# Biomedical Machine Translation for Low-Resource Arabic-Script Languages via Cross-Lingual Transfer and LoRA Adapter Merging

arXiv: https://arxiv.org/abs/2607.22300 · HTML: https://mlsanity.com/paper/2607.22300/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 51 references that are in the map.
- First posted on arXiv 24 Jul 2026.

## Builds on (most cited first, 23 of 23)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Neural Machine Translation by Jointly Learning to Align and Translate](https://mlsanity.com/paper/1409.0473/) (2014)
- [Parameter-Efficient Transfer Learning for NLP](https://mlsanity.com/paper/1902.00751/) (2019)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [QLoRA: Efficient Finetuning of Quantized LLMs](https://mlsanity.com/paper/2305.14314/) (2023)
- [Sequence to Sequence Learning with Neural Networks](https://mlsanity.com/paper/1409.3215/) (2014)
- [Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time](https://mlsanity.com/paper/2203.05482/) (2022)
- [COMET: A Neural Framework for MT Evaluation](https://mlsanity.com/paper/2009.09025/) (2020)
- [Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch](https://mlsanity.com/paper/2311.03099/) (2023)
- [Google's Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation](https://mlsanity.com/paper/1611.04558/) (2016)
- [Massively Multilingual Neural Machine Translation](https://mlsanity.com/paper/1903.00089/) (2019)
- [The unreasonable effectiveness of few-shot learning for machine translation](https://mlsanity.com/paper/2302.01398/) (2023)
- [TICO-19: the Translation Initiative for Covid-19](https://mlsanity.com/paper/2007.01788/) (2020)
- [AdaMergeX: Cross-Lingual Transfer with Large Language Models via Adaptive Adapter Merging](https://mlsanity.com/paper/2402.18913/) (2024)
- [A Comparison of Transformer and Recurrent Neural Networks on Multilingual Neural Machine Translation](https://mlsanity.com/paper/1806.06957/) (2018)
- [Physician Detection of Clinical Harm in Machine Translation: Quality Estimation Aids in Reliance and Backtranslation Identifies Critical Errors](https://mlsanity.com/paper/2310.16924/) (2023)
- [Language-Family Adapters for Low-Resource Multilingual Neural Machine Translation](https://mlsanity.com/paper/2209.15236/) (2022)
- [The Unreasonable Effectiveness of Model Merging for Cross-Lingual Transfer in LLMs](https://mlsanity.com/paper/2505.18356/) (2025)
- [Graphemic Normalization of the Perso-Arabic Script](https://mlsanity.com/paper/2210.12273/) (2022)
- [Advancing Dialectal Arabic to Modern Standard Arabic Machine Translation](https://mlsanity.com/paper/2507.20301/) (2025)
- [The interplay between language similarity and script on a novel multi-layer Algerian dialect corpus](https://mlsanity.com/paper/2105.07400/) (2021)
- [Pivot Language for Low-Resource Machine Translation](https://mlsanity.com/paper/2505.14553/) (2025)
- [Cross-Lingual Empirical Evaluation of Large Language Models for Arabic Medical Tasks](https://mlsanity.com/paper/2602.05374/) (2026)

## Abstract (arXiv)

We present a systematic study of healthcare-domain cross-lingual transfer to address the scarcity of biomedical NMT resources for Arabic-script languages. We use Arabic and Persian as higher-resource pivots to improve translation for four severely low-resource targets: Dari (Afghan Persian, a standardised variety of Persian), Pashto, Sorani Kurdish (Central Kurdish, a major standardized variety of Kurdish), and Urdu (closely related to Hindi). Using LoRA fine-tuning on small decoder-only LLMs, we train textitdomain-specific pivot adapters and evaluate three transfer strategies: few-shot in-context learning, minimal supervised adaptation, and, to the best of our knowledge, for the first time in this setting, zero-data LoRA adapter merging. Supervised adaptation with just 500 sentences achieves near pivot-language quality for Dari (CHrF++ 41.01) and meaningful gains for Urdu (28.88), while adapter merging reaches within 3.5 CHrF++ of supervised adaptation for Dari at zero additional cost. Pashto and Sorani Kurdish remain insufficient for high-stakes clinical deployment exposing the limits of cross-lingual transfer when structural distance from the pivots is too great. LoRA adapter merging works surprisingly well for closely related languages, even without target-language biomedical data.