Searcharxiv⌕ Search

arXiv subjects

Luca Maria Aiello

Publications and source records attributed to Luca Maria Aiello.

At least 19 recordsLinked to original sources

Place and Political Behavior Online during the Los Angeles Blaze

Understanding how online communication during extreme weather crises enables or constrains climate mobilization is critical. During a crisis, the online world connects distant audiences and complicates the relationship between place and political behavior. We analyze all sociopolitical content from Bluesky about the January 2025 Los Angeles wildfire event, reporting how participation in collective action and allocation of attention vary by geographic distance. Participation varies nonlinearly with proximity, with local and international users participating more than regional and national users. Attention concentrates around regional and national users engaging primarily in viral, non-participatory political discussion. Analysis reveals a geographic mismatch between participation and attention, illuminating how online communication concentrates attention away from collective action amid an exacerbating climate crisis.

cs.SI↗

Network topology shapes consensus and vulnerability in LLM populations

As large language models (LLMs) are deployed in interacting populations, their collective behaviour depends not only on the models but also on who interacts with whom. Studies of convention formation in LLM populations have mainly tested well-mixed interactions. We place policies extracted from four LLMs on fully connected, random regular, and heterogeneous networks, with $10^3$ to $10^5$ agents playing the naming game. Topology affects how quickly consensus emerges, sometimes whether it emerges at all, and, in rare cases, which convention wins. Sparsest networks slow consensus up to twelvefold. Yet, by preserving correlations between persistent neighbours, they restore consensus in some cases that fail to converge under well-mixed interactions. Networks also govern vulnerability to adversarial attack. We measure the critical mass of committed agents needed to overturn an established convention. Tipping thresholds follow the same ordering across models and conditions: hubs $\leq$ random $\leq$ low-degree nodes. On the most heterogeneous topology, median critical masses are 3%, 16%, and 35%, respectively; targeting hubs requires one-sixth as many committed agents as random placement. For most models, networked populations are easier to tip than well-mixed ones, so well-mixed estimates can overstate robustness. Assessing the safety of multi-agent AI requires mapping who interacts with whom.

physics.soc-ph↗

Validating LLM-generated political networks with survey-reported partisan exposure

Large language models (LLMs) are increasingly used to simulate human behavior in social science research, including decisions, interactions, and social relationships. However, in network formation tasks, existing evaluations often focus on aggregate network properties, while networks with plausible structural properties may still fail to reproduce individuals' social and political environments. Here, we generate LLM-based political networks using personas sampled from the nationally representative ANES 2020--2022 Social Media Study and use respondents' survey-reported partisan exposure as an empirical benchmark for the political environments generated for the corresponding personas. We first characterize the structural properties and political assortativity of the generated networks, and then compare ego-level partisan exposure with the survey. We find that the structure and political assortativity of the generated networks vary across experimental configurations. More importantly, when comparing generated partisan exposure with the survey, some models generate stronger separation between Democratic and Republican egos than reported in the survey, with higher same-party and lower cross-party exposure, while others mainly underestimate Republican exposure across ego groups. Moreover, networks with similar political assortativity can still generate different partisan environments around the same personas, including among networks with plausible structural properties. Finally, generated networks show group-specific errors, under- or overestimating exposure to either Democrats or Republicans for specific social groups. These findings show that evaluating LLM-based social simulations only through aggregate outcomes can hide errors in the local environments through which those outcomes emerge, highlighting the importance of validation at both aggregate and individual levels.

