# Incorporating Domain Knowledge into Materials Tokenization

arXiv: https://arxiv.org/abs/2506.11115 · HTML: https://mlsanity.com/paper/2506.11115/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 54 references that are in the map.
- First posted on arXiv 9 Jun 2025.

## Builds on (most cited first, 22 of 22)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Neural Machine Translation of Rare Words with Subword Units](https://mlsanity.com/paper/1508.07909/) (2015)
- [Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation](https://mlsanity.com/paper/1609.08144/) (2016)
- [SciBERT: A Pretrained Language Model for Scientific Text](https://mlsanity.com/paper/1903.10676/) (2019)
- [Domain-Specific Language Model Pretraining for Biomedical Natural Language Processing](https://mlsanity.com/paper/2007.15779/) (2020)
- [How Good is Your Tokenizer? On the Monolingual Performance of Multilingual Language Models](https://mlsanity.com/paper/2012.15613/) (2020)
- [Byte Pair Encoding is Suboptimal for Language Model Pretraining](https://mlsanity.com/paper/2004.03720/) (2020)
- [Construction of the Literature Graph in Semantic Scholar](https://mlsanity.com/paper/1805.02262/) (2018)
- [Tokenization counts: the impact of tokenization on arithmetic in frontier LLMs](https://mlsanity.com/paper/2402.14903/) (2024)
- [XLM-V: Overcoming the Vocabulary Bottleneck in Multilingual Masked Language Models](https://mlsanity.com/paper/2301.10472/) (2023)
- [Superbizarre Is Not Superb: Derivational Morphology Improves BERT's Interpretation of Complex Words](https://mlsanity.com/paper/2101.00403/) (2021)
- [The Materials Science Procedural Text Corpus: Annotating Materials Synthesis Procedures with Shallow Semantic Structures](https://mlsanity.com/paper/1905.06939/) (2019)
- [MatSci-NLP: Evaluating Scientific Language Models on Materials Science Language Tasks Using Text-to-Schema Modeling](https://mlsanity.com/paper/2305.08264/) (2023)
- [The SOFC-Exp Corpus and Neural Approaches to Information Extraction in the Materials Science Domain](https://mlsanity.com/paper/2006.03039/) (2020)
- [Tokenization Matters! Degrading Large Language Models through Challenging Their Tokenization](https://mlsanity.com/paper/2405.17067/) (2024)
- [BPE Gets Picky: Efficient Vocabulary Refinement During Tokenizer Training](https://mlsanity.com/paper/2409.04599/) (2024)
- [The SIGMORPHON 2022 Shared Task on Morpheme Segmentation](https://mlsanity.com/paper/2206.07615/) (2022)
- [HoneyBee: Progressive Instruction Finetuning of Large Language Models for Materials Science](https://mlsanity.com/paper/2310.08511/) (2023)
- [Looking Through Glass: Knowledge Discovery from Materials Science Literature using Natural Language Processing](https://mlsanity.com/paper/2101.01508/) (2021)
- [How Vocabulary Sharing Facilitates Multilingualism in LLaMA?](https://mlsanity.com/paper/2311.09071/) (2023)
- [Biomedical Language Models are Robust to Sub-optimal Tokenization](https://mlsanity.com/paper/2306.17649/) (2023)
- [MELT: Materials-aware Continued Pre-training for Language Model Adaptation to Materials Science](https://mlsanity.com/paper/2410.15126/) (2024)

## Built on by (most cited first, 2 of 2)

- [REPAIR: Resolving Long-Tail Confusion in Scientific Retrievers via Fact-Verified Iterative Refinement](https://mlsanity.com/paper/2609.18262/) (2026)
- [Exploring LLMs and RAG for Plausible and Explainable Material Prediction of Vehicle Components](https://mlsanity.com/paper/2609.18437/) (2026)

## Abstract (arXiv)

While language models are increasingly utilized in materials science, typical models rely on frequency-centric tokenization methods originally developed for natural language processing. However, these methods frequently produce excessive fragmentation and semantic loss, failing to maintain the structural and semantic integrity of material concepts. To address this issue, we propose MATTER, a novel tokenization approach that integrates material knowledge into tokenization. Based on MatDetector trained on our materials knowledge base and a re-ranking method prioritizing material concepts in token merging, MATTER maintains the structural integrity of identified material concepts and prevents fragmentation during tokenization, ensuring their semantic meaning remains intact. The experimental results demonstrate that MATTER outperforms existing tokenization methods, achieving an average performance gain of 4% and 2% in the generation and classification tasks, respectively. These results underscore the importance of domain knowledge for tokenization strategies in scientific text processing. Our code is available at https://github.com/yerimoh/MATTER