# ByteSpan: Information-Driven Subword Tokenisation

arXiv: https://arxiv.org/abs/2506.18639 · HTML: https://mlsanity.com/paper/2506.18639/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 45 references that are in the map.
- First posted on arXiv 23 Jun 2025.

## Builds on (most cited first, 24 of 24)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Neural Machine Translation of Rare Words with Subword Units](https://mlsanity.com/paper/1508.07909/) (2015)
- [The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale](https://mlsanity.com/paper/2406.17557/) (2024)
- [Scaling Vision Transformers](https://mlsanity.com/paper/2106.04560/) (2021)
- [MiniCPM: Unveiling the Potential of Small Language Models with Scalable Training Strategies](https://mlsanity.com/paper/2404.06395/) (2024)
- [Subword Regularization: Improving Neural Network Translation Models with Multiple Subword Candidates](https://mlsanity.com/paper/1804.10959/) (2018)
- [BLiMP: The Benchmark of Linguistic Minimal Pairs for English](https://mlsanity.com/paper/1912.00582/) (2019)
- [How Good is Your Tokenizer? On the Monolingual Performance of Multilingual Language Models](https://mlsanity.com/paper/2012.15613/) (2020)
- [MEGABYTE: Predicting Million-byte Sequences with Multiscale Transformers](https://mlsanity.com/paper/2305.07185/) (2023)
- [Funnel-Transformer: Filtering out Sequential Redundancy for Efficient Language Processing](https://mlsanity.com/paper/2006.03236/) (2020)
- [Hierarchical Transformers Are More Efficient Language Models](https://mlsanity.com/paper/2110.13711/) (2021)
- [Understanding Warmup-Stable-Decay Learning Rates: A River Valley Loss Landscape Perspective](https://mlsanity.com/paper/2410.05192/) (2024)
- [Efficient Transformers with Dynamic Token Pooling](https://mlsanity.com/paper/2211.09761/) (2022)
- [The MiniPile Challenge for Data-Efficient Language Models](https://mlsanity.com/paper/2304.08442/) (2023)
- [Tokenization and the Noiseless Channel](https://mlsanity.com/paper/2306.16842/) (2023)
- [MYTE: Morphology-Driven Byte Encoding for Better and Fairer Multilingual Language Modeling](https://mlsanity.com/paper/2403.10691/) (2024)
- [Greed is All You Need: An Evaluation of Tokenizer Inference Methods](https://mlsanity.com/paper/2403.01289/) (2024)
- [Analyzing Cognitive Plausibility of Subword Tokenization](https://mlsanity.com/paper/2310.13348/) (2023)
- [Tokenization Falling Short: On Subword Robustness in Large Language Models](https://mlsanity.com/paper/2406.11687/) (2024)
- [Improving Tokenisation by Alternative Treatment of Spaces](https://mlsanity.com/paper/2204.04058/) (2022)
- [CompoundPiece: Evaluating and Improving Decompounding Performance of Language Models](https://mlsanity.com/paper/2305.14214/) (2023)
- [Lost in Space Marking](https://mlsanity.com/paper/2208.01561/) (2022)

## Built on by (most cited first, 1 of 1)

- [Easy to anticipate, hard to compute: boundary dependence finds the computed outputs that entropy patching misses](https://mlsanity.com/paper/2610.11790/) (2026)

## Abstract (arXiv)

Recent dynamic tokenisation methods operate directly on bytes and pool their latent representations into patches. This bears similarities to computational models of word segmentation that determine lexical boundaries using spikes in an autoregressive model's prediction error. Inspired by this connection, we explore whether grouping predictable bytes - rather than pooling their representations - can yield a useful fixed subword vocabulary. We propose a new information-driven subword tokeniser, ByteSpan, that uses an external byte-level LM during training to identify contiguous predictable byte sequences and group them into subwords. Experiments show that ByteSpan yields efficient vocabularies with higher morphological alignment scores than BPE for English. Multilingual experiments show similar compression and R'enyi efficiency for 25 languages.