# VariViT: A Vision Transformer for Variable Image Sizes

arXiv: https://arxiv.org/abs/2602.14615 · HTML: https://mlsanity.com/paper/2602.14615/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 30 references that are in the map.
- First posted on arXiv 16 Feb 2026.

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

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [SGDR: Stochastic Gradient Descent with Warm Restarts](https://mlsanity.com/paper/1608.03983/) (2016)
- [TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation](https://mlsanity.com/paper/2102.04306/) (2021)
- [Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour](https://mlsanity.com/paper/1706.02677/) (2017)
- [Self-Attention with Relative Position Representations](https://mlsanity.com/paper/1803.02155/) (2018)
- [The RSNA-ASNR-MICCAI BraTS 2021 Benchmark on Brain Tumor Segmentation and Radiogenomic Classification](https://mlsanity.com/paper/2107.02314/) (2021)
- [Pix2Struct: Screenshot Parsing as Pretraining for Visual Language Understanding](https://mlsanity.com/paper/2210.03347/) (2022)
- [Transformers in Medical Imaging: A Survey](https://mlsanity.com/paper/2201.09873/) (2022)
- [Patch n' Pack: NaViT, a Vision Transformer for any Aspect Ratio and Resolution](https://mlsanity.com/paper/2307.06304/) (2023)
- [Rethinking and Improving Relative Position Encoding for Vision Transformer](https://mlsanity.com/paper/2107.14222/) (2021)
- [FlexiViT: One Model for All Patch Sizes](https://mlsanity.com/paper/2212.08013/) (2022)
- [UTNet: A Hybrid Transformer Architecture for Medical Image Segmentation](https://mlsanity.com/paper/2107.00781/) (2021)
- [The University of California San Francisco Preoperative Diffuse Glioma MRI (UCSF-PDGM) Dataset](https://mlsanity.com/paper/2109.00356/) (2021)
- [Global-Local Transformer for Brain Age Estimation](https://mlsanity.com/paper/2109.01663/) (2021)
- [COVID-VIT: Classification of COVID-19 from CT chest images based on vision transformer models](https://mlsanity.com/paper/2107.01682/) (2021)
- [Super Vision Transformer](https://mlsanity.com/paper/2205.11397/) (2022)
- [ViT-AE++: Improving Vision Transformer Autoencoder for Self-supervised Medical Image Representations](https://mlsanity.com/paper/2301.07382/) (2023)

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

- [Applications of Neural Cellular Automata: State of the Art, Challenges and Opportunities](https://mlsanity.com/paper/2609.24595/) (2026)

## Abstract (arXiv)

Vision Transformers (ViTs) have emerged as the state-of-the-art architecture in representation learning, leveraging self-attention mechanisms to excel in various tasks. ViTs split images into fixed-size patches, constraining them to a predefined size and necessitating pre-processing steps like resizing, padding, or cropping. This poses challenges in medical imaging, particularly with irregularly shaped structures like tumors. A fixed bounding box crop size produces input images with highly variable foreground-to-background ratios. Resizing medical images can degrade information and introduce artefacts, impacting diagnosis. Hence, tailoring variable-sized crops to regions of interest can enhance feature representation capabilities. Moreover, large images are computationally expensive, and smaller sizes risk information loss, presenting a computation-accuracy tradeoff. We propose VariViT, an improved ViT model crafted to handle variable image sizes while maintaining a consistent patch size. VariViT employs a novel positional embedding resizing scheme for a variable number of patches. We also implement a new batching strategy within VariViT to reduce computational complexity, resulting in faster training and inference times. In our evaluations on two 3D brain MRI datasets, VariViT surpasses vanilla ViTs and ResNet in glioma genotype prediction and brain tumor classification. It achieves F1-scores of 75.5% and 76.3%, respectively, learning more discriminative features. Our proposed batching strategy reduces computation time by up to 30% compared to conventional architectures. These findings underscore the efficacy of VariViT in image representation learning. Our code can be found here: https://github.com/Aswathi-Varma/varivit