Personalized Interpretability -- Interactive Alignment of Prototypical Parts Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Michalski, Tomasz, Wróbel, Adam, Bontempelli, Andrea, Luśtyk, Jakub, Kniejski, Mikolaj, Teso, Stefano, Passerini, Andrea, Zieliński, Bartosz, Rymarczyk, Dawid
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908395464818688
author Michalski, Tomasz
Wróbel, Adam
Bontempelli, Andrea
Luśtyk, Jakub
Kniejski, Mikolaj
Teso, Stefano
Passerini, Andrea
Zieliński, Bartosz
Rymarczyk, Dawid
author_facet Michalski, Tomasz
Wróbel, Adam
Bontempelli, Andrea
Luśtyk, Jakub
Kniejski, Mikolaj
Teso, Stefano
Passerini, Andrea
Zieliński, Bartosz
Rymarczyk, Dawid
contents Concept-based interpretable neural networks have gained significant attention due to their intuitive and easy-to-understand explanations based on case-based reasoning, such as "this bird looks like those sparrows". However, a major limitation is that these explanations may not always be comprehensible to users due to concept inconsistency, where multiple visual features are inappropriately mixed (e.g., a bird's head and wings treated as a single concept). This inconsistency breaks the alignment between model reasoning and human understanding. Furthermore, users have specific preferences for how concepts should look, yet current approaches provide no mechanism for incorporating their feedback. To address these issues, we introduce YoursProtoP, a novel interactive strategy that enables the personalization of prototypical parts - the visual concepts used by the model - according to user needs. By incorporating user supervision, YoursProtoP adapts and splits concepts used for both prediction and explanation to better match the user's preferences and understanding. Through experiments on both the synthetic FunnyBirds dataset and a real-world scenario using the CUB, CARS, and PETS datasets in a comprehensive user study, we demonstrate the effectiveness of YoursProtoP in achieving concept consistency without compromising the accuracy of the model.
format Preprint
id arxiv_https___arxiv_org_abs_2506_05533
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Personalized Interpretability -- Interactive Alignment of Prototypical Parts Networks
Michalski, Tomasz
Wróbel, Adam
Bontempelli, Andrea
Luśtyk, Jakub
Kniejski, Mikolaj
Teso, Stefano
Passerini, Andrea
Zieliński, Bartosz
Rymarczyk, Dawid
Computer Vision and Pattern Recognition
Human-Computer Interaction
Concept-based interpretable neural networks have gained significant attention due to their intuitive and easy-to-understand explanations based on case-based reasoning, such as "this bird looks like those sparrows". However, a major limitation is that these explanations may not always be comprehensible to users due to concept inconsistency, where multiple visual features are inappropriately mixed (e.g., a bird's head and wings treated as a single concept). This inconsistency breaks the alignment between model reasoning and human understanding. Furthermore, users have specific preferences for how concepts should look, yet current approaches provide no mechanism for incorporating their feedback. To address these issues, we introduce YoursProtoP, a novel interactive strategy that enables the personalization of prototypical parts - the visual concepts used by the model - according to user needs. By incorporating user supervision, YoursProtoP adapts and splits concepts used for both prediction and explanation to better match the user's preferences and understanding. Through experiments on both the synthetic FunnyBirds dataset and a real-world scenario using the CUB, CARS, and PETS datasets in a comprehensive user study, we demonstrate the effectiveness of YoursProtoP in achieving concept consistency without compromising the accuracy of the model.
title Personalized Interpretability -- Interactive Alignment of Prototypical Parts Networks
topic Computer Vision and Pattern Recognition
Human-Computer Interaction
url https://arxiv.org/abs/2506.05533