Equity with Efficiency: An Empirical Study of Tokenizers for Multilingual Large Language Models

• URL: http://arxiv.org/abs/2606.15044v1
• Date: Sat, 13 Jun 2026 01:10:42 GMT
• Title: Equity with Efficiency: An Empirical Study of Tokenizers for Multilingual Large Language Models
• Abstract summary: Byte-level Byte-Pair tokenizers structurally favor high-resource languages and Latin scripts.<n>For speakers of underrepresented languages, this bias inflates inference costs and widens cross-lingual capability gaps.<n>We present the first systematic comparison of equitable tokenizers on a benchmark spanning 11 Southeast Asian languages.
• Score: 30.7730386396743
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multilingual large language models (LLMs) depend on subword tokenization to bridge discrete text and continuous neural representation. State-of-the-art multilingual LLMs often use Byte-level Byte-Pair Encoding (BPE) tokenizers that structurally favor high-resource languages and Latin scripts. For speakers of underrepresented languages, particularly those across Southeast Asia, this bias inflates inference costs and widens cross-lingual capability gaps. We present the first systematic comparison of equitable tokenizers on a unified benchmark spanning 11 Southeast Asian languages. Beyond tokenizer-level analysis of compression efficiency and cross-lingual equity, we assess downstream task performance through controlled 1.5B-parameter language model training using the same training data. Our results show that Parity-aware BPE lies on the Pareto frontier of the efficiency-equity trade-off, achieving strong compression parity at competitive cost. Morphology-Driven Byte Encoding delivers the best semantic reasoning performance through morphologically richer representations, albeit at a higher computational expense. Byte Latent Transformer underperforms on downstream tasks, possibly because its architectural assumptions misalign with the constraints of limited low-resource training data. Together, our findings demonstrate that cross-lingual fairness and tokenization efficiency are not fundamentally at odds, and offer practical guidance for designing equitable multilingual models.

Related papers

• UNICS: Multilingual Code Search via Unified Pseudocode and Contrastive Transfer Learning [20.215764918447643]
  We introduce UNICS, a framework for multilingual code search built on a two-stage training strategy.<n>In the first stage, UNICS is pre-trained on a novel dataset we constructed, which uses pseudo-code as a unified representation.<n>The second stage employs a multi-task transfer learning strategy that adapts this general knowledge to specific languages.
  arXiv  Detail & Related papers  (2026-06-26T06:03:33Z)
• Beyond Many-Shot Translation: Scaling In-Context Demonstrations For Low-Resource Machine Translation [49.82863380286994]
  In-context learning may offer novel ways to adapt Large Language Models for low-resource machine translation.<n>In this study, we explore scaling low-resource machine translation ICL beyond the few-shot setting to thousands of examples with long-context models.<n>Our experiments on Javanese and Sundanese show that gains from additional context saturate quickly and can degrade near the maximum context window.
  arXiv  Detail & Related papers  (2026-02-04T17:02:22Z)
• Tokenization Disparities as Infrastructure Bias: How Subword Systems Create Inequities in LLM Access and Efficiency [6.943451388015595]
  Tokenization disparities pose a significant barrier to achieving equitable access to artificial intelligence.<n>This study conducts a large-scale cross-linguistic evaluation of tokenization efficiency in over 200 languages.
  arXiv  Detail & Related papers  (2025-10-14T11:14:38Z)
• Parity-Aware Byte-Pair Encoding: Improving Cross-lingual Fairness in Tokenization [53.22544362024936]
  Tokenization is the first -- and often least scrutinized -- step of most NLP pipelines.<n>Standard algorithms for learning tokenizers rely on frequency-based objectives.<n>We introduce Parity-aware Byte Pair.<n>We find empirically that Parity-aware BPE leads to more equitable token counts across languages.
  arXiv  Detail & Related papers  (2025-08-06T18:14:43Z)
• When Every Token Counts: Optimal Segmentation for Low-Resource Language Models [0.0]
  We show that an optimal Byte-Pair (BPE) configuration significantly reduces token count compared to greedy segmentation.<n>Our findings suggest that compression-optimized tokenization strategies could provide substantial advantages for multilingual and low-resource language applications.
  arXiv  Detail & Related papers  (2024-12-09T19:11:54Z)
• Tokenizer Choice For LLM Training: Negligible or Crucial? [30.33170936148845]
  We study the influence of tokenizer choice on Large Language Models (LLMs) downstream performance by training 24 mono- and multilingual LLMs. We find that the tokenizer choice can significantly impact the model's downstream performance and training costs. We show that multilingual tokenizers trained on the five most frequent European languages require vocabulary size increases of factor three in comparison to English.
  arXiv  Detail & Related papers  (2023-10-12T22:44:19Z)
• VECO 2.0: Cross-lingual Language Model Pre-training with Multi-granularity Contrastive Learning [56.47303426167584]
  We propose a cross-lingual pre-trained model VECO2.0 based on contrastive learning with multi-granularity alignments. Specifically, the sequence-to-sequence alignment is induced to maximize the similarity of the parallel pairs and minimize the non-parallel pairs. token-to-token alignment is integrated to bridge the gap between synonymous tokens excavated via the thesaurus dictionary from the other unpaired tokens in a bilingual instance.
  arXiv  Detail & Related papers  (2023-04-17T12:23:41Z)
• High-resource Language-specific Training for Multilingual Neural Machine Translation [109.31892935605192]
  We propose the multilingual translation model with the high-resource language-specific training (HLT-MT) to alleviate the negative interference. Specifically, we first train the multilingual model only with the high-resource pairs and select the language-specific modules at the top of the decoder. HLT-MT is further trained on all available corpora to transfer knowledge from high-resource languages to low-resource languages.
  arXiv  Detail & Related papers  (2022-07-11T14:33:13Z)
• Distributionally Robust Multilingual Machine Translation [94.51866646879337]
  We propose a new learning objective for Multilingual neural machine translation (MNMT) based on distributionally robust optimization. We show how to practically optimize this objective for large translation corpora using an iterated best response scheme. Our method consistently outperforms strong baseline methods in terms of average and per-language performance under both many-to-one and one-to-many translation settings.
  arXiv  Detail & Related papers  (2021-09-09T03:48:35Z)
• Cross-lingual Transferring of Pre-trained Contextualized Language Models [73.97131976850424]
  We propose a novel cross-lingual model transferring framework for PrLMs: TreLM. To handle the symbol order and sequence length differences between languages, we propose an intermediate TRILayer" structure. We show the proposed framework significantly outperforms language models trained from scratch with limited data in both performance and efficiency.
  arXiv  Detail & Related papers  (2021-07-27T06:51:13Z)

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.