Carlo Alberto Bono

Carlo Bono, Mehmet Oğuz Mülâyim, Cinzia Cappiello et al.

Social media have the potential to provide timely information about emergency situations and sudden events. However, finding relevant information among millions of posts being posted every day can be difficult, and developing a data analysis project usually requires time and technical skills. This study presents an approach that provides flexible support for analyzing social media, particularly during emergencies. Different use cases in which social media analysis can be adopted are introduced, and the challenges of retrieving information from large sets of posts are discussed. The focus is on analyzing images and text contained in social media posts and a set of automatic data processing tools for filtering, classification, and geolocation of content with a human-in-the-loop approach to support the data analyst. Such support includes both feedback and suggestions to configure automated tools, and crowdsourcing to gather inputs from citizens. The results are validated by discussing three case studies developed within the Crowd4SDG H2020 European project.

Carlo Bono, Nick Liu, Giuseppe Russo et al.

Moderation and blocking behavior, both closely related to the mitigation of abuse and misinformation on social platforms, are fundamental mechanisms for maintaining healthy online communities. However, while centralized platforms typically employ top-down moderation, decentralized networks rely on users to self-regulate through mechanisms like blocking actions to safeguard their online experience. Given the novelty of the decentralized paradigm, addressing self-moderation is critical for understanding how community safety and user autonomy can be effectively balanced. This study examines user blocking on Bluesky, a decentralized social networking platform, providing a comprehensive analysis of over three months of user activity through the lens of blocking behaviour. We define profiles based on 86 features that describe user activity, content characteristics, and network interactions, addressing two primary questions: (1) Is the likelihood of a user being blocked inferable from their online behavior? and (2) What behavioral features are associated with an increased likelihood of being blocked? Our findings offer valuable insights and contribute with a robust analytical framework to advance research in moderation on decentralized social networks.

Geng Liu, Li Feng, Carlo Alberto Bono et al.

Recent research has highlighted that assigning specific personas to large language models (LLMs) can significantly increase harmful content generation. However, limited attention has been given to persona-driven toxicity in non-Western contexts, particularly in Chinese-based LLMs. In this paper, we perform a large-scale, cross-model analysis of refusal behavior and persona-driven toxicity amplification across four Chinese LLMs, leveraging a comprehensive dataset of over 1,400,000 generated texts. We identify significant disparities in persona-driven refusal behavior, including systematic gender differences in refusal triggering across the evaluated Chinese LLMs. Furthermore, we provide quantitative evidence of persona-driven toxicity amplification with respect to model default baselines. We show that this amplification--whose magnitude varies substantially across models--is driven by interactions across several factors, involving persona conditioning, prompting strategy, target social group, and model-specific safety mechanisms. Leveraging model-specific regression analyses, we systematically characterize how persona categories, target social groups, and prompt templates independently and jointly shape both refusal behavior and output toxicity. As a complementary case study, we further explore an iterative, evaluator-guided mitigation strategy based on model feedback with an external LLM evaluator, demonstrating that highly toxic outputs can be substantially reduced without costly model retraining. Overall, our findings highlight the importance of culturally contextualized safety evaluations for Chinese-language LLMs and provide a structured framework for assessing persona-induced risks and exploratory mitigation strategies in LLM-generated content.

Carlo Bono, Federico Belotti, Matteo Palmonari

Linking textual values in tabular data to their corresponding entities in a Knowledge Base is a core task across a variety of data integration and enrichment applications. Although Large Language Models (LLMs) have shown State-of-The-Art performance in Entity Linking (EL) tasks, their deployment in real-world scenarios requires not only accurate predictions but also reliable uncertainty estimates, which require resource-demanding multi-shot inference, posing serious limits to their actual applicability. As a more efficient alternative, we investigate a self-supervised approach for estimating uncertainty from single-shot LLM outputs using token-level features, reducing the need for multiple generations. Evaluation is performed on an EL task on tabular data across multiple LLMs, showing that the resulting uncertainty estimates are highly effective in detecting low-accuracy outputs. This is achieved at a fraction of the computational cost, ultimately supporting a cost-effective integration of uncertainty measures into LLM-based EL workflows. The method offers a practical way to incorporate uncertainty estimation into EL workflows with limited computational overhead.