@inproceedings{kirk-etal-2021-memes,
title = "Memes in the Wild: Assessing the Generalizability of the Hateful Memes Challenge Dataset",
author = "Kirk, Hannah and
Jun, Yennie and
Rauba, Paulius and
Wachtel, Gal and
Li, Ruining and
Bai, Xingjian and
Broestl, Noah and
Doff-Sotta, Martin and
Shtedritski, Aleksandar and
Asano, Yuki M",
booktitle = "Proceedings of the 5th Workshop on Online Abuse and Harms (WOAH 2021)",
month = aug,
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.woah-1.4",
doi = "10.18653/v1/2021.woah-1.4",
pages = "26--35",
abstract = "Hateful memes pose a unique challenge for current machine learning systems because their message is derived from both text- and visual-modalities. To this effect, Facebook released the Hateful Memes Challenge, a dataset of memes with pre-extracted text captions, but it is unclear whether these synthetic examples generalize to {`}memes in the wild{'}. In this paper, we collect hateful and non-hateful memes from Pinterest to evaluate out-of-sample performance on models pre-trained on the Facebook dataset. We find that {`}memes in the wild{'} differ in two key aspects: 1) Captions must be extracted via OCR, injecting noise and diminishing performance of multimodal models, and 2) Memes are more diverse than {`}traditional memes{'}, including screenshots of conversations or text on a plain background. This paper thus serves as a reality-check for the current benchmark of hateful meme detection and its applicability for detecting real world hate.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kirk-etal-2021-memes">
<titleInfo>
<title>Memes in the Wild: Assessing the Generalizability of the Hateful Memes Challenge Dataset</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hannah</namePart>
<namePart type="family">Kirk</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yennie</namePart>
<namePart type="family">Jun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Paulius</namePart>
<namePart type="family">Rauba</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gal</namePart>
<namePart type="family">Wachtel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruining</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xingjian</namePart>
<namePart type="family">Bai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Noah</namePart>
<namePart type="family">Broestl</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Martin</namePart>
<namePart type="family">Doff-Sotta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aleksandar</namePart>
<namePart type="family">Shtedritski</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuki</namePart>
<namePart type="given">M</namePart>
<namePart type="family">Asano</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 5th Workshop on Online Abuse and Harms (WOAH 2021)</title>
</titleInfo>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Online</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Hateful memes pose a unique challenge for current machine learning systems because their message is derived from both text- and visual-modalities. To this effect, Facebook released the Hateful Memes Challenge, a dataset of memes with pre-extracted text captions, but it is unclear whether these synthetic examples generalize to ‘memes in the wild’. In this paper, we collect hateful and non-hateful memes from Pinterest to evaluate out-of-sample performance on models pre-trained on the Facebook dataset. We find that ‘memes in the wild’ differ in two key aspects: 1) Captions must be extracted via OCR, injecting noise and diminishing performance of multimodal models, and 2) Memes are more diverse than ‘traditional memes’, including screenshots of conversations or text on a plain background. This paper thus serves as a reality-check for the current benchmark of hateful meme detection and its applicability for detecting real world hate.</abstract>
<identifier type="citekey">kirk-etal-2021-memes</identifier>
<identifier type="doi">10.18653/v1/2021.woah-1.4</identifier>
<location>
<url>https://aclanthology.org/2021.woah-1.4</url>
</location>
<part>
<date>2021-08</date>
<extent unit="page">
<start>26</start>
<end>35</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Memes in the Wild: Assessing the Generalizability of the Hateful Memes Challenge Dataset
%A Kirk, Hannah
%A Jun, Yennie
%A Rauba, Paulius
%A Wachtel, Gal
%A Li, Ruining
%A Bai, Xingjian
%A Broestl, Noah
%A Doff-Sotta, Martin
%A Shtedritski, Aleksandar
%A Asano, Yuki M.
%S Proceedings of the 5th Workshop on Online Abuse and Harms (WOAH 2021)
%D 2021
%8 August
%I Association for Computational Linguistics
%C Online
%F kirk-etal-2021-memes
%X Hateful memes pose a unique challenge for current machine learning systems because their message is derived from both text- and visual-modalities. To this effect, Facebook released the Hateful Memes Challenge, a dataset of memes with pre-extracted text captions, but it is unclear whether these synthetic examples generalize to ‘memes in the wild’. In this paper, we collect hateful and non-hateful memes from Pinterest to evaluate out-of-sample performance on models pre-trained on the Facebook dataset. We find that ‘memes in the wild’ differ in two key aspects: 1) Captions must be extracted via OCR, injecting noise and diminishing performance of multimodal models, and 2) Memes are more diverse than ‘traditional memes’, including screenshots of conversations or text on a plain background. This paper thus serves as a reality-check for the current benchmark of hateful meme detection and its applicability for detecting real world hate.
%R 10.18653/v1/2021.woah-1.4
%U https://aclanthology.org/2021.woah-1.4
%U https://doi.org/10.18653/v1/2021.woah-1.4
%P 26-35
Markdown (Informal)
[Memes in the Wild: Assessing the Generalizability of the Hateful Memes Challenge Dataset](https://aclanthology.org/2021.woah-1.4) (Kirk et al., WOAH 2021)
ACL
- Hannah Kirk, Yennie Jun, Paulius Rauba, Gal Wachtel, Ruining Li, Xingjian Bai, Noah Broestl, Martin Doff-Sotta, Aleksandar Shtedritski, and Yuki M Asano. 2021. Memes in the Wild: Assessing the Generalizability of the Hateful Memes Challenge Dataset. In Proceedings of the 5th Workshop on Online Abuse and Harms (WOAH 2021), pages 26–35, Online. Association for Computational Linguistics.