Belief Is All You Need: Signed Belief Graph Neural Networks for Topic Modeling in Conspiratorial Discourse

Researchers developed a Signed Belief Graph Neural Network (SiBeGNN) to model conspiratorial discourse in Singapore-based Telegram groups, identifying seven topic clusters where conspiracy content appears across everyday discussions rather than isolated echo chambers. The method outperformed standard clustering approaches, achieving a cDBI of 8.38 versus 13.60–67.27, with 88% inter-rater agreement in cluster interpretation.

Belief Is All You Need: Signed Belief Graph Neural Networks for Topic Modeling in Conspiratorial Discourse https://aclanthology.org/2026.nlpcss-1.20.pdf Soorya Ram Shimgekar /people/soorya-ram-shimgekar/ , Abhay Goyal /people/abhay-goyal/unverified/ , Roy Ka-Wei Lee /people/roy-ka-wei-lee/ , Koustuv Saha /people/koustuv-saha/ , Pi Zonooz /people/pi-zonooz/unverified/ , Edson C Tandoc Jr /people/edson-c-tandoc-jr/unverified/ , Navin Kumar /people/navin-kumar/unverified/ Abstract Conspiratorial discourse is increasingly present in online communication, yet how it is organized across discussion topics remains unclear. We analyze Singapore-based Telegram groups to examine how conspiratorial content appears within everyday conversations rather than isolated echo chambers. To better capture the structure of such discussions, we propose a two-stage framework for topic modeling tailored to conspiratorial posts. First, a RoBERTa-large classifier identifies conspiratorial messages F1 = 0.866 using 2,000 expert-annotated examples. We then construct a graph where connections reflect textual similarity and conspiratorial stance. This graph is modeled using a Signed Belief Graph Neural Network SiBeGNN , which learns message embeddings that distinguish conspiratorial from non-conspiratorial content. We apply hierarchical clustering on these embeddings to perform topic modeling over 553,648 Telegram messages, producing seven topic clusters: General Legal Topics, Medical Concerns, Media Discussions, Banking and Finance, Contradictions in Authority, Group Moderation, and General Discussions. Our method substantially outperforms standard embedding-based clustering approaches cDBI = 8.38 vs. 13.60–67.27 , with manual evaluation showing 88% inter-rater agreement in cluster interpretation. The results show that conspiratorial content appears across multiple everyday topics, including finance, law, and daily life, rather than forming isolated thematic communities.- Anthology ID: - 2026.nlpcss-1.20 - Volume: Proceedings of the Seventh Workshop on Natural Language Processing and Computational Social Science /volumes/2026.nlpcss-1/ - Month: - July - Year: - 2026 - Address: - San Diego - Editors: Dallas Card /people/dallas-card/ , Anjalie Field /people/anjalie-field/ , Katherine Keith /people/katherine-keith/ , Julia Mendelsohn /people/julia-mendelsohn/ - Venues: NLP+CSS /venues/nlpcss/ | WS /venues/ws/ - SIG: - Publisher: - Association for Computational Linguistics - Note: - Pages: - 341–355 - Language: - URL: https://aclanthology.org/2026.nlpcss-1.20/ https://aclanthology.org/2026.nlpcss-1.20/ - DOI: - Cite ACL : - Soorya Ram Shimgekar, Abhay Goyal, Roy Ka-Wei Lee, Koustuv Saha, Pi Zonooz, Edson C Tandoc Jr, and Navin Kumar. 2026. Belief Is All You Need: Signed Belief Graph Neural Networks for Topic Modeling in Conspiratorial Discourse https://aclanthology.org/2026.nlpcss-1.20/ . In Proceedings of the Seventh Workshop on Natural Language Processing and Computational Social Science , pages 341–355, San Diego. Association for Computational Linguistics. - Cite Informal : Belief Is All You Need: Signed Belief Graph Neural Networks for Topic Modeling in Conspiratorial Discourse https://aclanthology.org/2026.nlpcss-1.20/ Shimgekar et al., NLP+CSS 2026 - PDF: https://aclanthology.org/2026.nlpcss-1.20.pdf https://aclanthology.org/2026.nlpcss-1.20.pdf