# Geographical Distance Is The New Hyperparameter: A Case Study Of Finding The Optimal Pre-trained Language For English-isiZulu Machine Translation

arXiv: https://arxiv.org/abs/2205.08621 · HTML: https://mlsanity.com/paper/2205.08621/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 29 references that are in the map.
- First posted on arXiv 17 May 2022.

## Builds on (most cited first, 16 of 16)

- [Improving Neural Machine Translation Models with Monolingual Data](https://mlsanity.com/paper/1511.06709/) (2015)
- [Six Challenges for Neural Machine Translation](https://mlsanity.com/paper/1706.03872/) (2017)
- [Marian: Fast Neural Machine Translation in C++](https://mlsanity.com/paper/1804.00344/) (2018)
- [Transfer Learning for Low-Resource Neural Machine Translation](https://mlsanity.com/paper/1604.02201/) (2016)
- [Choosing Transfer Languages for Cross-Lingual Learning](https://mlsanity.com/paper/1905.12688/) (2019)
- [Survey of Low-Resource Machine Translation](https://mlsanity.com/paper/2109.00486/) (2021)
- [Rapid Adaptation of Neural Machine Translation to New Languages](https://mlsanity.com/paper/1808.04189/) (2018)
- [Tagged Back-Translation](https://mlsanity.com/paper/1906.06442/) (2019)
- [Revisiting Low-Resource Neural Machine Translation: A Case Study](https://mlsanity.com/paper/1905.11901/) (2019)
- [Transfer Learning across Low-Resource, Related Languages for Neural Machine Translation](https://mlsanity.com/paper/1708.09803/) (2017)
- [On The Evaluation of Machine Translation Systems Trained With Back-Translation](https://mlsanity.com/paper/1908.05204/) (2019)
- [Dynamic Data Selection and Weighting for Iterative Back-Translation](https://mlsanity.com/paper/2004.03672/) (2020)
- [A Focus on Neural Machine Translation for African Languages](https://mlsanity.com/paper/1906.05685/) (2019)
- [Low Resource Neural Machine Translation: A Benchmark for Five African Languages](https://mlsanity.com/paper/2003.14402/) (2020)
- [Canonical and Surface Morphological Segmentation for Nguni Languages](https://mlsanity.com/paper/2104.00767/) (2021)
- [Low-Resource Neural Machine Translation for Southern African Languages](https://mlsanity.com/paper/2104.00366/) (2021)

## Built on by (most cited first, 5 of 5)

- [LLMatic: Neural Architecture Search via Large Language Models and Quality Diversity Optimization](https://mlsanity.com/paper/2306.01102/) (2023)
- [NLNDE at SemEval-2023 Task 12: Adaptive Pretraining and Source Language Selection for Low-Resource Multilingual Sentiment Analysis](https://mlsanity.com/paper/2305.00090/) (2023)
- [Pre-Trained Language Models Represent Some Geographic Populations Better Than Others](https://mlsanity.com/paper/2403.11025/) (2024)
- [MTet: Multi-domain Translation for English and Vietnamese](https://mlsanity.com/paper/2210.05610/) (2022)
- [Enriching Biomedical Knowledge for Low-resource Language Through Large-Scale Translation](https://mlsanity.com/paper/2210.05598/) (2022)

## Abstract (arXiv)

Stemming from the limited availability of datasets and textual resources for low-resource languages such as isiZulu, there is a significant need to be able to harness knowledge from pre-trained models to improve low resource machine translation. Moreover, a lack of techniques to handle the complexities of morphologically rich languages has compounded the unequal development of translation models, with many widely spoken African languages being left behind. This study explores the potential benefits of transfer learning in an English-isiZulu translation framework. The results indicate the value of transfer learning from closely related languages to enhance the performance of low-resource translation models, thus providing a key strategy for low-resource translation going forward. We gathered results from 8 different language corpora, including one multi-lingual corpus, and saw that isiXhosa-isiZulu outperformed all languages, with a BLEU score of 8.56 on the test set which was better from the multi-lingual corpora pre-trained model by 2.73. We also derived a new coefficient, Nasir's Geographical Distance Coefficient (NGDC) which provides an easy selection of languages for the pre-trained models. NGDC also indicated that isiXhosa should be selected as the language for the pre-trained model.