# When your Cousin has the Right Connections: Unsupervised Bilingual Lexicon Induction for Related Data-Imbalanced Languages

arXiv: https://arxiv.org/abs/2305.14012 · HTML: https://mlsanity.com/paper/2305.14012/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 46 references that are in the map.
- First posted on arXiv 23 May 2023.

## Builds on (most cited first, 19 of 19)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Efficient Estimation of Word Representations in Vector Space](https://mlsanity.com/paper/1301.3781/) (2013)
- [Enriching Word Vectors with Subword Information](https://mlsanity.com/paper/1607.04606/) (2016)
- [Word Translation Without Parallel Data](https://mlsanity.com/paper/1710.04087/) (2017)
- [Are All Languages Created Equal in Multilingual BERT?](https://mlsanity.com/paper/2005.09093/) (2020)
- [A robust self-learning method for fully unsupervised cross-lingual mappings of word embeddings](https://mlsanity.com/paper/1805.06297/) (2018)
- [A Survey Of Cross-lingual Word Embedding Models](https://mlsanity.com/paper/1706.04902/) (2017)
- [Enabling Language Models to Fill in the Blanks](https://mlsanity.com/paper/2005.05339/) (2020)
- [Bilingual Lexicon Induction with Semi-supervision in Non-Isometric Embedding Spaces](https://mlsanity.com/paper/1908.06625/) (2019)
- [L3Cube-HindBERT and DevBERT: Pre-Trained BERT Transformer models for Devanagari based Hindi and Marathi Languages](https://mlsanity.com/paper/2211.11418/) (2022)
- [Do We Really Need Fully Unsupervised Cross-Lingual Embeddings?](https://mlsanity.com/paper/1909.01638/) (2019)
- [It's not Greek to mBERT: Inducing Word-Level Translations from Multilingual BERT](https://mlsanity.com/paper/2010.08275/) (2020)
- [Socially Aware Bias Measurements for Hindi Language Representations](https://mlsanity.com/paper/2110.07871/) (2021)
- [Bilingual Lexicon Induction through Unsupervised Machine Translation](https://mlsanity.com/paper/1907.10761/) (2019)
- [Interactive Refinement of Cross-Lingual Word Embeddings](https://mlsanity.com/paper/1911.03070/) (2019)
- [Cross-Lingual Word Embeddings for Turkic Languages](https://mlsanity.com/paper/2005.08340/) (2020)
- [Combining Static Word Embeddings and Contextual Representations for Bilingual Lexicon Induction](https://mlsanity.com/paper/2106.03084/) (2021)
- [Linguistic Resources for Bhojpuri, Magahi and Maithili: Statistics about them, their Similarity Estimates, and Baselines for Three Applications](https://mlsanity.com/paper/2004.13945/) (2020)
- [Anchor-based Bilingual Word Embeddings for Low-Resource Languages](https://mlsanity.com/paper/2010.12627/) (2020)

## Built on by (most cited first, 3 of 3)

- [LexC-Gen: Generating Data for Extremely Low-Resource Languages with Large Language Models and Bilingual Lexicons](https://mlsanity.com/paper/2402.14086/) (2024)
- [LexGen: Domain-aware Multilingual Lexicon Generation](https://mlsanity.com/paper/2405.11200/) (2024)
- [DialUp! Modeling the Language Continuum by Adapting Models to Dialects and Dialects to Models](https://mlsanity.com/paper/2501.16581/) (2025)

## Abstract (arXiv)

Most existing approaches for unsupervised bilingual lexicon induction (BLI) depend on good quality static or contextual embeddings requiring large monolingual corpora for both languages. However, unsupervised BLI is most likely to be useful for low-resource languages (LRLs), where large datasets are not available. Often we are interested in building bilingual resources for LRLs against related high-resource languages (HRLs), resulting in severely imbalanced data settings for BLI. We first show that state-of-the-art BLI methods in the literature exhibit near-zero performance for severely data-imbalanced language pairs, indicating that these settings require more robust techniques. We then present a new method for unsupervised BLI between a related LRL and HRL that only requires inference on a masked language model of the HRL, and demonstrate its effectiveness on truly low-resource languages Bhojpuri and Magahi (with <5M monolingual tokens each), against Hindi. We further present experiments on (mid-resource) Marathi and Nepali to compare approach performances by resource range, and release our resulting lexicons for five low-resource Indic languages: Bhojpuri, Magahi, Awadhi, Braj, and Maithili, against Hindi.