The Curse of Multilinguality in Lexical Normalization

• URL: http://arxiv.org/abs/2609.00329v1
• Date: Thu, 27 Aug 2026 22:48:02 GMT
• Title: The Curse of Multilinguality in Lexical Normalization
• Abstract summary: A popular shortcut is to train a single model on many languages at once.<n>We ask how many languages should such a model be trained on?<n>We find a clear curse of multilinguality: accuracy is highest when a language is trained with only a few others.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Lexical normalization rewrites the noisy, non-standard words that fill user-generated text (tmrw, u, gr8) into their standard forms. Because labelled data is scarce for most languages, a popular shortcut is to train a single model on many languages at once. We ask a simple question: how many languages should such a model be trained on? Using one fixed-capacity character-level model and twelve languages from a standard benchmark, we vary the number of jointly trained languages from one to twelve and measure per-language accuracy. We find a clear curse of multilinguality: accuracy is highest when a language is trained with only a few others, often just one to four, and then falls steadily and substantially, dropping by about forty percent as the rest are piled on. A control that holds the total amount of training data constant makes the decline arrive sooner and fall further, which points to competition among the languages for one fixed-size model rather than to how much data is available. We also test whether a language's typological distance from the others predicts its ideal number of co-training languages, and find no dependable rule: any apparent relationship rests on a couple of languages and does not hold up. For compact normalization models, less can be more: a few languages beat pooling everything into a single model.

Related papers

• Flexing in 73 Languages: A Single Small Model for Multilingual Inflection [0.5934049163128877]
  We present a compact, single-model approach to multilingual inflection.<n>Our model is lightweight, robust to unseen words, and outperforms monolingual baselines in most languages.<n>Our work addresses the lack of an open-source, general-purpose, multilingual morphological inflection system.
  arXiv  Detail & Related papers  (2025-10-27T08:34:41Z)
• Code-mixed Sentiment and Hate-speech Prediction [2.9140539998069803]
  Large language models have dominated most natural language processing tasks. We created four new bilingual pre-trained masked language models for English-Hindi and English-Slovene languages. We performed an evaluation of monolingual, bilingual, few-lingual, and massively multilingual models on several languages.
  arXiv  Detail & Related papers  (2024-05-21T16:56:36Z)
• The Role of Language Imbalance in Cross-lingual Generalisation: Insights from Cloned Language Experiments [57.273662221547056]
  In this study, we investigate an unintuitive novel driver of cross-lingual generalisation: language imbalance. We observe that the existence of a predominant language during training boosts the performance of less frequent languages. As we extend our analysis to real languages, we find that infrequent languages still benefit from frequent ones, yet whether language imbalance causes cross-lingual generalisation there is not conclusive.
  arXiv  Detail & Related papers  (2024-04-11T17:58:05Z)
• Poro 34B and the Blessing of Multilinguality [3.270981284471548]
  Poro 34B is a 34 billion parameter model trained for 1 trillion tokens of Finnish, English, and programming languages.<n>We show that a multilingual training approach can produce a model that substantially advances over the capabilities of existing models for Finnish.
  arXiv  Detail & Related papers  (2024-04-02T11:34:12Z)
• Lifting the Curse of Multilinguality by Pre-training Modular Transformers [72.46919537293068]
  multilingual pre-trained models suffer from the curse of multilinguality, which causes per-language performance to drop as they cover more languages. We introduce language-specific modules, which allows us to grow the total capacity of the model, while keeping the total number of trainable parameters per language constant. Our approach enables adding languages post-hoc with no measurable drop in performance, no longer limiting the model usage to the set of pre-trained languages.
  arXiv  Detail & Related papers  (2022-05-12T17:59:56Z)
• Discovering Representation Sprachbund For Multilingual Pre-Training [139.05668687865688]
  We generate language representation from multilingual pre-trained models and conduct linguistic analysis. We cluster all the target languages into multiple groups and name each group as a representation sprachbund. Experiments are conducted on cross-lingual benchmarks and significant improvements are achieved compared to strong baselines.
  arXiv  Detail & Related papers  (2021-09-01T09:32:06Z)
• Are Multilingual Models the Best Choice for Moderately Under-resourced Languages? A Comprehensive Assessment for Catalan [0.05277024349608833]
  This work focuses on Catalan with the aim of exploring what extent a medium-sized monolingual language model is competitive with state-of-the-art large multilingual models. We build a clean, high-quality textual Catalan corpus (CaText), train a Transformer-based language model for Catalan (BERTa), and devise a thorough evaluation in a diversity of settings. The result is a new benchmark, the Catalan Language Understanding Benchmark (CLUB), which we publish as an open resource.
  arXiv  Detail & Related papers  (2021-07-16T13:52:01Z)
• How Good is Your Tokenizer? On the Monolingual Performance of Multilingual Language Models [96.32118305166412]
  We study a set of nine typologically diverse languages with readily available pretrained monolingual models on a set of five diverse monolingual downstream tasks. We find that languages which are adequately represented in the multilingual model's vocabulary exhibit negligible performance decreases over their monolingual counterparts.
  arXiv  Detail & Related papers  (2020-12-31T14:11:00Z)
• When Being Unseen from mBERT is just the Beginning: Handling New Languages With Multilingual Language Models [2.457872341625575]
  Transfer learning based on pretraining language models on a large amount of raw data has become a new norm to reach state-of-the-art performance in NLP. We show that such models behave in multiple ways on unseen languages.
  arXiv  Detail & Related papers  (2020-10-24T10:15:03Z)
• Multilingual Translation with Extensible Multilingual Pretraining and Finetuning [77.33262578776291]
  Previous work has demonstrated that machine translation systems can be created by finetuning on bitext. We show that multilingual translation models can be created through multilingual finetuning. We demonstrate that pretrained models can be extended to incorporate additional languages without loss of performance.
  arXiv  Detail & Related papers  (2020-08-02T05:36:55Z)
• XCOPA: A Multilingual Dataset for Causal Commonsense Reasoning [68.57658225995966]
  Cross-lingual Choice of Plausible Alternatives (XCOPA) is a typologically diverse multilingual dataset for causal commonsense reasoning in 11 languages. We evaluate a range of state-of-the-art models on this novel dataset, revealing that the performance of current methods falls short compared to translation-based transfer.
  arXiv  Detail & Related papers  (2020-05-01T12:22:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.