Alessia Antelmi

Alessia Antelmi, Alessia Galdeman, Lucio La Cava et al.

Large Language Models (LLMs) are increasingly employed as proxies for human behavior in computational social science. However, their tendency to internalize biases from training data raises concerns about their reliability in politically sensitive domains, specifically in regard to their susceptibility to persuasive language. In this work, we examine whether LLMs endorse persuasion-infused messages and whether partisan persona prompting modulates such endorsement. We evaluate six LLMs from different geographic regions on content annotated with persuasion techniques drawn from real-world media sources, measuring the likelihood of endorsement using a five-point Likert scale. The models are prompted as either a neutral social media user or as a user with left- or right-leaning political views. Results show that without political conditioning, LLMs generally do not endorse messages containing persuasion techniques, though model-level differences emerge, and that partisan persona prompting increases polarization of endorsement, particularly for persuasion-infused content. Endorsement further varies by persuasion technique and topic. These findings raise concerns about agentic LLM deployments in politically sensitive environments and complicate their use as reliable simulators of human political cognition.

Sunwoo Kim, Soo Yong Lee, Yue Gao et al.

Higher-order interactions (HOIs) are ubiquitous in real-world complex systems and applications. Investigation of deep learning for HOIs, thus, has become a valuable agenda for the data mining and machine learning communities. As networks of HOIs are expressed mathematically as hypergraphs, hypergraph neural networks (HNNs) have emerged as a powerful tool for representation learning on hypergraphs. Given the emerging trend, we present the first survey dedicated to HNNs, with an in-depth and step-by-step guide. Broadly, the present survey overviews HNN architectures, training strategies, and applications. First, we break existing HNNs down into four design components: (i) input features, (ii) input structures, (iii) message-passing schemes, and (iv) training strategies. Second, we examine how HNNs address and learn HOIs with each of their components. Third, we overview the recent applications of HNNs in recommendation, bioinformatics and medical science, time series analysis, and computer vision. Lastly, we conclude with a discussion on limitations and future directions.

Enrica Loria, Alessia Antelmi, Johanna Pirker

In most games, social connections are an essential part of the gaming experience. Players connect in communities inside or around games and form friendships, which can be translated into other games or even in the real world. Recent research has investigated social phenomena within the player social network of several multiplayer games, yet we still know very little about how these networks are shaped and formed. Specifically, we are unaware of how the game type and its mechanics are related to its community structure and how those structures vary in different games. This paper presents an initial analysis of Steam users and how friendships on Steam are formed around 200 games. We examine the friendship graphs of these 200 games by dividing them into clusters to compare their network properties and their specific characteristics (e.g., genre, game elements, and mechanics). We found how the Steam user-defined tags better characterized the clusters than the game genre, suggesting that how players perceive and use the game also reflects how they connect in the community. Moreover, team-based games are associated with more cohesive and clustered networks than games with a stronger single-player focus, supporting the idea that playing together in teams more likely produces social capital (i.e., Steam friendships).