@inproceedings{toughrai-etal-2026-makes,
title = "What Makes User Reactions Useful for Fake News Detection?",
author = "Toughrai, Yassine and
Langlois, David and
Smaili",
editor = "Zhang, Yue and
Xu, Ruifeng and
Purwarianti, Ayu and
Rong, Tong and
Suzuki, Jun",
booktitle = "Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the {A}sia-Pacific Chapter of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
month = nov,
year = "2026",
address = "Hengqin, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.aacl-long.192/",
pages = "4282--4297",
ISBN = "979-8-89176-439-2",
abstract = "Fake news detection is often treated as a content-centered task, yet social media posts are embedded in user reactions that express skepticism, endorsement, correction, ridicule, and affective responses. This paper asks what makes user reactions useful for fake news detection. We study performance according to different dimensions: text only, emotions in comments, and number of available comments. We study this question across Fakeddit and Weibo20 under both standard and temporal splits (in a temporal split, most recent posts are included in the testing split). Robustness of results is measured across two encoder families (BERT-base, RoBERTa-base) and two English emotion taxonomies (dair-ai/emotion, GoEmotions). Across these configurations, comment-only representations outperform post-only baselines in most evaluation regimes, though this gain depends on dataset, split, and number of available comments. Emotion-fine-tuned comment encoders produce a temporal-shift-specific significant benefit compared to non-temporal splits.\par We compare post-only representations, generic comment representations, emotion-fine-tuned comment representations, and post{--}reaction fusion models, while varying comment availability, showing that a minimum number of comments is necessary to improve detection of fakes. In addition, we show that fine-tuning comment or reaction-encoders on relevant auxiliary tasks such as emotions has a significant impact on the distribution of embeddings of fake and real posts in their representation space.\par In light of these findings, we believe that user responses{---}and their emotional structure for example{---}should be taken into account when detecting fake news or compiling datasets for this purpose."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="toughrai-etal-2026-makes">
<titleInfo>
<title>What Makes User Reactions Useful for Fake News Detection?</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yassine</namePart>
<namePart type="family">Toughrai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Langlois</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name>
<namePart>Smaili</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruifeng</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ayu</namePart>
<namePart type="family">Purwarianti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tong</namePart>
<namePart type="family">Rong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Suzuki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Hengqin, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-439-2</identifier>
</relatedItem>
<abstract>Fake news detection is often treated as a content-centered task, yet social media posts are embedded in user reactions that express skepticism, endorsement, correction, ridicule, and affective responses. This paper asks what makes user reactions useful for fake news detection. We study performance according to different dimensions: text only, emotions in comments, and number of available comments. We study this question across Fakeddit and Weibo20 under both standard and temporal splits (in a temporal split, most recent posts are included in the testing split). Robustness of results is measured across two encoder families (BERT-base, RoBERTa-base) and two English emotion taxonomies (dair-ai/emotion, GoEmotions). Across these configurations, comment-only representations outperform post-only baselines in most evaluation regimes, though this gain depends on dataset, split, and number of available comments. Emotion-fine-tuned comment encoders produce a temporal-shift-specific significant benefit compared to non-temporal splits.\par We compare post-only representations, generic comment representations, emotion-fine-tuned comment representations, and post–reaction fusion models, while varying comment availability, showing that a minimum number of comments is necessary to improve detection of fakes. In addition, we show that fine-tuning comment or reaction-encoders on relevant auxiliary tasks such as emotions has a significant impact on the distribution of embeddings of fake and real posts in their representation space.\par In light of these findings, we believe that user responses—and their emotional structure for example—should be taken into account when detecting fake news or compiling datasets for this purpose.</abstract>
<identifier type="citekey">toughrai-etal-2026-makes</identifier>
<location>
<url>https://aclanthology.org/2026.aacl-long.192/</url>
</location>
<part>
<date>2026-11</date>
<extent unit="page">
<start>4282</start>
<end>4297</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T What Makes User Reactions Useful for Fake News Detection?
%A Toughrai, Yassine
%A Langlois, David
%Y Zhang, Yue
%Y Xu, Ruifeng
%Y Purwarianti, Ayu
%Y Rong, Tong
%Y Suzuki, Jun
%A Smaili
%S Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 November
%I Association for Computational Linguistics
%C Hengqin, China
%@ 979-8-89176-439-2
%F toughrai-etal-2026-makes
%X Fake news detection is often treated as a content-centered task, yet social media posts are embedded in user reactions that express skepticism, endorsement, correction, ridicule, and affective responses. This paper asks what makes user reactions useful for fake news detection. We study performance according to different dimensions: text only, emotions in comments, and number of available comments. We study this question across Fakeddit and Weibo20 under both standard and temporal splits (in a temporal split, most recent posts are included in the testing split). Robustness of results is measured across two encoder families (BERT-base, RoBERTa-base) and two English emotion taxonomies (dair-ai/emotion, GoEmotions). Across these configurations, comment-only representations outperform post-only baselines in most evaluation regimes, though this gain depends on dataset, split, and number of available comments. Emotion-fine-tuned comment encoders produce a temporal-shift-specific significant benefit compared to non-temporal splits.\par We compare post-only representations, generic comment representations, emotion-fine-tuned comment representations, and post–reaction fusion models, while varying comment availability, showing that a minimum number of comments is necessary to improve detection of fakes. In addition, we show that fine-tuning comment or reaction-encoders on relevant auxiliary tasks such as emotions has a significant impact on the distribution of embeddings of fake and real posts in their representation space.\par In light of these findings, we believe that user responses—and their emotional structure for example—should be taken into account when detecting fake news or compiling datasets for this purpose.
%U https://aclanthology.org/2026.aacl-long.192/
%P 4282-4297
Markdown (Informal)
[What Makes User Reactions Useful for Fake News Detection?](https://aclanthology.org/2026.aacl-long.192/) (Toughrai et al., AACL-IJCNLP 2026)
ACL
- Yassine Toughrai, David Langlois, and Smaili. 2026. What Makes User Reactions Useful for Fake News Detection?. In Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers), pages 4282–4297, Hengqin, China. Association for Computational Linguistics.