cs.SI↗

Peer Influence across Heterogeneous AI Models

When two AI agents disagree, who persuades whom? As multi-agent systems increasingly combine language models of different families and sizes, the answer can determine which judgments survive interaction. Measuring persuasion as the probabilistic shift in an agent's decision after a single exchange with a dissenting peer, we test seven open-weight models across three language understanding tasks. We find that persuasion is strong: when models disagree, receivers often abandon their initial judgment after seeing a peer's answer and explanation. Surprisingly, however, neither standalone certainty nor model scale reliably predicts persuasion dynamics. Models producing almost perfectly consistent decisions in isolation can be among the most susceptible to persuasion, and small models can match larger ones as persuaders and resist their influence just as effectively. Furthermore, we show that the size of the shift depends more on the susceptibility of the listener than on the persuasiveness of the speaker. Persuasion patterns are therefore specific to each model pairing, with heterogeneity amplifying persuasion in some combinations and suppressing it in others, allowing a dissenting agent running a small model to overturn the judgments of a much larger one. These findings show that the behavior of interacting models cannot be inferred from their individual properties but must be evaluated in the combinations in which they will operate.

cs.AI↗

Indirect tipping: a social attack surface in AI agent populations

As generative AI agents are deployed at scale, safety will depend not only on technical safeguards and individual model design, but also on collective equilibria that determine how agent populations process information, prioritize actions, and respond to uncertainty. Yet the same equilibria that enable agents to coordinate also create a social attack surface. The standard framework to assess this vulnerability is critical mass dynamics: the minimum fraction of adversarial agents required to overturn an equilibrium through direct competition. Here, we show that this approach risks underestimating system vulnerability by reducing the problem to the identification of singular tipping points, and ignoring indirect but potentially more efficient routes through which collective behavior can be redirected. Through experiments with populations of LLM agents and an analytic framework that captures their collective dynamics at scale, we map critical-mass thresholds that define a directed, weighted topology over the space of coordination equilibria, and treat this topology as a navigable landscape. We show that indirect tipping through intermediate stepping-stone equilibria can reduce the committed minority required to reach an alternative state, bypass majority requirements, and make possible transitions inaccessible through direct challenges. The diversity of available alternatives and timing of the attack further reshape this landscape, creating opportunities for control as well as risks of unintended destabilization. These results show that an equilibrium's resistance to committed intervention is not an intrinsic property but a structural feature of its competitive relations with alternative states. Securing populations of interacting AI agents therefore requires mapping this social landscape alongside individual agent capabilities and the technical channels through which they interact.

cs.MA↗

Overreliance on AI in Information-seeking from Video Content

The ubiquity of multimedia content is reshaping online information spaces, particularly in social media environments. At the same time, search is being rapidly transformed by generative AI, with large language models (LLMs) routinely deployed as intermediaries between users and multimedia content to retrieve and summarize information. Despite their growing influence, the impact of LLM inaccuracies and potential vulnerabilities on multimedia information-seeking tasks remains largely unexplored. We investigate how generative AI affects accuracy, efficiency, and confidence in information retrieval from videos. We conduct an experiment with around 900 participants on 8,000+ video-based information-seeking tasks, comparing behavior across three conditions: (1) access to videos only, (2) access to videos with LLM-based AI assistance, and (3) access to videos with a deceiving AI assistant designed to provide false answers. We find that AI assistance increases accuracy by 3-7% when participants viewed the relevant video segment, and by 27-35% when they did not. Efficiency increases by 10% for short videos and 25% for longer ones. However, participants tend to over-rely on AI outputs, resulting in accuracy drops of up to 32% when interacting with the deceiving AI. Alarmingly, self-reported confidence in answers remains stable across all three conditions. Our findings expose fundamental safety risks in AI-mediated video information retrieval.

cs.CY↗

Measuring Narrative Polarization in Online Discourse

Polarization research has demonstrated how people cluster in homogeneous groups with opposing opinions. However, this effect emerges not only through interaction between people, limiting communication between groups, but also between narratives, shaping opinions and partisan identities. Yet, how polarized information environments portray opposing interpretations of reality, and whether narratives move between content environments despite limited interactions, remains unexplored. To address this gap, we formalize the concept of narrative polarization and demonstrate its measurement in 212 YouTube videos and 90,029 comments on the Israeli-Palestinian conflict. Based on structural narrative theory and implemented through a large language model, we extract the narrative roles assigned to central actors in two partisan information environments. We find that while videos produce highly polarized narratives, comments exhibit significantly lower levels of narrative polarization, converging on shared core narrative structures. However, recurring narrative motifs capturing more complex actor constellations reveal additional differences between partisan environments.

