Sara Tonelli

Leonardo Bottona, Nicolò Penzo, Bruno Lepri et al.

We present LLMberjack, a platform for creating multi-party conversations starting from existing debates, originally structured as reply trees. The system offers an interactive interface that visualizes discussion trees and enables users to construct coherent linearized dialogue sequences while preserving participant identity and discourse relations. It integrates optional large language model (LLM) assistance to support automatic editing of the messages and speakers' descriptions. We demonstrate the platform's utility by showing how tree visualization facilitates the creation of coherent, meaningful conversation threads and how LLM support enhances output quality while reducing human effort. The tool is open-source and designed to promote transparent and reproducible workflows to create multi-party conversations, addressing a lack of resources of this type.

Nicolò Penzo, Maryam Sajedinia, Bruno Lepri et al.

Assessing the performance of systems to classify Multi-Party Conversations (MPC) is challenging due to the interconnection between linguistic and structural characteristics of conversations. Conventional evaluation methods often overlook variances in model behavior across different levels of structural complexity on interaction graphs. In this work, we propose a methodological pipeline to investigate model performance across specific structural attributes of conversations. As a proof of concept we focus on Response Selection and Addressee Recognition tasks, to diagnose model weaknesses. To this end, we extract representative diagnostic subdatasets with a fixed number of users and a good structural variety from a large and open corpus of online MPCs. We further frame our work in terms of data minimization, avoiding the use of original usernames to preserve privacy, and propose alternatives to using original text messages. Results show that response selection relies more on the textual content of conversations, while addressee recognition requires capturing their structural dimension. Using an LLM in a zero-shot setting, we further highlight how sensitivity to prompt variations is task-dependent.

Alan Ramponi, Agnese Daffara, Sara Tonelli

We introduce Faina, the first dataset for fallacy detection that embraces multiple plausible answers and natural disagreement. Faina includes over 11K span-level annotations with overlaps across 20 fallacy types on social media posts in Italian about migration, climate change, and public health given by two expert annotators. Through an extensive annotation study that allowed discussion over multiple rounds, we minimize annotation errors whilst keeping signals of human label variation. Moreover, we devise a framework that goes beyond "single ground truth" evaluation and simultaneously accounts for multiple (equally reliable) test sets and the peculiarities of the task, i.e., partial span matches, overlaps, and the varying severity of labeling errors. Our experiments across four fallacy detection setups show that multi-task and multi-label transformer-based approaches are strong baselines across all settings. We release our data, code, and annotation guidelines to foster research on fallacy detection and human label variation more broadly.

Nicolò Penzo, Marco Guerini, Bruno Lepri et al.

Multi-Party Conversations (MPCs) are widely studied across disciplines, with social media as a primary data source due to their accessibility. However, these datasets raise privacy concerns and often reflect platform-specific properties. For example, interactions between speakers may be limited due to rigid platform structures (e.g., threads, tree-like discussions), which yield overly simplistic interaction patterns (e.g., as a consequence of ``reply-to'' links). This work explores the feasibility of generating diverse MPCs with instruction-tuned Large Language Models (LLMs) by providing deterministic constraints such as dialogue structure and participants' stance. We investigate two complementary strategies of leveraging LLMs in this context: (i.) LLMs as MPC generators, where we task the LLM to generate a whole MPC at once and (ii.) LLMs as MPC parties, where the LLM generates one turn of the conversation at a time, provided the conversation history. We next introduce an analytical framework to evaluate compliance with the constraints, content quality, and interaction complexity for both strategies. Finally, we assess the quality of obtained MPCs via human annotation and LLM-as-a-judge evaluations. We find stark differences among LLMs, with only some being able to generate high-quality MPCs. We also find that turn-by-turn generation yields better conformance to constraints and higher linguistic variability than generating MPCs in one pass. Nonetheless, our structural and qualitative evaluation indicates that both generation strategies can yield high-quality MPCs.

Ivan Srba, Olesya Razuvayevskaya, João A. Leite et al.

