On the reliability of feature attribution methods for speech classification

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shen, Gaofei, Mohebbi, Hosein, Bisazza, Arianna, Alishahi, Afra, Chrupała, Grzegorz
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912387230072832
author Shen, Gaofei
Mohebbi, Hosein
Bisazza, Arianna
Alishahi, Afra
Chrupała, Grzegorz
author_facet Shen, Gaofei
Mohebbi, Hosein
Bisazza, Arianna
Alishahi, Afra
Chrupała, Grzegorz
contents As the capabilities of large-scale pre-trained models evolve, understanding the determinants of their outputs becomes more important. Feature attribution aims to reveal which parts of the input elements contribute the most to model outputs. In speech processing, the unique characteristics of the input signal make the application of feature attribution methods challenging. We study how factors such as input type and aggregation and perturbation timespan impact the reliability of standard feature attribution methods, and how these factors interact with characteristics of each classification task. We find that standard approaches to feature attribution are generally unreliable when applied to the speech domain, with the exception of word-aligned perturbation methods when applied to word-based classification tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2505_16406
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle On the reliability of feature attribution methods for speech classification
Shen, Gaofei
Mohebbi, Hosein
Bisazza, Arianna
Alishahi, Afra
Chrupała, Grzegorz
Computation and Language
Audio and Speech Processing
As the capabilities of large-scale pre-trained models evolve, understanding the determinants of their outputs becomes more important. Feature attribution aims to reveal which parts of the input elements contribute the most to model outputs. In speech processing, the unique characteristics of the input signal make the application of feature attribution methods challenging. We study how factors such as input type and aggregation and perturbation timespan impact the reliability of standard feature attribution methods, and how these factors interact with characteristics of each classification task. We find that standard approaches to feature attribution are generally unreliable when applied to the speech domain, with the exception of word-aligned perturbation methods when applied to word-based classification tasks.
title On the reliability of feature attribution methods for speech classification
topic Computation and Language
Audio and Speech Processing
url https://arxiv.org/abs/2505.16406