# Disentangling Language Modeling and Boundaries

arXiv: https://arxiv.org/abs/2608.03599 · HTML: https://mlsanity.com/paper/2608.03599/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 13 references that are in the map.
- First posted on arXiv 4 Aug 2026.

## Builds on (most cited first, 11 of 11)

- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [ByT5: Towards a token-free future with pre-trained byte-to-byte models](https://mlsanity.com/paper/2105.13626/) (2021)
- [Byte Latent Transformer: Patches Scale Better Than Tokens](https://mlsanity.com/paper/2412.09871/) (2024)
- [Dynamic Chunking for End-to-End Hierarchical Sequence Modeling](https://mlsanity.com/paper/2507.07955/) (2025)
- [Towards Cross-Tokenizer Distillation: the Universal Logit Distillation Loss for LLMs](https://mlsanity.com/paper/2402.12030/) (2024)
- [Universal Cross-Tokenizer Distillation via Approximate Likelihood Matching](https://mlsanity.com/paper/2503.20083/) (2025)
- [Bolmo: Byteifying the Next Generation of Language Models](https://mlsanity.com/paper/2512.15586/) (2025)
- [Cross-Tokenizer LLM Distillation through a Byte-Level Interface](https://mlsanity.com/paper/2604.07466/) (2026)
- [Token Distillation: Attention-aware Input Embeddings For New Tokens](https://mlsanity.com/paper/2505.20133/) (2025)
- [Model-Aware Tokenizer Transfer](https://mlsanity.com/paper/2510.21954/) (2025)
- [Recovered in Translation: Efficient Pipeline for Automated Translation of Benchmarks and Datasets](https://mlsanity.com/paper/2602.22207/) (2026)

## Abstract (arXiv)

Byte-level language models are usually argued for on the grounds of robustness, multilingual fairness, and character-level skills. We point to a different, structural advantage: because they read and write bytes, any two of them share an output space, so knowledge transfer between them is exact and independent of how either was originally tokenized. We hypothesize that the two distributions a byte-level model produces, one over the next byte, one over where its patch boundaries fall, can be disentangled and changed almost independently. A model could absorb a teacher's capability while keeping its own boundaries, or change how it places those boundaries while keeping its capabilities. We lay out the two experiments that would settle the hypothesis, alongside preliminary measurements of the properties they rest on. We argue that the community should move toward a byte-level interface as a shared standard: if the hypothesis holds, then once byte-level models are the norm, transferring capabilities and reshaping boundaries between them become cheap and routine, free of the per-model tokenizer that blocks them today.