# Embedding Initialization for Unseen Low-resource Languages in Multilingual NMT: A Case Study on Limbum-English Translation

arXiv: https://arxiv.org/abs/2608.07629 · HTML: https://mlsanity.com/paper/2608.07629/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 19 references that are in the map.
- First posted on arXiv 7 Aug 2026.

## Builds on (most cited first, 13 of 13)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Neural Machine Translation of Rare Words with Subword Units](https://mlsanity.com/paper/1508.07909/) (2015)
- [A Call for Clarity in Reporting BLEU Scores](https://mlsanity.com/paper/1804.08771/) (2018)
- [No Language Left Behind: Scaling Human-Centered Machine Translation](https://mlsanity.com/paper/2207.04672/) (2022)
- [The State and Fate of Linguistic Diversity and Inclusion in the NLP World](https://mlsanity.com/paper/2004.09095/) (2020)
- [Google's Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation](https://mlsanity.com/paper/1611.04558/) (2016)
- [MAD-X: An Adapter-Based Framework for Multi-Task Cross-Lingual Transfer](https://mlsanity.com/paper/2005.00052/) (2020)
- [Transfer Learning for Low-Resource Neural Machine Translation](https://mlsanity.com/paper/1604.02201/) (2016)
- [Participatory Research for Low-resourced Machine Translation: A Case Study in African Languages](https://mlsanity.com/paper/2010.02353/) (2020)
- [A Few Thousand Translations Go a Long Way! Leveraging Pre-trained Models for African News Translation](https://mlsanity.com/paper/2205.02022/) (2022)
- [Rapid Adaptation of Neural Machine Translation to New Languages](https://mlsanity.com/paper/1808.04189/) (2018)
- [Improving Cross-Lingual Word Embeddings by Meeting in the Middle](https://mlsanity.com/paper/1808.08780/) (2018)
- [MMTAfrica: Multilingual Machine Translation for African Languages](https://mlsanity.com/paper/2204.04306/) (2022)

## Abstract (arXiv)

Multilingual neural machine translation models such as NLLB-200 cover 200 languages but leave thousands unsupported, including most Grassfields Bantu languages of Cameroon. When fine-tuning these models for an unseen language, practitioners must choose a proxy language token, yet no principled method exists for this selection. We implemented an embedding initialization strategy where a language token is the average of embeddings from multiple typologically related languages already in the mod el. We evaluate this approach on Limbum-to-English translation using a parallel corpus of 8,837 sentence pairs from New Testament text and a bilingual dictionary. We compare models: NLLB-200 zero-shot (chrF2++ = 12.5), a Transformer trained from scratch (chrF2++ = 14.5), NLLB-200 fine-tuned with a Swahili proxy token (chrF2++ = 47.3), and NLLB-200 with our averaged embedding initialization (chrF2++ = 46.7). We find that the multi-language initialization achieves performance comparable to the best single-language proxy. Both NLLB-200 variants improve over the from-scratch baseline by over 32 chrF2++ points. These results show that multilingual transfer is the dominant factor in extremely low-resource Bantu translation while eliminating the need for heuristic proxy selection. However, all systems fail to preserve tonal diacritics, highlighting an open challenge. We make our dataset and code available to support further research.