@inproceedings{henrique-luz-de-araujo-roth-2022-checking,
title = "Checking {H}ate{C}heck: a cross-functional analysis of behaviour-aware learning for hate speech detection",
author = "Henrique Luz de Araujo, Pedro and
Roth, Benjamin",
editor = "Shavrina, Tatiana and
Mikhailov, Vladislav and
Malykh, Valentin and
Artemova, Ekaterina and
Serikov, Oleg and
Protasov, Vitaly",
booktitle = "Proceedings of NLP Power! The First Workshop on Efficient Benchmarking in NLP",
month = may,
year = "2022",
address = "Dublin, Ireland",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.nlppower-1.8",
doi = "10.18653/v1/2022.nlppower-1.8",
pages = "75--83",
abstract = "Behavioural testing{---}verifying system capabilities by validating human-designed input-output pairs{---}is an alternative evaluation method of natural language processing systems proposed to address the shortcomings of the standard approach: computing metrics on held-out data. While behavioural tests capture human prior knowledge and insights, there has been little exploration on how to leverage them for model training and development. With this in mind, we explore behaviour-aware learning by examining several fine-tuning schemes using HateCheck, a suite of functional tests for hate speech detection systems. To address potential pitfalls of training on data originally intended for evaluation, we train and evaluate models on different configurations of HateCheck by holding out categories of test cases, which enables us to estimate performance on potentially overlooked system properties. The fine-tuning procedure led to improvements in the classification accuracy of held-out functionalities and identity groups, suggesting that models can potentially generalise to overlooked functionalities. However, performance on held-out functionality classes and i.i.d. hate speech detection data decreased, which indicates that generalisation occurs mostly across functionalities from the same class and that the procedure led to overfitting to the HateCheck data distribution.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="henrique-luz-de-araujo-roth-2022-checking">
<titleInfo>
<title>Checking HateCheck: a cross-functional analysis of behaviour-aware learning for hate speech detection</title>
</titleInfo>
<name type="personal">
<namePart type="given">Pedro</namePart>
<namePart type="family">Henrique Luz de Araujo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Benjamin</namePart>
<namePart type="family">Roth</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of NLP Power! The First Workshop on Efficient Benchmarking in NLP</title>
</titleInfo>
<name type="personal">
<namePart type="given">Tatiana</namePart>
<namePart type="family">Shavrina</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vladislav</namePart>
<namePart type="family">Mikhailov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Valentin</namePart>
<namePart type="family">Malykh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Artemova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Oleg</namePart>
<namePart type="family">Serikov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vitaly</namePart>
<namePart type="family">Protasov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Dublin, Ireland</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Behavioural testing—verifying system capabilities by validating human-designed input-output pairs—is an alternative evaluation method of natural language processing systems proposed to address the shortcomings of the standard approach: computing metrics on held-out data. While behavioural tests capture human prior knowledge and insights, there has been little exploration on how to leverage them for model training and development. With this in mind, we explore behaviour-aware learning by examining several fine-tuning schemes using HateCheck, a suite of functional tests for hate speech detection systems. To address potential pitfalls of training on data originally intended for evaluation, we train and evaluate models on different configurations of HateCheck by holding out categories of test cases, which enables us to estimate performance on potentially overlooked system properties. The fine-tuning procedure led to improvements in the classification accuracy of held-out functionalities and identity groups, suggesting that models can potentially generalise to overlooked functionalities. However, performance on held-out functionality classes and i.i.d. hate speech detection data decreased, which indicates that generalisation occurs mostly across functionalities from the same class and that the procedure led to overfitting to the HateCheck data distribution.</abstract>
<identifier type="citekey">henrique-luz-de-araujo-roth-2022-checking</identifier>
<identifier type="doi">10.18653/v1/2022.nlppower-1.8</identifier>
<location>
<url>https://aclanthology.org/2022.nlppower-1.8</url>
</location>
<part>
<date>2022-05</date>
<extent unit="page">
<start>75</start>
<end>83</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Checking HateCheck: a cross-functional analysis of behaviour-aware learning for hate speech detection
%A Henrique Luz de Araujo, Pedro
%A Roth, Benjamin
%Y Shavrina, Tatiana
%Y Mikhailov, Vladislav
%Y Malykh, Valentin
%Y Artemova, Ekaterina
%Y Serikov, Oleg
%Y Protasov, Vitaly
%S Proceedings of NLP Power! The First Workshop on Efficient Benchmarking in NLP
%D 2022
%8 May
%I Association for Computational Linguistics
%C Dublin, Ireland
%F henrique-luz-de-araujo-roth-2022-checking
%X Behavioural testing—verifying system capabilities by validating human-designed input-output pairs—is an alternative evaluation method of natural language processing systems proposed to address the shortcomings of the standard approach: computing metrics on held-out data. While behavioural tests capture human prior knowledge and insights, there has been little exploration on how to leverage them for model training and development. With this in mind, we explore behaviour-aware learning by examining several fine-tuning schemes using HateCheck, a suite of functional tests for hate speech detection systems. To address potential pitfalls of training on data originally intended for evaluation, we train and evaluate models on different configurations of HateCheck by holding out categories of test cases, which enables us to estimate performance on potentially overlooked system properties. The fine-tuning procedure led to improvements in the classification accuracy of held-out functionalities and identity groups, suggesting that models can potentially generalise to overlooked functionalities. However, performance on held-out functionality classes and i.i.d. hate speech detection data decreased, which indicates that generalisation occurs mostly across functionalities from the same class and that the procedure led to overfitting to the HateCheck data distribution.
%R 10.18653/v1/2022.nlppower-1.8
%U https://aclanthology.org/2022.nlppower-1.8
%U https://doi.org/10.18653/v1/2022.nlppower-1.8
%P 75-83
Markdown (Informal)
[Checking HateCheck: a cross-functional analysis of behaviour-aware learning for hate speech detection](https://aclanthology.org/2022.nlppower-1.8) (Henrique Luz de Araujo & Roth, nlppower 2022)
ACL