# LEMON-ZEST: Evolution-Informed Tokenization for Efficient Protein Language Modeling

arXiv: https://arxiv.org/abs/2609.37675 · HTML: https://mlsanity.com/paper/2609.37675/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 14 of its 58 references that are in the map.
- First posted on arXiv 29 Sep 2026.

## Builds on (most cited first, 14 of 14)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Training Compute-Optimal Large Language Models](https://mlsanity.com/paper/2203.15556/) (2022)
- [Neural Machine Translation of Rare Words with Subword Units](https://mlsanity.com/paper/1508.07909/) (2015)
- [GLU Variants Improve Transformer](https://mlsanity.com/paper/2002.05202/) (2020)
- [Train Short, Test Long: Attention with Linear Biases Enables Input Length Extrapolation](https://mlsanity.com/paper/2108.12409/) (2021)
- [BPE-Dropout: Simple and Effective Subword Regularization](https://mlsanity.com/paper/1910.13267/) (2019)
- [A Closer Look at AUROC and AUPRC under Class Imbalance](https://mlsanity.com/paper/2401.06091/) (2024)
- [Ankh: Optimized Protein Language Model Unlocks General-Purpose Modelling](https://mlsanity.com/paper/2301.06568/) (2023)
- [Protein Structure Tokenization via Geometric Byte Pair Encoding](https://mlsanity.com/paper/2511.11758/) (2025)
- [PUMA: Learning a Mutation-Aware Vocabulary of Protein Units](https://mlsanity.com/paper/2503.08838/) (2025)

## Abstract (arXiv)

Protein Language Models (PLMs) have made remarkable progress following scaling laws established in natural language processing across sequence- and structure-based tasks, yet the potential of tokenization remains underexploited. Unlike human language, proteins preserve structure despite extensive sequence variation a property standard tokenization strategies fundamentally fail to capture. We introduce ZEST (Zoned Encoding of Sequence Traits), an evolution-informed vocabulary derived from conserved regions of multiple sequence alignments. ZEST allows embedding domain-level biological priors directly at the tokenization stage rather than learning them implicitly through scale. ZEST natively compresses sequences to an average token length of 4 residues, enabling our model to process 4,000 residues within a standard 1024-token context window. Building on this, we present LEMON (Layered Extraction of Molecular Ordering from Nature), a compact 200M-parameter sequence-based model for detection of remote homology between protein sequences trained on a single H100 GPU for one week. Despite its modest size, LEMON outperforms state-of-the-art models ranging from 600M to 3B parameters. Our results demonstrate that evolution-informed tokenization can substitute for massive parameter scaling, opening a new direction for efficient, biologically-grounded protein representation learning. All code, model weights, and results are publicly available under the MIT license.