cs.CY↗

Unmasking Conversational Bias in AI Multiagent Systems

Detecting biases in the outputs produced by generative models is essential to reduce the potential risks associated with their application in critical settings. However, the majority of existing methodologies for identifying biases in generated text consider the models in isolation and neglect their contextual applications. Specifically, the biases that may arise in multi-agent systems involving generative models remain under-researched. To address this gap, we present a framework designed to quantify biases within multi-agent systems of conversational Large Language Models (LLMs). Our approach involves simulating small echo chambers, where pairs of LLMs, initialized with aligned perspectives on a polarizing topic, engage in discussions. Contrary to expectations, we observe significant shifts in the stance expressed in the generated messages, particularly within echo chambers where all agents initially express conservative viewpoints, in line with the well-documented political bias of many LLMs toward liberal positions. Crucially, the bias observed in the echo-chamber experiment remains undetected by current state-of-the-art bias detection methods that rely on questionnaires. This highlights a critical need for the development of a more sophisticated toolkit for bias detection and mitigation for AI multi-agent systems. The code to perform the experiments is publicly available.

cs.CL↗

Billions of Sketches Reveal Hidden Cultural Variation in Human Concepts

Claims about the universality of human concepts have been predominantly assessed through linguistic similarity across languages and cultures. However, words are effective as communication devices because they compress rich experiential variation into shared conventions, potentially obscuring hidden individual and cultural differences in how concepts are mentally represented. Here, we analyse 2.6 billion human-made sketches of common concepts from 236 countries and territories to examine conceptual structure through people's visual imagination. Consistent with recent work on image-based cognition, we find that single concepts unfold into multiple distinct visual exemplars, revealing latent information about similarities and differences in conceptual structure across cultures. This variation is strongest for concepts involving haptic interaction, suggesting that visual imagery reflects variation in embodied experience as much as conventional definitions. Comparing embedding models of sketches with word embedding models across languages, we find that their geometries diverge, with visual representations preserving rich semantic and cultural structure that language models compress. Cross-cultural similarities derived from sketches align 32% more closely with established cultural distances than do text-based measures. Together, these results suggest that patterns of human conceptual universality may depend critically on the modality through which concepts are measured, with large-scale sketching providing a direct, high-resolution probe of conceptual diversity across embodied and cultural dimensions of thought.

cs.CY↗

AI Persuasive Framing in Collective Dilemmas

AI agents are promising tools that can act as flexible behavioral nudges to enhance human cooperation in addressing large-scale societal problems. However, evidence on whether AI agents can effectively boost cooperation remains mixed. We recruited 1,283 participants to play iterated Collective Risk Games in small groups, testing whether AI assistants could nudge participants toward cooperation. By using persuasive framing personalized to each player's Social Value Orientation profile, the AI interventions significantly increased contributions and group success rates. These cooperative effects were short-lived, however, fading after the first few rounds. Strikingly, when the AI treatments were reconfigured to promote selfish behavior through exculpatory framing, the negative effects on contributions and group success were larger and substantially more persistent, particularly for personalized interventions. This asymmetry between prosocial and antisocial persuasion highlights the dual-use risks of AI systems designed to influence group behavior in collective action settings.

cs.CY↗

P1SCO: Social Dimensions from a Perspectivist Lens

We introduce P1SCO, a dataset of social media comments collected from three distinct platforms, annotated according to ten social dimensions to capture the diversity of social interactions and perceptions. The dataset is carefully disaggregated to allow analysis at the level of individual comments, annotators, and platforms. In addition to the social dimension labels, we include rich metadata on the annotators, including demographics, Big Five personality profiles, and political affiliation. This combination of comment-level annotations and annotator-level features enables nuanced analyses of how social perception varies across platforms, individual differences, and demographic factors. By preserving the diversity of annotator perspectives, our dataset supports studies of inter- and intra-annotator agreement, the influence of personality and political orientation on social interpretation, and the cross-platform dynamics of social discourse.

