# TokenFormer: Rethinking Transformer Scaling with Tokenized Model Parameters

arXiv: https://arxiv.org/abs/2410.23168 · HTML: https://mlsanity.com/paper/2410.23168/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Builds on 35 of its 49 references that are in the map.
- First posted on arXiv 30 Oct 2024.

## Builds on (most cited first, 30 of 35)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Layer Normalization](https://mlsanity.com/paper/1607.06450/) (2016)
- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [OPT: Open Pre-trained Transformer Language Models](https://mlsanity.com/paper/2205.01068/) (2022)
- [Gaussian Error Linear Units (GELUs)](https://mlsanity.com/paper/1606.08415/) (2016)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [The Pile: An 800GB Dataset of Diverse Text for Language Modeling](https://mlsanity.com/paper/2101.00027/) (2020)
- [Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling](https://mlsanity.com/paper/2304.01373/) (2023)
- [Decision Transformer: Reinforcement Learning via Sequence Modeling](https://mlsanity.com/paper/2106.01345/) (2021)
- [Root Mean Square Layer Normalization](https://mlsanity.com/paper/1910.07467/) (2019)
- [Transformer Feed-Forward Layers Are Key-Value Memories](https://mlsanity.com/paper/2012.14913/) (2020)
- [RWKV: Reinventing RNNs for the Transformer Era](https://mlsanity.com/paper/2305.13048/) (2023)
- [Efficient Transformers: A Survey](https://mlsanity.com/paper/2009.06732/) (2020)
- [OFA: Unifying Architectures, Tasks, and Modalities Through a Simple Sequence-to-Sequence Learning Framework](https://mlsanity.com/paper/2202.03052/) (2022)
- [GPT-NeoX-20B: An Open-Source Autoregressive Language Model](https://mlsanity.com/paper/2204.06745/) (2022)
- [Graph Transformer Networks](https://mlsanity.com/paper/1911.06455/) (2019)
- [Hungry Hungry Hippos: Towards Language Modeling with State Space Models](https://mlsanity.com/paper/2212.14052/) (2022)
- [Net2Net: Accelerating Learning via Knowledge Transfer](https://mlsanity.com/paper/1511.05641/) (2015)
- [DSVT: Dynamic Sparse Voxel Transformer with Rotated Sets](https://mlsanity.com/paper/2301.06051/) (2023)
- [Augmenting Self-attention with Persistent Memory](https://mlsanity.com/paper/1907.01470/) (2019)
- [Learning to Grow Pretrained Models for Efficient Transformer Training](https://mlsanity.com/paper/2303.00980/) (2023)

## Built on by (most cited first, 9 of 9)

- [UFO: A Unified Approach to Fine-grained Visual Perception via Open-ended Language Interface](https://mlsanity.com/paper/2503.01342/) (2025)
- [Depth-Recurrent Attention Mixtures: Giving Latent Reasoning the Attention it Deserves](https://mlsanity.com/paper/2601.21582/) (2026)
- [Improving Progressive Generation with Decomposable Flow Matching](https://mlsanity.com/paper/2506.19839/) (2025)
- [Grow Up and Merge: Scaling Strategies for Efficient Language Adaptation](https://mlsanity.com/paper/2512.10772/) (2025)
- [T2S: Tokenized Skill Scaling for Lifelong Imitation Learning](https://mlsanity.com/paper/2508.01167/) (2025)
- [Flash Multi-Head Feed-Forward Network](https://mlsanity.com/paper/2512.06989/) (2025)
- [Key-Value Means: Transformers with Expandable Block-Recurrent Compressed Memory](https://mlsanity.com/paper/2605.09877/) (2026)
- [Millions of States: Designing a Scalable MoE Architecture with RWKV-7 Meta-learner](https://mlsanity.com/paper/2504.08247/) (2025)
- [Nexusformer: Nonlinear Attention Expansion for Stable and Inheritable Transformer Scaling](https://mlsanity.com/paper/2604.19147/) (2026)

## Abstract (arXiv)

Transformers have become the predominant architecture in foundation models due to their excellent performance across various domains. However, the substantial cost of scaling these models remains a significant concern. This problem arises primarily from their dependence on a fixed number of parameters within linear projections. When architectural modifications (e.g., channel dimensions) are introduced, the entire model typically requires retraining from scratch. As model sizes continue growing, this strategy results in increasingly high computational costs and becomes unsustainable. To overcome this problem, we introduce TokenFormer, a natively scalable architecture that leverages the attention mechanism not only for computations among input tokens but also for interactions between tokens and model parameters, thereby enhancing architectural flexibility. By treating model parameters as tokens, we replace all the linear projections in Transformers with our token-parameter attention layer, where input tokens act as queries and model parameters as keys and values. This reformulation allows for progressive and efficient scaling without necessitating retraining from scratch. Our model scales from 124M to 1.4B parameters by incrementally adding new key-value parameter pairs, achieving performance comparable to Transformers trained from scratch while greatly reducing training costs. Code and models are available at https://github.com/Haiyang-W/TokenFormer.