mlsanityData to 14 Oct 2026

Rethinking Tokenization: Crafting Better Tokenizers for Large Language Models

arXiv 2403.00417 ↗First posted 1 Mar 2024cs.CLcited by 8

On the map: LLM reasoning → Reasoning benchmarks and evaluation

Abstract (arXiv)

Tokenization significantly influences language models(LMs)' performance. This paper traces the evolution of tokenizers from word-level to subword-level, analyzing how they balance tokens and types to enhance model adaptability while controlling complexity. Despite subword tokenizers like Byte Pair Encoding (BPE) overcoming many word tokenizer limitations, they encounter difficulties in handling non-Latin languages and depend heavily on extensive training data and computational resources to grasp the nuances of multiword expressions (MWEs). This article argues that tokenizers, more than mere technical tools, should drawing inspiration from the cognitive science about human language processing. This study then introduces the "Principle of Least Effort" from cognitive science, that humans naturally seek to reduce cognitive effort, and discusses the benefits of this principle for tokenizer development. Based on this principle, the paper proposes that the Less-is-Better (LiB) model could be a new approach for LLM tokenizer. The LiB model can autonomously learn an integrated vocabulary consisting of subwords, words, and MWEs, which effectively reduces both the numbers of tokens and types. Comparative evaluations show that the LiB tokenizer outperforms existing word and BPE tokenizers, presenting an innovative method for tokenizer development, and hinting at the possibility of future cognitive science-based tokenizers being more efficient.

Key facts. Cited by 8 papers in the map (data through 14 Oct 2026). Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning. Builds on 14 of its 38 references that are in the map. First posted on arXiv 1 Mar 2024.

Builds on

14 of 38 references.

References · 14

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2022Training language models to follow instructions with human feedbackcited by 15,971
  3. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  4. 2013Efficient Estimation of Word Representations in Vector Spacecited by 4,375
  5. 2015Neural Machine Translation of Rare Words with Subword Unitscited by 2,918
  6. 2019XLNet: Generalized Autoregressive Pretraining for Language Understandingcited by 2,811
  7. 2019ALBERT: A Lite BERT for Self-supervised Learning of Language Representationscited by 2,790
  8. 2018SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processingcited by 1,989
  9. 2018Subword Regularization: Improving Neural Network Translation Models with Multiple Subword Candidatescited by 639
  10. 2023Large Language Models Are Zero-Shot Time Series Forecasterscited by 500
  11. 2019ERNIE: Enhanced Representation through Knowledge Integrationcited by 302
  12. 2023Language Modeling Is Compressioncited by 225
2 more
  1. 2023Randomized Positional Encodings Boost Length Generalization of Transformerscited by 117
  2. 2023Analyzing Cognitive Plausibility of Subword Tokenizationcited by 30

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 5; 2025: 1; 2026: 2.

8 papers.

  1. 2024Data Mixture Inference: What do BPE Tokenizers Reveal about their Training Data?cited by 13
  2. 2024Problematic Tokens: Tokenizer Bias in Large Language Modelscited by 8
  3. 2024Elucidating the design space of language models for image generationcited by 6
  4. 2024CharED: Character-wise Ensemble Decoding for Large Language Modelscited by 3
  5. 2026Proxy Compression for Language Modelingcited by 3
  6. 2025KG-CF: Knowledge Graph Completion with Context Filtering under the Guidance of Large Language Modelscited by 1
  7. 2024Batching BPE Tokenization Mergescited by 0
  8. 2026Zonkey: A Hierarchical Diffusion Language Model with Differentiable Tokenization and Probabilistic Attentioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025The Art of Breaking Words: Rethinking Multilingual Tokenizer Design
  2. 2026In-Place Tokenizer Expansion for Pre-trained LLMs
  3. 2026TokEval: A Tokenizer Evaluation Suite
  4. 2026Why do Large Language Models Fail in Low-resource Translation? Unraveling the Token Dynamics of Large Language Models for Machine Translation
  5. 2026A Family of LLMs Liberated from Static Vocabularies
  6. 2025Length-MAX Tokenizer for Language Models
  7. 2026How Tokenization Limits Phonological Knowledge Representation in Language Models and How to Improve Them
  8. 2025The Token Tax: Systematic Bias in Multilingual Tokenization