cs.CL↗

Learning Perspectivist Social Meaning via Demographic-Conditioned Fusion Embeddings

Social meaning in language is inherently perspectival, varying across annotator backgrounds, demographics, and ideological positions. However, most NLP systems collapse this variation into a single ground-truth label, ignoring the diversity of interpretations. In this work, we model social dimensions along a perspectivist spectrum, capturing how interpretations vary across demographic groups on a dataset consisting of 28k human annotations. We benchmark multiple modeling paradigms, including zero-shot, few-shot, and fine-tuned approaches, and propose fusion embeddings that integrate textual and demographic representations. Our fusion models yield consistent and statistically significant improvements over text-only baselines across all fusion strategies (+5.9-6.5% relative macro PR-AUC), with shuffle ablations confirming that demographic profiles carry genuine predictive signal rather than spurious correlations.

cs.CL↗

Failure of contextual invariance in large language models

Standard evaluation practices assume that large language model (LLM) outputs are stable when prompts are embedded in contextually equivalent discourses. Here, we test this assumption in the setting of gender inference. Using a controlled pronoun selection task, we introduce minimal, theoretically uninformative discourse context and find that this induces large, systematic shifts in model outputs. Correlations with cultural gender stereotypes, present in decontextualized settings, weaken or disappear once context is introduced, while theoretically irrelevant features, such as the gender of a pronoun for an unrelated referent, become the most informative predictors of model behavior. A Contextuality-by-Default analysis reveals that, in 19--52\% of cases across models, this dependence persists after accounting for all marginal effects of context on individual outputs and cannot be attributed to simple pronoun repetition. These findings show that LLM outputs violate contextual invariance even under near-identical syntactic formulations, with implications for bias benchmarking and deployment in high-stakes settings.

cs.CL↗

Auditing the Reliability of Multimodal Generative Search

Multimodal Large Language Models (MLLMs) increasingly function as generative search systems that retrieve and synthesize answers from multimedia content, including YouTube videos. Although these systems project authority by citing specific videos as evidence, the extent to which these citations genuinely substantiate the generated claims remains underexplored. We present a large-scale audit of the Gemini 2.5 Pro multimodal search system, analyzing 11,943 claim-video pairs generated across Medical, Economic, and General domains. Through automated verification using three independent LLM judges (87.7\% inter-rater agreement), validated against human annotations, we find that depending on the judge's strictness, between 3.7\% and 18.7\% of video-grounded claims are not supported by their cited sources. The dominant failure modes are not outright contradictions but rather unverifiable specificities and overstated claims, suggesting the system injects precise but ungrounded details from parametric knowledge while citing videos as evidence. Exploratory post-hoc analysis via logistic regression reveals properties associated with these failures: claims departing from source vocabulary ($β= -1.6$ to $-3.1$, $p < 0.01$) and claims with low semantic similarity to the video transcript ($β= -2.1$ to $-11.6$, $p < 0.01$) are significantly more likely to be unsupported. These findings characterize the current trustworthiness of video-based generative search and highlight the gap between the confidence these systems project and the fidelity of their outputs. The dataset is available at https://anonymous.4open.science/r/icwsm-gemini-audit-04DF .