In the age of social media and generative AI, the ability to automatically assess the credibility of online content has become increasingly critical, complementing traditional approaches to false information detection. Credibility assessment relies on aggregating diverse credibility signals - small units of information, such as content subjectivity, bias, or a presence of persuasion techniques - into a final credibility label/score. However, current research in automatic credibility assessment and credibility signals detection remains highly fragmented, with many signals studied in isolation and lacking integration. Notably, there is a scarcity of approaches that detect and aggregate multiple credibility signals simultaneously. These challenges are further exacerbated by the absence of a comprehensive and up-to-date overview of research works that connects these research efforts under a common framework and identifies shared trends, challenges, and open problems. In this survey, we address this gap by presenting a systematic and comprehensive literature review of 175 research papers, focusing on textual credibility signals within the field of Natural Language Processing (NLP), which undergoes a rapid transformation due to advancements in Large Language Models (LLMs). While positioning the NLP research into the the broader multidisciplinary landscape, we examine both automatic credibility assessment methods as well as the detection of nine categories of credibility signals. We provide an in-depth analysis of three key categories: 1) factuality, subjectivity and bias, 2) persuasion techniques and logical fallacies, and 3) check-worthy and fact-checked claims. In addition to summarising existing methods, datasets, and tools, we outline future research direction and emerging opportunities, with particular attention to evolving challenges posed by generative AI.

Alan Ramponi, Marco Rovera, Robert Moro et al.

Retrieval of previously fact-checked claims is a well-established task, whose automation can assist professional fact-checkers in the initial steps of information verification. Previous works have mostly tackled the task monolingually, i.e., having both the input and the retrieved claims in the same language. However, especially for languages with a limited availability of fact-checks and in case of global narratives, such as pandemics, wars, or international politics, it is crucial to be able to retrieve claims across languages. In this work, we examine strategies to improve the multilingual and crosslingual performance, namely selection of negative examples (in the supervised) and re-ranking (in the unsupervised setting). We evaluate all approaches on a dataset containing posts and claims in 47 languages (283 language combinations). We observe that the best results are obtained by using LLM-based re-ranking, followed by fine-tuning with negative examples sampled using a sentence similarity-based strategy. Most importantly, we show that crosslinguality is a setup with its own unique characteristics compared to the multilingual setup.

Nicolò Penzo, Antonio Longa, Bruno Lepri et al.

Current text classification approaches usually focus on the content to be classified. Contextual aspects (both linguistic and extra-linguistic) are usually neglected, even in tasks based on online discussions. Still in many cases the multi-party and multi-turn nature of the context from which these elements are selected can be fruitfully exploited. In this work, we propose a series of experiments on a large dataset for stance detection in English, in which we evaluate the contribution of different types of contextual information, i.e. linguistic, structural and temporal, by feeding them as natural language input into a transformer-based model. We also experiment with different amounts of training data and analyse the topology of local discussion networks in a privacy-compliant way. Results show that structural information can be highly beneficial to text classification but only under certain circumstances (e.g. depending on the amount of training data and on discussion chain complexity). Indeed, we show that contextual information on smaller datasets from other classification tasks does not yield significant improvements. Our framework, based on local discussion networks, allows the integration of structural information, while minimising user profiling, thus preserving their privacy.

Elisa Leonardelli, Sara Tonelli

The online diffusion of information related to Europe and migration has been little investigated from an external point of view. However, this is a very relevant topic, especially if users have had no direct contact with Europe and its perception depends solely on information retrieved online. In this work we analyse the information circulating online about Europe and migration after retrieving a large amount of data from social media (Twitter), to gain new insights into topics, magnitude, and dynamics of their diffusion. We combine retweets and hashtags network analysis with geolocation of users, linking thus data to geography and allowing analysis from an "outside Europe" perspective, with a special focus on Africa. We also introduce a novel approach based on cross-lingual quotes, i.e. when content in a language is commented and retweeted in another language, assuming these interactions are a proxy for connections between very distant communities. Results show how the majority of online discussions occurs at a national level, especially when discussing migration. Language (English) is pivotal for information to become transnational and reach far. Transnational information flow is strongly unbalanced, with content mainly produced in Europe and amplified outside. Conversely Europe-based accounts tend to be self-referential when they discuss migration-related topics. Football is the most exported topic from Europe worldwide. Moreover, important nodes in the communities discussing migration-related topics include accounts of official institutions and international agencies, together with journalists, news, commentators and activists.

