The proliferation of conspiracy theories and hateful messages on social media poses significant challenges for content moderation and public discourse. Despite their societal impact, existing datasets for automated conspiracy detection remain limited in scope and language coverage. We present a multilingual dataset of conspiracy content on Telegram comprising 5750 messages across English, Dutch, Italian, Spanish and Portuguese from 87 channels documented as disseminating conspiracist and extremist content. Domain experts annotated messages for conspiracist tone, population replacement conspiracy theories, vaccine conspiracies, and hate speech. We extensively report on difficulties and caveats when creating and annotating this type of dataset. We establish classification baselines by evaluating six models in zero-shot fashion and fine-tuning three encoder models, achieving F1 scores up to 0.800 for conspiracist tone, 0.846 for PRCT, 0.843 for vaccine-related conspiracy theories, and 0.734 for hate speech. Inter-annotator agreement was moderate, consistent with the complexity documented in similar annotation tasks.

Disinformation has become a serious threat to the democratic stability of Western societies, with various conspiracy theories spreading from fringe spaces to mainstream media and politics. While some of these theories may seem merely absurd and harmless, others pose significant risks. Among the most dangerous are Population Replacement Conspiracy Theories (PRCTs), which promote the false narrative of a deliberate demographic substitution through immigration. Despite their disinformative nature, increasing widespread and documented connections to extremist violence and political polarization, current computational detection models primarily target COVID-19 or general conspiracy theories, lacking specialized annotated corpora and approaches for identifying PRCTs in multilingual contexts. In this work, we present the first systematic benchmark for PRCT detection in Portuguese and Italian.

This position paper critiques the dominance of detection-centered approaches in misinformation research. We argue that the prevailing paradigm treats information disorders as a content-level anomaly to be identified and suppressed, thereby obscuring the structural conditions under which different forms of information disorders emerge and resonate. Drawing on critical anthropology, we propose an alternative “clinical” model: information disorders should be understood not only as informational distortion, but as a syndrome with complex causes embedded in contexts of economic precarity, institutional distrust, and informational inequality. Treating detection as the ends rather than the means of intervention risks misguiding our efforts. Rather than positioning NLP primarily as a tool for boundary enforcement, we outline a reorientation toward structural diagnosis: diversifying data beyond WEIRD contexts, extracting socioeconomic and trust-related signals from discourse, and integrating computational outputs within interdisciplinary causal frameworks. Under this model, detection becomes a means for an epidemiology of discourse, subordinated to the broader objective of cultivating long-term epistemic resilience in our online environments.

Detecting hyperpartisan narratives and Population Replacement Conspiracy Theories (PRCT) is essential to addressing the spread of misinformation. These complex narratives pose a significant threat, as hyperpartisanship drives political polarisation and institutional distrust, while PRCTs directly motivate real-world extremist violence, making their identification critical for social cohesion and public safety. However, existing resources are scarce, predominantly English-centric, and often analyse hyperpartisanship, stance, and rhetorical bias in isolation rather than as interrelated aspects of political discourse. To bridge this gap, we introduce PartisanLens, the first multilingual dataset of 1617 hyperpartisan news headlines in Spanish, Italian, and Portuguese, annotated in multiple political discourse aspects. We first evaluate the classification performance of widely used Large Language Models (LLMs) on this dataset, establishing robust baselines for the classification of hyperpartisan and PRCT narratives. In addition, we assess the viability of using LLMs as automatic annotators for this task, analysing their ability to approximate human annotation. Results highlight both their potential and current limitations. Next, moving beyond standard judgments, we explore whether LLMs can emulate human annotation patterns by conditioning them on socio-economic and ideological profiles that simulate annotator perspectives. At last, we provide our resources and evaluation; PartisanLens supports future research on detecting partisan and conspiratorial narratives in European contexts.

Online discussions can either bridge differences through constructive dialogue or amplify divisions through destructive interactions. paper proposes a computational approach to analyze dialogical relation patterns in YouTube comments, offering a fine-grained framework for controversy detection, enabling also analysis of individual contributions. experiments demonstrate that shallow learning methods, when equipped with these theoretically-grounded features, consistently outperform more complex language models in characterizing discourse quality at both comment-pair and conversation-chain levels.studies confirm that divisive rhetorical techniques serve as strong predictors of destructive communication patterns. work advances understanding of how communicative choices shape online discourse, moving beyond engagement metrics toward nuanced examination of constructive versus destructive dialogue patterns.

Les parseurs de discours ont suscité un intérêt considérable dans les récentes applications de traitement automatique du langage naturel. Cette approche dépasse les limites traditionnelles de la phrase et peut s’étendre pour englober l’identification de relation de discours. Il existe plusieurs parseurs spécialisés dans le traitement autmatique du discours, mais ces derniers ont été principalement évalués sur des corpus anglais. Par conséquent, il n’est pas évident de bien cerner les éléments linguistiques importants sur lesquels les parseurs se basent pour classifier les relations de discours en dehors de l’anglais. Cet article évalue les performances du parseur DMRST sur le corpus RST-DT traduit en français. Nous constatons que les performances de classification des relations de discours en français sont comparables à celles obtenues pour d’autres langues. En analysant les succès et échecs de la classification des relations, nous soulignons l’impact des marqueurs de discours et des structures syntaxiques sur la précision du parseur.