# Over-Tokenized Transformer: Vocabulary is Generally Worth Scaling

arXiv: https://arxiv.org/abs/2501.16975 · HTML: https://mlsanity.com/paper/2501.16975/

## Key facts

- Cited by 44 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 26 references that are in the map.
- First posted on arXiv 28 Jan 2025.

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

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Neural Machine Translation of Rare Words with Subword Units](https://mlsanity.com/paper/1508.07909/) (2015)
- [PIQA: Reasoning about Physical Commonsense in Natural Language](https://mlsanity.com/paper/1911.11641/) (2019)
- [PyTorch FSDP: Experiences on Scaling Fully Sharded Data Parallel](https://mlsanity.com/paper/2304.11277/) (2023)
- [Subword Regularization: Improving Neural Network Translation Models with Multiple Subword Candidates](https://mlsanity.com/paper/1804.10959/) (2018)
- [ByT5: Towards a token-free future with pre-trained byte-to-byte models](https://mlsanity.com/paper/2105.13626/) (2021)
- [Better & Faster Large Language Models via Multi-token Prediction](https://mlsanity.com/paper/2404.19737/) (2024)
- [Blockwise Parallel Decoding for Deep Autoregressive Models](https://mlsanity.com/paper/1811.03115/) (2018)
- [OLMoE: Open Mixture-of-Experts Language Models](https://mlsanity.com/paper/2409.02060/) (2024)
- [CANINE: Pre-training an Efficient Tokenization-Free Encoder for Language Representation](https://mlsanity.com/paper/2103.06874/) (2021)
- [ProphetNet: Predicting Future N-gram for Sequence-to-Sequence Pre-training](https://mlsanity.com/paper/2001.04063/) (2020)
- [MEGABYTE: Predicting Million-byte Sequences with Multiscale Transformers](https://mlsanity.com/paper/2305.07185/) (2023)
- [Infini-gram: Scaling Unbounded n-gram Language Models to a Trillion Tokens](https://mlsanity.com/paper/2401.17377/) (2024)
- [Byte Latent Transformer: Patches Scale Better Than Tokens](https://mlsanity.com/paper/2412.09871/) (2024)
- [Scaling Laws with Vocabulary: Larger Models Deserve Larger Vocabularies](https://mlsanity.com/paper/2407.13623/) (2024)
- [Colossal-AI: A Unified Deep Learning System For Large-Scale Parallel Training](https://mlsanity.com/paper/2110.14883/) (2021)
- [N-Grammer: Augmenting Transformers with latent n-grams](https://mlsanity.com/paper/2207.06366/) (2022)
- [Lookup-Table Recurrent Language Models for Long Tail Speech Recognition](https://mlsanity.com/paper/2104.04552/) (2021)
- [A Frequency-aware Software Cache for Large Recommendation System Embeddings](https://mlsanity.com/paper/2208.05321/) (2022)

