K citations · Full text
LUKE: Deep Contextualized Entity Representations with Entity-aware Self-attention
Ikuya Yamada, Akari Asai, Hiroyuki Shindo et al. · 2020 · 564 citations · Full text
1–5 of 397
1.1K
8
h-index
6
h-index
6
h-index
6
h-index
5
h-index
1–5 of 1.1K
396
Google (United States)
Mountain View, United States
14 h-index
41 publications
2.8K citations
University of Washington
Seattle, United States
10 h-index
17 publications
1.1K citations
Carnegie Mellon University
Pittsburgh, United States
9 h-index
25 publications
3K citations
University of Southern California
Los Angeles, United States
8 h-index
22 publications
957 citations
Technion – Israel Institute of Technology
Haifa, Israel
7 h-index
19 publications
963 citations

Johns Hopkins University
Baltimore, United States
7 h-index
21 publications
1.4K citations
Westlake University
Hangzhou, China
7 h-index
18 publications
861 citations
Stanford University
Stanford, United States
7 h-index
15 publications
501 citations
1–8 of 396
Repository · 80 publications · 3.6K citations
32
h-index
Conference · 14 publications · 651 citations
12
h-index
10 publications · 523 citations
10
h-index
TASLP · 6 publications · 437 citations
6
h-index
Conference · 5 publications · 105 citations
5
h-index
1–5