# Explaining and Mitigating Crosslingual Tokenizer Inequities

arXiv: https://arxiv.org/abs/2510.21909 · HTML: https://mlsanity.com/paper/2510.21909/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 34 of its 52 references that are in the map.
- First posted on arXiv 24 Oct 2025.

## Builds on (most cited first, 30 of 34)

- [Neural Machine Translation of Rare Words with Subword Units](https://mlsanity.com/paper/1508.07909/) (2015)
- [SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing](https://mlsanity.com/paper/1808.06226/) (2018)
- [Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling](https://mlsanity.com/paper/2304.01373/) (2023)
- [No Language Left Behind: Scaling Human-Centered Machine Translation](https://mlsanity.com/paper/2207.04672/) (2022)
- [Subword Regularization: Improving Neural Network Translation Models with Multiple Subword Candidates](https://mlsanity.com/paper/1804.10959/) (2018)
- [OLMo: Accelerating the Science of Language Models](https://mlsanity.com/paper/2402.00838/) (2024)
- [ByT5: Towards a token-free future with pre-trained byte-to-byte models](https://mlsanity.com/paper/2105.13626/) (2021)
- [SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model](https://mlsanity.com/paper/2502.02737/) (2025)
- [How Good is Your Tokenizer? On the Monolingual Performance of Multilingual Language Models](https://mlsanity.com/paper/2012.15613/) (2020)
- [Think before you speak: Training Language Models With Pause Tokens](https://mlsanity.com/paper/2310.02226/) (2023)
- [Quality at a Glance: An Audit of Web-Crawled Multilingual Datasets](https://mlsanity.com/paper/2103.12028/) (2021)
- [CANINE: Pre-training an Efficient Tokenization-Free Encoder for Language Representation](https://mlsanity.com/paper/2103.06874/) (2021)
- [Language Modeling Is Compression](https://mlsanity.com/paper/2309.10668/) (2023)
- [Language Model Tokenizers Introduce Unfairness Between Languages](https://mlsanity.com/paper/2305.15425/) (2023)
- [Byte Pair Encoding is Suboptimal for Language Model Pretraining](https://mlsanity.com/paper/2004.03720/) (2020)
- [Do All Languages Cost the Same? Tokenization in the Era of Commercial Language Models](https://mlsanity.com/paper/2305.13707/) (2023)
- [Tokenization Is More Than Compression](https://mlsanity.com/paper/2402.18376/) (2024)
- [XLM-V: Overcoming the Vocabulary Bottleneck in Multilingual Masked Language Models](https://mlsanity.com/paper/2301.10472/) (2023)
- [Getting the most out of your tokenizer for pre-training and domain adaptation](https://mlsanity.com/paper/2402.01035/) (2024)
- [SuperBPE: Space Travel for Language Models](https://mlsanity.com/paper/2503.13423/) (2025)
- [Tokenization Impacts Multilingual Language Modeling: Assessing Vocabulary Allocation and Overlap Across Languages](https://mlsanity.com/paper/2305.17179/) (2023)
- [Why do language models perform worse for morphologically complex languages?](https://mlsanity.com/paper/2411.14198/) (2024)
- [Unpacking Tokenization: Evaluating Text Compression and its Correlation with Model Performance](https://mlsanity.com/paper/2403.06265/) (2024)
- [MYTE: Morphology-Driven Byte Encoding for Better and Fairer Multilingual Language Modeling](https://mlsanity.com/paper/2403.10691/) (2024)
- [Boundless Byte Pair Encoding: Breaking the Pre-tokenization Barrier](https://mlsanity.com/paper/2504.00178/) (2025)
- [MAGNET: Improving the Multilingual Fairness of Language Models with Adaptive Gradient-Based Tokenization](https://mlsanity.com/paper/2407.08818/) (2024)
- [Goldfish: Monolingual Language Models for 350 Languages](https://mlsanity.com/paper/2408.10441/) (2024)
- [A Bit of a Problem: Measurement Disparities in Dataset Sizes Across Languages](https://mlsanity.com/paper/2403.00686/) (2024)
- [Bridging the Gap for Tokenizer-Free Language Models](https://mlsanity.com/paper/1908.10322/) (2019)
- [Egalitarian Language Representation in Language Models: It All Begins with Tokenizers](https://mlsanity.com/paper/2409.11501/) (2024)

## Built on by (most cited first, 7 of 7)

- [The Roots of Performance Disparity in Multilingual Language Models: Intrinsic Modeling Difficulty or Design Choices?](https://mlsanity.com/paper/2601.07220/) (2026)
- [Position: Don't Just "Fix it in Post": A Science of AI Must Study Training Dynamics](https://mlsanity.com/paper/2606.06533/) (2026)
- [Apples to Apples? Towards Comparable Crosslingual Language Model Evaluation](https://mlsanity.com/paper/2608.25089/) (2026)
- [TokEval: A Tokenizer Evaluation Suite](https://mlsanity.com/paper/2608.18062/) (2026)
- [An In-Vitro Study on Cross-Lingual Generalization in Language Models](https://mlsanity.com/paper/2605.26683/) (2026)
- [Fewer Words, Not Fewer Tokens: Measuring the Sanskrit Tokenization Penalty per Proposition](https://mlsanity.com/paper/2609.12960/) (2026)
- [To Each Language Its Tokenizer: Modular Tokenizers for Efficient Multilingual LLMs](https://mlsanity.com/paper/2609.15528/) (2026)

## Abstract (arXiv)

The number of tokens it takes to encode parallel text in different languages is known to vary. These disparities are called token premiums. Having high token premiums leads to less throughput during training and increases costs at inference. In this paper, we show that even after controlling for dataset size, vocabulary size, and data content, monolingual tokenizers exhibit a wide range of token premiums across languages. To understand the cross-linguistic differences that cause these token premiums, we train a suite of approximately 7,000 comparable monolingual tokenizers for 97 languages, manipulating tokenization algorithm, vocabulary size, and dataset size. We measure token premiums and test for a relationship between factors such as data similarity (between tokenizer training and evaluation), vocabulary size, and pre-tokenization. We also investigate the role of language-specific features such as writing system and word length. We find that similarity between training and test data does not impact token premiums, but vocabulary size and pre-tokenization do. While simply increasing vocabulary size does not lead to reduced token premium effects, we can determine an ``optimal'' vocabulary size for each language to achieve significantly reduced token premium effects. We also train superword tokenizers which allow merges over whitespaces, and we find that they both reduce token premium effects and improve compression overall. Thus, intervening on the vocabulary size or the pre-tokenizer significantly reduces crosslingual token premium effects.