Elisa Leonardelli, Stefano Menini, Alessio Palmero Aprosio et al.

Since state-of-the-art approaches to offensive language detection rely on supervised learning, it is crucial to quickly adapt them to the continuously evolving scenario of social media. While several approaches have been proposed to tackle the problem from an algorithmic perspective, so to reduce the need for annotated data, less attention has been paid to the quality of these data. Following a trend that has emerged recently, we focus on the level of agreement among annotators while selecting data to create offensive language datasets, a task involving a high level of subjectivity. Our study comprises the creation of three novel datasets of English tweets covering different topics and having five crowd-sourced judgments each. We also present an extensive set of experiments showing that selecting training and test data according to different levels of annotators' agreement has a strong effect on classifiers performance and robustness. Our findings are further validated in cross-domain experiments and studied using a popular benchmark dataset. We show that such hard cases, where low agreement is present, are not necessarily due to poor-quality annotation and we advocate for a higher presence of ambiguous cases in future datasets, particularly in test sets, to better account for the different points of view expressed online.

Daniela Trotta, Raffaele Guarasci, Elisa Leonardelli et al.

The development of automated approaches to linguistic acceptability has been greatly fostered by the availability of the English CoLA corpus, which has also been included in the widely used GLUE benchmark. However, this kind of research for languages other than English, as well as the analysis of cross-lingual approaches, has been hindered by the lack of resources with a comparable size in other languages. We have therefore developed the ItaCoLA corpus, containing almost 10,000 sentences with acceptability judgments, which has been created following the same approach and the same steps as the English one. In this paper we describe the corpus creation, we detail its content, and we present the first experiments on this new resource. We compare in-domain and out-of-domain classification, and perform a specific evaluation of nine linguistic phenomena. We also present the first cross-lingual experiments, aimed at assessing whether multilingual transformerbased approaches can benefit from using sentences in two languages during fine-tuning.

Yi-Ling Chung, Serra Sinem Tekiroglu, Sara Tonelli et al.

Studies on online hate speech have mostly focused on the automated detection of harmful messages. Little attention has been devoted so far to the development of effective strategies to fight hate speech, in particular through the creation of counter-messages. While existing manual scrutiny and intervention strategies are time-consuming and not scalable, advances in natural language processing have the potential to provide a systematic approach to hatred management. In this paper, we introduce a novel ICT platform that NGO operators can use to monitor and analyze social media data, along with a counter-narrative suggestion tool. Our platform aims at increasing the efficiency and effectiveness of operators' activities against islamophobia. We test the platform with more than one hundred NGO operators in three countries through qualitative and quantitative evaluation. Results show that NGOs favor the platform solution with the suggestion tool, and that the time required to produce counter-narratives significantly decreases.

Stefano Menini, Alessio Palmero Aprosio, Sara Tonelli

The datasets most widely used for abusive language detection contain lists of messages, usually tweets, that have been manually judged as abusive or not by one or more annotators, with the annotation performed at message level. In this paper, we investigate what happens when the hateful content of a message is judged also based on the context, given that messages are often ambiguous and need to be interpreted in the context of occurrence. We first re-annotate part of a widely used dataset for abusive language detection in English in two conditions, i.e. with and without context. Then, we compare the performance of three classification algorithms obtained on these two types of dataset, arguing that a context-aware classification is more challenging but also more similar to a real application scenario.

Alessio Palmero Aprosio, Stefano Menini, Sara Tonelli

In order to study online hate speech, the availability of datasets containing the linguistic phenomena of interest are of crucial importance. However, when it comes to specific target groups, for example teenagers, collecting such data may be problematic due to issues with consent and privacy restrictions. Furthermore, while text-only datasets of this kind have been widely used, limitations set by image-based social media platforms like Instagram make it difficult for researchers to experiment with multimodal hate speech data. We therefore developed CREENDER, an annotation tool that has been used in school classes to create a multimodal dataset of images and abusive comments, which we make freely available under Apache 2.0 license. The corpus, with Italian comments, has been analysed from different perspectives, to investigate whether the subject of the images plays a role in triggering a comment. We find that users judge the same images in different ways, although the presence of a person in the picture increases the probability to get an offensive comment.