# Powerful Design of Small Vision Transformer on CIFAR10

arXiv: https://arxiv.org/abs/2501.06220 · HTML: https://mlsanity.com/paper/2501.06220/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 24 references that are in the map.
- First posted on arXiv 7 Jan 2025.

## Builds on (most cited first, 19 of 19)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- [mixup: Beyond Empirical Risk Minimization](https://mlsanity.com/paper/1710.09412/) (2017)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features](https://mlsanity.com/paper/1905.04899/) (2019)
- [Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention](https://mlsanity.com/paper/2006.16236/) (2020)
- [Vision Transformers for Dense Prediction](https://mlsanity.com/paper/2103.13413/) (2021)
- [Generating Long Sequences with Sparse Transformers](https://mlsanity.com/paper/1904.10509/) (2019)
- [Transformer Feed-Forward Layers Are Key-Value Memories](https://mlsanity.com/paper/2012.14913/) (2020)
- [Random Erasing Data Augmentation](https://mlsanity.com/paper/1708.04896/) (2017)
- [DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model](https://mlsanity.com/paper/2405.04434/) (2024)
- [Deep Networks with Stochastic Depth](https://mlsanity.com/paper/1603.09382/) (2016)
- [Efficient Training of Visual Transformers with Small Datasets](https://mlsanity.com/paper/2106.03746/) (2021)
- [SiT: Self-supervised vIsion Transformer](https://mlsanity.com/paper/2104.03602/) (2021)
- [How to Train Vision Transformer on Small-scale Datasets?](https://mlsanity.com/paper/2210.07240/) (2022)
- [MCTformer+: Multi-Class Token Transformer for Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/2308.03005/) (2023)
- [94% on CIFAR-10 in 3.29 Seconds on a Single GPU](https://mlsanity.com/paper/2404.00498/) (2024)
- [Depth-Wise Convolutions in Vision Transformers for Efficient Training on Small Datasets](https://mlsanity.com/paper/2407.19394/) (2024)

## Built on by (most cited first, 1 of 1)

- [Exploring Kolmogorov-Arnold Network Expansions in Vision Transformers for Mitigating Catastrophic Forgetting in Continual Learning](https://mlsanity.com/paper/2507.04020/) (2025)

## Abstract (arXiv)

Vision Transformers (ViTs) have demonstrated remarkable success on large-scale datasets, but their performance on smaller datasets often falls short of convolutional neural networks (CNNs). This paper explores the design and optimization of Tiny ViTs for small datasets, using CIFAR-10 as a benchmark. We systematically evaluate the impact of data augmentation, patch token initialization, low-rank compression, and multi-class token strategies on model performance. Our experiments reveal that low-rank compression of queries in Multi-Head Latent Attention (MLA) incurs minimal performance loss, indicating redundancy in ViTs. Additionally, introducing multiple CLS tokens improves global representation capacity, boosting accuracy. These findings provide a comprehensive framework for optimizing Tiny ViTs, offering practical insights for efficient and effective designs. Code is available at https://github.com/erow/PoorViTs.