Learning Geometric Word Meta-Embeddings
Not on the map. Why?
Abstract (arXiv)
We propose a geometric framework for learning meta-embeddings of words from different embedding sources. Our framework transforms the embeddings into a common latent space, where, for example, simple averaging of different embeddings (of a given word) is more amenable. The proposed latent space arises from two particular geometric transformations - the orthogonal rotations and the Mahalanobis metric scaling. Empirical results on several word similarity and word analogy benchmarks illustrate the efficacy of the proposed framework.
Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). First posted on arXiv 20 Apr 2020.
Builds on
No reference list yet.
Built on by
Citing papers per year: 2021: 2; 2022: 3; 2023: 1.
6 papers.
- 2021On Riemannian Optimization over Positive Definite Matrices with the Bures-Wasserstein Geometrycited by 32
- 2022A Survey on Word Meta-Embedding Learningcited by 7
- 2022Learning Meta Word Embeddings by Unsupervised Weighted Concatenation of Source Embeddingscited by 5
- 2022Unsupervised Attention-based Sentence-Level Meta-Embeddings from Contextualised Language Modelscited by 4
- 2021A Meta-embedding-based Ensemble Approach for ICD Coding Predictioncited by 2
- 2023Together We Make Sense -- Learning Meta-Sense Embeddings from Pretrained Static Sense Embeddingscited by 1
- 2023Together We Make Sense -- Learning Meta-Sense Embeddings from Pretrained Static Sense Embeddings30 May 2023
- 2022Learning Meta Word Embeddings by Unsupervised Weighted Concatenation of Source Embeddings26 Apr 2022
- 2022A Survey on Word Meta-Embedding Learning25 Apr 2022
- 2022Unsupervised Attention-based Sentence-Level Meta-Embeddings from Contextualised Language Models16 Apr 2022
- 2021On Riemannian Optimization over Positive Definite Matrices with the Bures-Wasserstein Geometry1 Jun 2021
- 2021A Meta-embedding-based Ensemble Approach for ICD Coding Prediction26 Feb 2021