# Universal or Language-Family-Specific Script Unification for Cross-Lingual Transfer? A Case Study on Turkic Languages

arXiv: https://arxiv.org/abs/2608.09356 · HTML: https://mlsanity.com/paper/2608.09356/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 22 references that are in the map.
- First posted on arXiv 10 Aug 2026.

## Builds on (most cited first, 10 of 10)

- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [Enriching Word Vectors with Subword Information](https://mlsanity.com/paper/1607.04606/) (2016)
- [Neural Architectures for Named Entity Recognition](https://mlsanity.com/paper/1603.01360/) (2016)
- [Learning Word Vectors for 157 Languages](https://mlsanity.com/paper/1802.06893/) (2018)
- [Universal Dependencies v2: An Evergrowing Multilingual Treebank Collection](https://mlsanity.com/paper/2004.10643/) (2020)
- [CANINE: Pre-training an Efficient Tokenization-Free Encoder for Language Representation](https://mlsanity.com/paper/2103.06874/) (2021)
- [Transfer Learning across Low-Resource, Related Languages for Neural Machine Translation](https://mlsanity.com/paper/1708.09803/) (2017)
- [Alternative Input Signals Ease Transfer in Multilingual Machine Translation](https://mlsanity.com/paper/2110.07804/) (2021)
- [Scripts Through Time: A Survey of the Evolving Role of Transliteration in NLP](https://mlsanity.com/paper/2604.18722/) (2026)
- [TurkicNLP: An NLP Toolkit for Turkic Languages](https://mlsanity.com/paper/2602.19174/) (2026)

## Abstract (arXiv)

Closely related languages written in different scripts expose little surface overlap to multilingual models, limiting cross-lingual transfer. We compare two approaches to script unification: the general-purpose uroman romanizer and the family-specific Common Turkic Script (CTS). We train matched fastText models on transliterated Wikipedia corpora from 11 Turkic languages and evaluate them on WikiANN named entity recognition and Universal Dependencies part-of-speech tagging. CTS and uroman show no significant difference on NER, while both substantially outperform the official monolingual fastText baselines. POS results reveal no universal winner: language-specific differences are associated with the cross-lingual character n-gram coverage induced by each representation, while within-language coverage becomes more important when target-language supervision is available. Although CANINE-c achieves higher overall POS averages, the substantially simpler fastText-based systems remain competitive on several treebanks. Overall, the effectiveness of script unification depends on the language, the induced subword overlap, and the available supervision.