Multilingual Word Embeddings for Low-Resource Languages using Anchors and a Chain of Related Languages
Not on the map. Why?
Abstract (arXiv)
Very low-resource languages, having only a few million tokens worth of data, are not well-supported by multilingual NLP approaches due to poor quality cross-lingual word representations. Recent work showed that good cross-lingual performance can be achieved if a source language is related to the low-resource target language. However, not all language pairs are related. In this paper, we propose to build multilingual word embeddings (MWEs) via a novel language chain-based approach, that incorporates intermediate related languages to bridge the gap between the distant source and target. We build MWEs one language at a time by starting from the resource rich source and sequentially adding each language in the chain till we reach the target. We extend a semi-joint bilingual approach to multiple languages in order to eliminate the main weakness of previous works, i.e., independently trained monolingual embeddings, by anchoring the target language around the multilingual space. We evaluate our method on bilingual lexicon induction for 4 language families, involving 4 very low-resource (<5M tokens) and 4 moderately low-resource (<50M) target languages, showing improved performance in both categories. Additionally, our analysis reveals the importance of good quality embeddings for intermediate languages as well as the importance of leveraging anchor points from all languages in the multilingual space.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 22 of its 36 references that are in the map. First posted on arXiv 21 Nov 2023.
Builds on
22 of 36 references.
References · 22
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
- 2013Efficient Estimation of Word Representations in Vector Spacecited by 4,346
- 2019Unsupervised Cross-lingual Representation Learning at Scalecited by 3,907
- 2017Word Translation Without Parallel Datacited by 480
- 2018Massively Multilingual Sentence Embeddings for Zero-Shot Cross-Lingual Transfer and Beyondcited by 434
- 2017Unsupervised Machine Translation Using Monolingual Corpora Onlycited by 272
- 2013Exploiting Similarities among Languages for Machine Translationcited by 187
- 2018A robust self-learning method for fully unsupervised cross-lingual mappings of word embeddingscited by 145
- 2017A Survey Of Cross-lingual Word Embedding Modelscited by 142
- 2018Phrase-Based & Neural Unsupervised Machine Translationcited by 138
- 2019Cross-Lingual Alignment of Contextual Word Embeddings, with Applications to Zero-shot Dependency Parsingcited by 69
- 2018On the Limitations of Unsupervised Bilingual Dictionary Inductioncited by 61
10 more
- 2017Learning Language Representations for Typology Predictioncited by 40
- 2019Bilingual Lexicon Induction with Semi-supervision in Non-Isometric Embedding Spacescited by 35
- 2019Cross-lingual Alignment vs Joint Training: A Comparative Study and A Simple Unified Frameworkcited by 26
- 2020LNMap: Departures from Isomorphic Assumption in Bilingual Lexicon Induction Through Non-Linear Mapping in Latent Spacecited by 23
- 2018Learning Multilingual Word Embeddings in Latent Metric Space: A Geometric Approachcited by 22
- 2018Unsupervised Multilingual Word Embeddingscited by 16
- 2019Bilingual Lexicon Induction through Unsupervised Machine Translationcited by 16
- 2019Should All Cross-Lingual Embeddings Speak English?cited by 14
- 2020Beyond Offline Mapping: Learning Cross Lingual Word Embeddings through Context Anchoringcited by 8
- 2020Anchor-based Bilingual Word Embeddings for Low-Resource Languagescited by 6
Built on by
Citing papers per year: 2026: 1.
1 paper.