# Super Vision Transformer

arXiv: https://arxiv.org/abs/2205.11397 · HTML: https://mlsanity.com/paper/2205.11397/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- Builds on 50 of its 63 references that are in the map.
- First posted on arXiv 23 May 2022.

## Builds on (most cited first, 30 of 50)

- [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)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [End-to-End Object Detection with Transformers](https://mlsanity.com/paper/2005.12872/) (2020)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://mlsanity.com/paper/1704.04861/) (2017)
- [Gaussian Error Linear Units (GELUs)](https://mlsanity.com/paper/1606.08415/) (2016)
- [Deformable DETR: Deformable Transformers for End-to-End Object Detection](https://mlsanity.com/paper/2010.04159/) (2020)
- [SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers](https://mlsanity.com/paper/2105.15203/) (2021)
- [Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions](https://mlsanity.com/paper/2102.12122/) (2021)
- [SwinIR: Image Restoration Using Swin Transformer](https://mlsanity.com/paper/2108.10257/) (2021)
- [Restormer: Efficient Transformer for High-Resolution Image Restoration](https://mlsanity.com/paper/2111.09881/) (2021)
- [Is Space-Time Attention All You Need for Video Understanding?](https://mlsanity.com/paper/2102.05095/) (2021)
- [ViViT: A Video Vision Transformer](https://mlsanity.com/paper/2103.15691/) (2021)
- [Deformable Convolutional Networks](https://mlsanity.com/paper/1703.06211/) (2017)
- [Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformers](https://mlsanity.com/paper/2012.15840/) (2020)
- [Unified Perceptual Parsing for Scene Understanding](https://mlsanity.com/paper/1807.10221/) (2018)
- [ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices](https://mlsanity.com/paper/1707.01083/) (2017)
- [Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNet](https://mlsanity.com/paper/2101.11986/) (2021)
- [Transformers in Vision: A Survey](https://mlsanity.com/paper/2101.01169/) (2021)
- [Revisiting Unreasonable Effectiveness of Data in Deep Learning Era](https://mlsanity.com/paper/1707.02968/) (2017)
- [DynamicViT: Efficient Vision Transformers with Dynamic Token Sparsification](https://mlsanity.com/paper/2106.02034/) (2021)
- [CrossViT: Cross-Attention Multi-Scale Vision Transformer for Image Classification](https://mlsanity.com/paper/2103.14899/) (2021)
- [Once-for-All: Train One Network and Specialize it for Efficient Deployment](https://mlsanity.com/paper/1908.09791/) (2019)
- [Going deeper with Image Transformers](https://mlsanity.com/paper/2103.17239/) (2021)
- [Learning Efficient Convolutional Networks through Network Slimming](https://mlsanity.com/paper/1708.06519/) (2017)
- [A Survey on Visual Transformer](https://mlsanity.com/paper/2012.12556/) (2020)
- [Transformer in Transformer](https://mlsanity.com/paper/2103.00112/) (2021)
- [Twins: Revisiting the Design of Spatial Attention in Vision Transformers](https://mlsanity.com/paper/2104.13840/) (2021)
- [LeViT: a Vision Transformer in ConvNet's Clothing for Faster Inference](https://mlsanity.com/paper/2104.01136/) (2021)

## Built on by (most cited first, 13 of 13)

- [FlexiViT: One Model for All Patch Sizes](https://mlsanity.com/paper/2212.08013/) (2022)
- [DiffRate : Differentiable Compression Rate for Efficient Vision Transformers](https://mlsanity.com/paper/2305.17997/) (2023)
- [Which Tokens to Use? Investigating Token Reduction in Vision Transformers](https://mlsanity.com/paper/2308.04657/) (2023)
- [ResFormer: Scaling ViTs with Multi-Resolution Training](https://mlsanity.com/paper/2212.00776/) (2022)
- [BA-SAM: Scalable Bias-Mode Attention Mask for Segment Anything Model](https://mlsanity.com/paper/2401.02317/) (2024)
- [Agglomerative Token Clustering](https://mlsanity.com/paper/2409.11923/) (2024)
- [Win-Win: Training High-Resolution Vision Transformers from Two Windows](https://mlsanity.com/paper/2310.00632/) (2023)
- [Sparse-Tuning: Adapting Vision Transformers with Efficient Fine-tuning and Inference](https://mlsanity.com/paper/2405.14700/) (2024)
- [SMMix: Self-Motivated Image Mixing for Vision Transformers](https://mlsanity.com/paper/2212.12977/) (2022)
- [Token Compensator: Altering Inference Cost of Vision Transformer without Re-Tuning](https://mlsanity.com/paper/2408.06798/) (2024)
- [COMMA: Coordinate-aware Modulated Mamba Network for 3D Dispersed Vessel Segmentation](https://mlsanity.com/paper/2503.02332/) (2025)
- [VariViT: A Vision Transformer for Variable Image Sizes](https://mlsanity.com/paper/2602.14615/) (2026)
- [RAViT: Resolution-Adaptive Vision Transformer](https://mlsanity.com/paper/2602.24159/) (2026)

## Abstract (arXiv)

We attempt to reduce the computational costs in vision transformers (ViTs), which increase quadratically in the token number. We present a novel training paradigm that trains only one ViT model at a time, but is capable of providing improved image recognition performance with various computational costs. Here, the trained ViT model, termed super vision transformer (SuperViT), is empowered with the versatile ability to solve incoming patches of multiple sizes as well as preserve informative tokens with multiple keeping rates (the ratio of keeping tokens) to achieve good hardware efficiency for inference, given that the available hardware resources often change from time to time. Experimental results on ImageNet demonstrate that our SuperViT can considerably reduce the computational costs of ViT models with even performance increase. For example, we reduce 2x FLOPs of DeiT-S while increasing the Top-1 accuracy by 0.2% and 0.7% for 1.5x reduction. Also, our SuperViT significantly outperforms existing studies on efficient vision transformers. For example, when consuming the same amount of FLOPs, our SuperViT surpasses the recent state-of-the-art (SOTA) EViT by 1.1% when using DeiT-S as their backbones. The project of this work is made publicly available at https://github.com/lmbxmu/SuperViT.