@inproceedings{rosenblatt-etal-2022-critical,
title = "Critical Perspectives: A Benchmark Revealing Pitfalls in {P}erspective{API}",
author = "Rosenblatt, Lucas and
Piedras, Lorena and
Wilkins, Julia",
editor = "Biester, Laura and
Demszky, Dorottya and
Jin, Zhijing and
Sachan, Mrinmaya and
Tetreault, Joel and
Wilson, Steven and
Xiao, Lu and
Zhao, Jieyu",
booktitle = "Proceedings of the Second Workshop on NLP for Positive Impact (NLP4PI)",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates (Hybrid)",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.nlp4pi-1.2",
doi = "10.18653/v1/2022.nlp4pi-1.2",
pages = "15--24",
abstract = "Detecting {``}toxic{''} language in internet content is a pressing social and technical challenge. In this work, we focus on Perspective API from Jigsaw, a state-of-the-art tool that promises to score the {``}toxicity{''} of text, with a recent model update that claims impressive results (Lees et al., 2022). We seek to challenge certain normative claims about toxic language by proposing a new benchmark, Selected Adversarial SemanticS, or SASS. We evaluate Perspective on SASS, and compare to low-effort alternatives, like zero-shot and few-shot GPT-3 prompt models, in binary classification settings. We find that Perspective exhibits troubling shortcomings across a number of our toxicity categories. SASS provides a new tool for evaluating performance on previously undetected toxic language that avoids common normative pitfalls. Our work leads us to emphasize the importance of questioning assumptions made by tools already in deployment for toxicity detection in order to anticipate and prevent disparate harms.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="rosenblatt-etal-2022-critical">
<titleInfo>
<title>Critical Perspectives: A Benchmark Revealing Pitfalls in PerspectiveAPI</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lucas</namePart>
<namePart type="family">Rosenblatt</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lorena</namePart>
<namePart type="family">Piedras</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Julia</namePart>
<namePart type="family">Wilkins</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Second Workshop on NLP for Positive Impact (NLP4PI)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Laura</namePart>
<namePart type="family">Biester</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dorottya</namePart>
<namePart type="family">Demszky</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhijing</namePart>
<namePart type="family">Jin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mrinmaya</namePart>
<namePart type="family">Sachan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joel</namePart>
<namePart type="family">Tetreault</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Steven</namePart>
<namePart type="family">Wilson</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lu</namePart>
<namePart type="family">Xiao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jieyu</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Abu Dhabi, United Arab Emirates (Hybrid)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Detecting “toxic” language in internet content is a pressing social and technical challenge. In this work, we focus on Perspective API from Jigsaw, a state-of-the-art tool that promises to score the “toxicity” of text, with a recent model update that claims impressive results (Lees et al., 2022). We seek to challenge certain normative claims about toxic language by proposing a new benchmark, Selected Adversarial SemanticS, or SASS. We evaluate Perspective on SASS, and compare to low-effort alternatives, like zero-shot and few-shot GPT-3 prompt models, in binary classification settings. We find that Perspective exhibits troubling shortcomings across a number of our toxicity categories. SASS provides a new tool for evaluating performance on previously undetected toxic language that avoids common normative pitfalls. Our work leads us to emphasize the importance of questioning assumptions made by tools already in deployment for toxicity detection in order to anticipate and prevent disparate harms.</abstract>
<identifier type="citekey">rosenblatt-etal-2022-critical</identifier>
<identifier type="doi">10.18653/v1/2022.nlp4pi-1.2</identifier>
<location>
<url>https://aclanthology.org/2022.nlp4pi-1.2</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>15</start>
<end>24</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Critical Perspectives: A Benchmark Revealing Pitfalls in PerspectiveAPI
%A Rosenblatt, Lucas
%A Piedras, Lorena
%A Wilkins, Julia
%Y Biester, Laura
%Y Demszky, Dorottya
%Y Jin, Zhijing
%Y Sachan, Mrinmaya
%Y Tetreault, Joel
%Y Wilson, Steven
%Y Xiao, Lu
%Y Zhao, Jieyu
%S Proceedings of the Second Workshop on NLP for Positive Impact (NLP4PI)
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates (Hybrid)
%F rosenblatt-etal-2022-critical
%X Detecting “toxic” language in internet content is a pressing social and technical challenge. In this work, we focus on Perspective API from Jigsaw, a state-of-the-art tool that promises to score the “toxicity” of text, with a recent model update that claims impressive results (Lees et al., 2022). We seek to challenge certain normative claims about toxic language by proposing a new benchmark, Selected Adversarial SemanticS, or SASS. We evaluate Perspective on SASS, and compare to low-effort alternatives, like zero-shot and few-shot GPT-3 prompt models, in binary classification settings. We find that Perspective exhibits troubling shortcomings across a number of our toxicity categories. SASS provides a new tool for evaluating performance on previously undetected toxic language that avoids common normative pitfalls. Our work leads us to emphasize the importance of questioning assumptions made by tools already in deployment for toxicity detection in order to anticipate and prevent disparate harms.
%R 10.18653/v1/2022.nlp4pi-1.2
%U https://aclanthology.org/2022.nlp4pi-1.2
%U https://doi.org/10.18653/v1/2022.nlp4pi-1.2
%P 15-24
Markdown (Informal)
[Critical Perspectives: A Benchmark Revealing Pitfalls in PerspectiveAPI](https://aclanthology.org/2022.nlp4pi-1.2) (Rosenblatt et al., NLP4PI 2022)
ACL