## Built on by (most cited first, 30 of 44)

- [Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models](https://mlsanity.com/paper/2601.07372/) (2026)
- [Dynamic Chunking for End-to-End Hierarchical Sequence Modeling](https://mlsanity.com/paper/2507.07955/) (2025)
- [SuperBPE: Space Travel for Language Models](https://mlsanity.com/paper/2503.13423/) (2025)
- [Scaling Behavior of Discrete Diffusion Language Models](https://mlsanity.com/paper/2512.10858/) (2025)
- [On the Design of Qwen3.8-Next Architecture: Evaluation, Efficiency, and Training Stability](https://mlsanity.com/paper/2608.30320/) (2026)
- [Boundless Byte Pair Encoding: Breaking the Pre-tokenization Barrier](https://mlsanity.com/paper/2504.00178/) (2025)
- [Scaling Embedding Layers in Language Models](https://mlsanity.com/paper/2502.01637/) (2025)
- [The Rise of Small Language Models in Healthcare: A Comprehensive Survey](https://mlsanity.com/paper/2504.17119/) (2025)
- [Scaling Embeddings Outperforms Scaling Experts in Language Models](https://mlsanity.com/paper/2601.21204/) (2026)
- [One Tokenizer To Rule Them All: Emergent Language Plasticity via Multilingual Tokenizers](https://mlsanity.com/paper/2506.10766/) (2025)
- [The Strawberry Problem: Emergence of Character-level Understanding in Tokenized Language Models](https://mlsanity.com/paper/2505.14172/) (2025)
- [Bolmo: Byteifying the Next Generation of Language Models](https://mlsanity.com/paper/2512.15586/) (2025)
- [Frac-Connections: Fractional Extension of Hyper-Connections](https://mlsanity.com/paper/2503.14125/) (2025)
- [Exploiting Vocabulary Frequency Imbalance in Language Model Pre-training](https://mlsanity.com/paper/2508.15390/) (2025)
- [Token Distillation: Attention-aware Input Embeddings For New Tokens](https://mlsanity.com/paper/2505.20133/) (2025)
- [XSpecMesh: Quality-Preserving Auto-Regressive Mesh Generation Acceleration via Multi-Head Speculative Decoding](https://mlsanity.com/paper/2507.23777/) (2025)
- [Stop Taking Tokenizers for Granted: They Are Core Design Decisions in Large Language Models](https://mlsanity.com/paper/2601.13260/) (2026)
- [Speculative Decoding with a Speculative Vocabulary](https://mlsanity.com/paper/2602.13836/) (2026)
- [UltraMemV2: Memory Networks Scaling to 120B Parameters with Superior Long-Context Learning](https://mlsanity.com/paper/2508.18756/) (2025)
- [DynaSpec: Context-aware Dynamic Speculative Sampling for Large-Vocabulary Language Models](https://mlsanity.com/paper/2510.13847/) (2025)
- [TokSuite: Measuring the Impact of Tokenizer Choice on Language Model Behavior](https://mlsanity.com/paper/2512.20757/) (2025)
- [Scale-Distribution Decoupling: Enabling Stable and Effective Training of Large Language Models](https://mlsanity.com/paper/2502.15499/) (2025)
- [SupraTok: Cross-Boundary Tokenization for Enhanced Language Model Performance](https://mlsanity.com/paper/2508.11857/) (2025)
- [Membership Inference Attacks on Tokenizers of Large Language Models](https://mlsanity.com/paper/2510.05699/) (2025)
- [Virtual Width Networks](https://mlsanity.com/paper/2511.11238/) (2025)
- [JTok: On Token Embedding as another Axis of Scaling Law via Joint Token Self-modulation](https://mlsanity.com/paper/2602.00800/) (2026)
- [Proxy Compression for Language Modeling](https://mlsanity.com/paper/2602.04289/) (2026)
- [LiteToken: Removing Intermediate Merge Residues From BPE Tokenizers](https://mlsanity.com/paper/2602.04706/) (2026)
- [Beyond the Last Layer: Multi-Layer Representation Fusion for Visual Tokenization](https://mlsanity.com/paper/2605.10780/) (2026)
- [Tensorizing Engram: Sharing Latents Across N-Gram Embeddings is Beneficial in LLMs](https://mlsanity.com/paper/2606.08347/) (2026)

## Abstract (arXiv)

Tokenization is a fundamental component of large language models (LLMs), yet its influence on model scaling and performance is not fully explored. In this paper, we introduce Over-Tokenized Transformers, a novel framework that decouples input and output vocabularies to improve language modeling performance. Specifically, our approach scales up input vocabularies to leverage multi-gram tokens. Through extensive experiments, we uncover a log-linear relationship between input vocabulary size and training loss, demonstrating that larger input vocabularies consistently enhance model performance, regardless of model size. Using a large input vocabulary, we achieve performance comparable to double-sized baselines with no additional cost. Our findings highlight the importance of tokenization in scaling laws and provide practical insight for tokenizer design, paving the way for more efficient and powerful LLMs.