@inproceedings{bandyopadhyay-etal-2025-enhancing,
title = "Enhancing Training Data Quality through Influence Scores for Generalizable Classification: A Case Study on Sexism Detection",
author = "Bandyopadhyay, Rabiraj and
Assenmacher, Dennis and
Alonso-Moral, Jose Maria and
Wagner, Claudia",
editor = "Inui, Kentaro and
Sakti, Sakriani and
Wang, Haofen and
Wong, Derek F. and
Bhattacharyya, Pushpak and
Banerjee, Biplab and
Ekbal, Asif and
Chakraborty, Tanmoy and
Singh, Dhirendra Pratap",
booktitle = "Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics",
month = dec,
year = "2025",
address = "Mumbai, India",
publisher = "The Asian Federation of Natural Language Processing and The Association for Computational Linguistics",
url = "https://aclanthology.org/2025.ijcnlp-long.77/",
pages = "1382--1403",
ISBN = "979-8-89176-298-5",
abstract = "The quality of training data is crucial for the performance of supervised machine learning models. In particular, poor annotation quality and spurious correlations between labels and features in text dataset can significantly degrade model generalization. This problem is especially pronounced in harmful language detection, where prior studies have revealed major deficiencies in existing datasets. In this work, we design and test data selection methods based on learnability measures to improve dataset quality. Using a sexism dataset with counterfactuals designed to avoid spurious correlations, we show that pruning with EL2N and PVI scores can lead to significant performance increases and outperforms submodular and random selection. Our analysis reveals that in presence of label imbalance models rely on dataset shortcuts; especially easy-to-classify sexist instances and hard-to-classify non-sexist instances contain shortcuts. Pruning these instances leads to performances increases. Pruning hard-to-classify instances is in general a promising strategy as well when shortcuts are not present."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="bandyopadhyay-etal-2025-enhancing">
<titleInfo>
<title>Enhancing Training Data Quality through Influence Scores for Generalizable Classification: A Case Study on Sexism Detection</title>
</titleInfo>
<name type="personal">
<namePart type="given">Rabiraj</namePart>
<namePart type="family">Bandyopadhyay</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dennis</namePart>
<namePart type="family">Assenmacher</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jose</namePart>
<namePart type="given">Maria</namePart>
<namePart type="family">Alonso-Moral</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Claudia</namePart>
<namePart type="family">Wagner</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kentaro</namePart>
<namePart type="family">Inui</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sakriani</namePart>
<namePart type="family">Sakti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haofen</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Derek</namePart>
<namePart type="given">F</namePart>
<namePart type="family">Wong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pushpak</namePart>
<namePart type="family">Bhattacharyya</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Biplab</namePart>
<namePart type="family">Banerjee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Asif</namePart>
<namePart type="family">Ekbal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tanmoy</namePart>
<namePart type="family">Chakraborty</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dhirendra</namePart>
<namePart type="given">Pratap</namePart>
<namePart type="family">Singh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>The Asian Federation of Natural Language Processing and The Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Mumbai, India</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-298-5</identifier>
</relatedItem>
<abstract>The quality of training data is crucial for the performance of supervised machine learning models. In particular, poor annotation quality and spurious correlations between labels and features in text dataset can significantly degrade model generalization. This problem is especially pronounced in harmful language detection, where prior studies have revealed major deficiencies in existing datasets. In this work, we design and test data selection methods based on learnability measures to improve dataset quality. Using a sexism dataset with counterfactuals designed to avoid spurious correlations, we show that pruning with EL2N and PVI scores can lead to significant performance increases and outperforms submodular and random selection. Our analysis reveals that in presence of label imbalance models rely on dataset shortcuts; especially easy-to-classify sexist instances and hard-to-classify non-sexist instances contain shortcuts. Pruning these instances leads to performances increases. Pruning hard-to-classify instances is in general a promising strategy as well when shortcuts are not present.</abstract>
<identifier type="citekey">bandyopadhyay-etal-2025-enhancing</identifier>
<location>
<url>https://aclanthology.org/2025.ijcnlp-long.77/</url>
</location>
<part>
<date>2025-12</date>
<extent unit="page">
<start>1382</start>
<end>1403</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Enhancing Training Data Quality through Influence Scores for Generalizable Classification: A Case Study on Sexism Detection
%A Bandyopadhyay, Rabiraj
%A Assenmacher, Dennis
%A Alonso-Moral, Jose Maria
%A Wagner, Claudia
%Y Inui, Kentaro
%Y Sakti, Sakriani
%Y Wang, Haofen
%Y Wong, Derek F.
%Y Bhattacharyya, Pushpak
%Y Banerjee, Biplab
%Y Ekbal, Asif
%Y Chakraborty, Tanmoy
%Y Singh, Dhirendra Pratap
%S Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics
%D 2025
%8 December
%I The Asian Federation of Natural Language Processing and The Association for Computational Linguistics
%C Mumbai, India
%@ 979-8-89176-298-5
%F bandyopadhyay-etal-2025-enhancing
%X The quality of training data is crucial for the performance of supervised machine learning models. In particular, poor annotation quality and spurious correlations between labels and features in text dataset can significantly degrade model generalization. This problem is especially pronounced in harmful language detection, where prior studies have revealed major deficiencies in existing datasets. In this work, we design and test data selection methods based on learnability measures to improve dataset quality. Using a sexism dataset with counterfactuals designed to avoid spurious correlations, we show that pruning with EL2N and PVI scores can lead to significant performance increases and outperforms submodular and random selection. Our analysis reveals that in presence of label imbalance models rely on dataset shortcuts; especially easy-to-classify sexist instances and hard-to-classify non-sexist instances contain shortcuts. Pruning these instances leads to performances increases. Pruning hard-to-classify instances is in general a promising strategy as well when shortcuts are not present.
%U https://aclanthology.org/2025.ijcnlp-long.77/
%P 1382-1403
Markdown (Informal)
[Enhancing Training Data Quality through Influence Scores for Generalizable Classification: A Case Study on Sexism Detection](https://aclanthology.org/2025.ijcnlp-long.77/) (Bandyopadhyay et al., IJCNLP-AACL 2025)
ACL