# Interpretable Image Classification with Adaptive Prototype-based Vision Transformers

Source: [Interpretable Image Classification with Adaptive Prototype-based Vision Transformers](<https://www.sota2.com/research/paper/68fa7b491e209131cea878b4>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-04T05:02:19.156Z

Authors: Chiyu Ma, Jon Donnelly, Wenjun Liu, Soroush Vosoughi, Cynthia Rudin, Chaofan Chen

Published: 2024-10-28T04:33:28.000Z

## Abstract

We present ProtoViT, a method for interpretable image classification combining deep learning and case-based reasoning. This method classifies an image by comparing it to a set of learned prototypes, providing explanations of the form \`\`this looks like that.'' In our model, a prototype consists of \\textit{parts}, which can deform over irregular geometries to create a better comparison between images. Unlike existing models that rely on Convolutional Neural Network (CNN) backbones and spatially rigid prototypes, our model integrates Vision Transformer (ViT) backbones into prototype based models, while offering spatially deformed prototypes that not only accommodate geometric variations of objects but also provide coherent and clear prototypical feature representations with an adaptive number of prototypical parts. Our experiments show that our model can generally achieve higher performance than the existing prototype based models. Our comprehensive analyses ensure that the prototypes are consistent and the interpretations are faithful.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Image Classification on Cars (test)](<https://www.sota2.com/research/sota/image-classification-on-cars-test.md>): Image Classification; Cars (test); Accuracy: 92.4.

- [Fine-grained Image Classification on CUB-200](<https://www.sota2.com/research/sota/fine-grained-image-classification-on-cub-200.md>): Fine-grained Image Classification; CUB-200; Accuracy (All): 85.4.

- [Image Classification on CUB-200-2011 (test)](<https://www.sota2.com/research/sota/image-classification-on-cub-200-2011-test.md>): Image Classification; CUB-200-2011 (test); Top-1 Acc: 85.82.

- [Fine-grained Image Classification on Stanford Cars](<https://www.sota2.com/research/sota/fine-grained-image-classification-on-stanford-cars.md>): Fine-grained Image Classification; Stanford Cars; Accuracy: 91.8.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2410.20722>)

[PDF](<https://arxiv.org/pdf/2410.20722.pdf>)

[Implementation](<https://github.com/Henrymachiyu/ProtoViT>)