cs.CY↗

Twitter climate discourse as a signal of pro-environmental behaviors

Fostering coordinated pro-environmental behaviors at scale is a key challenge for climate mitigation. Individual actions only generate meaningful impact when they diffuse widely and become socially coordinated, yet monitoring such processes remains difficult with traditional survey-based tools alone. In this study, we examine whether large-scale online climate discourse is associated with differences in offline pro-environmental behavior across European regions. We combine geolocated Twitter data from the Climate Change Twitter Dataset (2017-2019) with survey-based measures from the 2019 Special Eurobarometer, focusing on the regional density of climate-related tweets and the average number of self-reported pro-environmental actions. We find a strong positive association between tweet density and pro-environmental behavior that remains robust to socio-economic controls, alternative spatial aggregations, and a wide range of robustness checks. To move beyond aggregate volume, we further decompose online discourse using Natural Language Processing tools that capture distinct social dimensions. While knowledge exchange shows no clear relationship with offline behavior, the prevalence of activism- and social support-related expressions is negatively associated with pro-environmental actions. Overall, our results suggest that online climate discourse can serve as an informative, attention-related signal of regional differences in pro-environmental behavior, but that different forms of online engagement relate to offline action in markedly different ways. More broadly, the study highlights the potential of integrating large-scale digital traces with survey data to investigate collective behavior in socio-environmental systems, while remaining explicitly observational in scope.

cs.SI↗

Framing Unionization on Facebook: Communication around Representation Elections in the United States

Digital media have become central to how labor unions communicate, organize, and sustain collective action. Yet little is known about how unions' online discourse relates to concrete outcomes such as representation elections. This study addresses the gap by combining National Labor Relations Board (NLRB) election data with 158k Facebook posts published by U.S. labor unions between 2015 and 2024. We focused on five discourse frames widely recognized in labor and social movement communication research: diagnostic (identifying problems), prognostic (proposing solutions), motivational (mobilizing action), community (emphasizing solidarity), and engagement (promoting social media interaction). Using a fine-tuned RoBERTa classifier, we systematically annotated unions' posts and analyzed patterns of frame usage around election events. Our findings showed that diagnostic and community frames dominated union communication overall, but that frame usage varied substantially across organizations. Greater use of diagnostic, prognostic, and community frames prior to an election was associated with higher odds of a successful outcome. After elections, framing patterns diverged depending on results: after wins, the use of prognostic and motivational frames decreased, whereas after losses, the use of prognostic and engagement frames increased. By examining variation in message-level framing, the study highlights how communication strategies correlate with organizational success, contributing open tools and data, and complementing prior research in understanding digital communication of unions and social movements.

cs.CY↗

Measuring Online Behavior Change with Observational Studies: a Review

Exploring online behavior change is imperative for societal progress in the context of 21st-century challenges. We analyze 148 articles (2000-2023) focusing on behavior change in the digital space and build a map that categorizes behaviors, behavior change detection methodologies, platforms of reference, and theoretical frameworks that characterize the analysis of online behavior change. Our findings reveal a focus on sentiment shifts, an emphasis on API-restricted platforms, and limited integration of theory. We call for methodologies able to capture a wider range of behavior types, diverse data sources, and stronger theory-practice alignment in the study of online behavior and its change.

cs.CY↗

Group size effects and collective misalignment in LLM multi-agent systems

Multi-agent systems of large language models (LLMs) are rapidly expanding across domains, introducing dynamics not captured by single-agent evaluations. Yet, existing work has mostly contrasted the behavior of a single agent with that of a collective of fixed size, leaving open a central question: how does group size shape dynamics? Here, we move beyond this dichotomy and systematically explore outcomes across the full range of group sizes. We focus on multi-agent misalignment, building on recent evidence that interacting LLMs playing a simple coordination game can generate collective biases absent in individual models. First, we show that collective bias is a deeper phenomenon than previously assessed: interaction can amplify individual biases, introduce new ones, or override model-level preferences. Second, we demonstrate that group size affects the dynamics in a non-linear way, revealing model-dependent dynamical regimes. Finally, we develop a mean-field analytical approach and show that, above a critical population size, simulations converge to deterministic predictions that expose the basins of attraction of competing equilibria. These findings establish group size as a key driver of multi-agent dynamics and highlight the need to consider population-level effects when deploying LLM-based systems at scale.

cs.MA↗