# Model-Aware Tokenizer Transfer

arXiv: https://arxiv.org/abs/2510.21954 · HTML: https://mlsanity.com/paper/2510.21954/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 47 of its 59 references that are in the map.
- First posted on arXiv 24 Oct 2025.

## Builds on (most cited first, 30 of 47)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [Qwen Technical Report](https://mlsanity.com/paper/2309.16609/) (2023)
- [Gemma 2: Improving Open Language Models at a Practical Size](https://mlsanity.com/paper/2408.00118/) (2024)
- [Decoupled Weight Decay Regularization](https://mlsanity.com/paper/1711.05101/) (2017)
- [Gemma 3 Technical Report](https://mlsanity.com/paper/2503.19786/) (2025)
- [Enriching Word Vectors with Subword Information](https://mlsanity.com/paper/1607.04606/) (2016)
- [Transformer Feed-Forward Layers Are Key-Value Memories](https://mlsanity.com/paper/2012.14913/) (2020)
- [No Language Left Behind: Scaling Human-Centered Machine Translation](https://mlsanity.com/paper/2207.04672/) (2022)
- [Gemma: Open Models Based on Gemini Research and Technology](https://mlsanity.com/paper/2403.08295/) (2024)
- [GradNorm: Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networks](https://mlsanity.com/paper/1711.02257/) (2017)
- [OLMo: Accelerating the Science of Language Models](https://mlsanity.com/paper/2402.00838/) (2024)
- [Knowledge Neurons in Pretrained Transformers](https://mlsanity.com/paper/2104.08696/) (2021)
- [On the Cross-lingual Transferability of Monolingual Representations](https://mlsanity.com/paper/1910.11856/) (2019)
- [The FLORES-101 Evaluation Benchmark for Low-Resource and Multilingual Machine Translation](https://mlsanity.com/paper/2106.03193/) (2021)
- [The Falcon Series of Open Language Models](https://mlsanity.com/paper/2311.16867/) (2023)
- [XL-Sum: Large-Scale Multilingual Abstractive Summarization for 44 Languages](https://mlsanity.com/paper/2106.13822/) (2021)
- [The Belebele Benchmark: a Parallel Reading Comprehension Dataset in 122 Language Variants](https://mlsanity.com/paper/2308.16884/) (2023)
- [Global MMLU: Understanding and Addressing Cultural and Linguistic Biases in Multilingual Evaluation](https://mlsanity.com/paper/2412.03304/) (2024)
- [Byte Latent Transformer: Patches Scale Better Than Tokens](https://mlsanity.com/paper/2412.09871/) (2024)
- [Aya Expanse: Combining Research Breakthroughs for a New Multilingual Frontier](https://mlsanity.com/paper/2412.04261/) (2024)
- [Tokenizer Choice For LLM Training: Negligible or Crucial?](https://mlsanity.com/paper/2310.08754/) (2023)
- [Language Contamination Helps Explain the Cross-lingual Capabilities of English Pretrained Models](https://mlsanity.com/paper/2204.08110/) (2022)
- [WECHSEL: Effective initialization of subword embeddings for cross-lingual transfer of monolingual language models](https://mlsanity.com/paper/2112.06598/) (2021)
- [XLM-V: Overcoming the Vocabulary Bottleneck in Multilingual Masked Language Models](https://mlsanity.com/paper/2301.10472/) (2023)
- [Dynamic Chunking for End-to-End Hierarchical Sequence Modeling](https://mlsanity.com/paper/2507.07955/) (2025)
- [The FLoRes Evaluation Datasets for Low-Resource Machine Translation: Nepali-English and Sinhala-English](https://mlsanity.com/paper/1902.01382/) (2019)
- [As Good as New. How to Successfully Recycle English GPT-2 to Make Models for Other Languages](https://mlsanity.com/paper/2012.05628/) (2020)
- [Understanding Factual Recall in Transformers via Associative Memories](https://mlsanity.com/paper/2412.06538/) (2024)

## Built on by (most cited first, 6 of 6)

- [Bolmo: Byteifying the Next Generation of Language Models](https://mlsanity.com/paper/2512.15586/) (2025)
- [Advancing Polish Language Modeling through Tokenizer Optimization in the Bielik v3 7B and 11B Series](https://mlsanity.com/paper/2604.10799/) (2026)
- [Writing-System-Level Tokenizer Adaptation for Byte-Level BPE](https://mlsanity.com/paper/2608.00582/) (2026)
- [KazByte: Adapting Qwen models to Kazakh via Byte-level Adapter](https://mlsanity.com/paper/2603.27859/) (2026)
- [Disentangling Language Modeling and Boundaries](https://mlsanity.com/paper/2608.03599/) (2026)
- [Lend Me Your Eyes: Instruction-Aware Text Embeddings via Attention Relay](https://mlsanity.com/paper/2610.05564/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) are trained to support an increasing number of languages, yet their predefined tokenizers remain a bottleneck for adapting models to lower-resource or distinct-script languages. Existing tokenizer transfer methods typically rely on semantic heuristics to initialize new embeddings, ignoring higher-layer model dynamics and limiting transfer quality. We propose Model-Aware Tokenizer Transfer (MATT), a method that incorporates model internals into the tokenizer transfer process. MATT introduces an Attention Influence Modeling (AIM) objective that distills inter-token communication patterns from a source model into a target model with a new tokenizer, providing an efficient warm-up before standard language modeling. Unlike approaches that focus solely on embedding similarity, MATT leverages attention behavior to guide embedding initialization and adaptation. Experiments across diverse linguistic settings show that MATT recovers a large fraction of the original model's performance within a few GPU hours, outperforming heuristic baselines. These results demonstrate that incorporating model-level signals offers a practical and effective path toward robust tokenizer transfer in multilingual LLMs.