# A Subword Embedding Approach for Variation Detection in Luxembourgish User Comments

arXiv: https://arxiv.org/abs/2602.11795 · HTML: https://mlsanity.com/paper/2602.11795/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 35 references that are in the map.
- First posted on arXiv 12 Feb 2026.

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

- [Enriching Word Vectors with Subword Information](https://mlsanity.com/paper/1607.04606/) (2016)
- [Multilingual Alignment of Contextual Word Representations](https://mlsanity.com/paper/2002.03518/) (2020)
- [False Friends Are Not Foes: Investigating Vocabulary Overlap in Multilingual Language Models](https://mlsanity.com/paper/2509.18750/) (2025)
- [LuxBank: The First Universal Dependency Treebank for Luxembourgish](https://mlsanity.com/paper/2411.04813/) (2024)
- [Text Generation Models for Luxembourgish with Limited Data: A Balanced Multilingual Strategy](https://mlsanity.com/paper/2412.09415/) (2024)
- [Neural Text Normalization for Luxembourgish using Real-Life Variation Data](https://mlsanity.com/paper/2412.09383/) (2024)
- [Unsupervised Context-Sensitive Spelling Correction of English and Dutch Clinical Free-Text with Word and Character N-Gram Embeddings](https://mlsanity.com/paper/1710.07045/) (2017)
- [Forget NLI, Use a Dictionary: Zero-Shot Topic Classification for Low-Resource Languages with Application to Luxembourgish](https://mlsanity.com/paper/2404.03912/) (2024)
- [Towards Better Inclusivity: A Diverse Tweet Corpus of English Varieties](https://mlsanity.com/paper/2401.11487/) (2024)
- [DialectGram: Detecting Dialectal Variation at Multiple Geographic Resolutions](https://mlsanity.com/paper/1910.01818/) (2019)

## Abstract (arXiv)

This paper presents an embedding-based approach to detecting variation without relying on prior normalisation or predefined variant lists. The method trains subword embeddings on raw text and groups related forms through combined cosine and n-gram similarity. This allows spelling and morphological diversity to be examined and analysed as linguistic structure rather than treated as noise. Using a large corpus of Luxembourgish user comments, the approach uncovers extensive lexical and orthographic variation that aligns with patterns described in dialectal and sociolinguistic research. The induced families capture systematic correspondences and highlight areas of regional and stylistic differentiation. The procedure does not strictly require manual annotation, but does produce transparent clusters that support both quantitative and qualitative analysis. The results demonstrate that distributional modelling can reveal meaningful patterns of variation even in ''noisy'' or low-resource settings, offering a reproducible methodological framework for studying language variety in multilingual and small-language contexts.