# Disentangling Visual Transformers: Patch-level Interpretability for Image Classification

Source: [Disentangling Visual Transformers: Patch-level Interpretability for Image Classification](<https://www.sota2.com/research/paper/68fa78301e209131cea77109>)

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

Data updated: 2025-10-26T04:33:25.608Z

Authors: Guillaume Jeanneret, Lo\\"ic Simon, Fr\\'ed\\'eric Jurie

Published: 2025-02-24T14:30:29.000Z

## Abstract

Visual transformers have achieved remarkable performance in image classification tasks, but this performance gain has come at the cost of interpretability. One of the main obstacles to the interpretation of transformers is the self-attention mechanism, which mixes visual information across the whole image in a complex way. In this paper, we propose Hindered Transformer (HiT), a novel interpretable by design architecture inspired by visual transformers. Our proposed architecture rethinks the design of transformers to better disentangle patch influences at the classification stage. Ultimately, HiT can be interpreted as a linear combination of patch-level information. We show that the advantages of our approach in terms of explicability come with a reasonable trade-off in performance, making it an attractive alternative for applications where interpretability is paramount.

## 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/2502.17196>)

[PDF](<https://arxiv.org/pdf/2502.17196.pdf>)