# Sanity checks for patch visualisation in prototype-based image classification

Source: [Sanity checks for patch visualisation in prototype-based image classification](<https://www.sota2.com/research/paper/68fa71451e209131cea50483>)

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

Data updated: 2025-10-28T13:57:02.657Z

Authors: Romain Xu-Darme, Georges Qu\\'enot, Zakaria Chihani, Marie-Christine Rousset

Published: 2023-10-25T08:13:02.000Z

## Abstract

In this work, we perform an analysis of the visualisation methods implemented in ProtoPNet and ProtoTree, two self-explaining visual classifiers based on prototypes. We show that such methods do not correctly identify the regions of interest inside of the images, and therefore do not reflect the model behaviour, which can create a false sense of bias in the model. We also demonstrate quantitatively that this issue can be mitigated by using other saliency methods that provide more faithful image patches.

## 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.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2311.16120>)

[PDF](<https://arxiv.org/pdf/2311.16120.pdf>)