Papers for
social media managers
Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.
Online climate action varies by distance in Los Angeles wildfire event
Place and Political Behavior Online during the Los Angeles Blaze
Abstract: 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.
Viral trends use image editing differently from education and marketing
How are viral trends different from educational and commercial uses of LLM image generation and editing? A privacy-preserving descriptive analysis
Abstract: Gemini's Nano Banana models, first introduced in late 2025, led to a shift in the capabilities and use of image generation and editing. We utilize Urania, a Google-developed framework for generating insights about LLM chatbot interactions with rigorous differential privacy (DP) guarantees, to analyze use cases and intents, techniques and ways of prompting, without ever looking at a single image or individual prompt. We select a subsample of 2.1M conversations in early 2026 which utilized Nano Banana models, into roughly 800 clusters in five categories: (1) Social & Viral Sharing, (2) Brand & Marketing Collateral, (3) Educational & Technical Information Media, (4) Product Visualizations & E-commerce, and (5) Architectural & Spatial Designs. We then compare these clusters on 21 features of use, measuring aspects of prompts' (a) editing techniques, (b) artistic style requests, (c) composition, and (d) chatbot interactions. This allows us to describe a baseline of how people are using image generation and editing tools for different use cases---(including growing and emergent use cases which were not possible before Nano Banana)---in a privacy-preserving manner.
Trust and transparency in ai-altered linkedin posts match human-written ones
Perceived Trust and Transparency in AI-Altered vs. Human-Written LinkedIn Posts
Abstract: People increasingly draft a social media post themselves and then let an AI model refine it (AI-altered), a common but understudied middle case between fully human-written and fully AI-generated content. We ask whether this changes how trustworthy and transparent readers perceive the post to be, and whether they can even tell. We first analyze real ChatGPT conversations (WildChat) to find how people actually alter social media posts, which we use to design realistic alteration prompts. We then run a pre-registered, within-subjects online experiment (n=214) in which participants rate unlabeled human-written and AI-altered LinkedIn-style posts for perceived trustworthiness and transparency, and separately judge whether each post was AI-altered. We find that AI-alteration itself has no detectable effect on either perceived trustworthiness or transparency (both estimates near zero, not significant). At the same time, readers did not reliably detect AI-alteration (49\% accuracy, near chance) and defaulted to assuming human authorship when unsure. However, our analysis shows significant differences in transparency and trustworthiness ratings across the specific post, the AI model, as well as the prompt used to alter it. For instance, more detailed prompts raise both perceived trustworthiness and transparency but also make the alteration more detectable. Surprisingly, the prompt that shifted ratings most also asked the model to add one new idea absent from the original post: readers nonetheless rated this AI-invented, unverifiable content as more trustworthy and transparent.
Ai suggestions help social media creators reply faster but need careful checking
The Impact of Personalized AI Response Suggestions on Content Creators' Direct Messaging
Abstract: Direct messaging is often employed by social media content creators to directly interact with followers. Those creators attempt to engage with followers in private communication, but they often encounter a large volume of messages. Through formative interviews with 15 creators, we found that creators frequently experience fatigue from writing suitable content, as they moderate audience expectations while concurrently integrating their media personas into responses. To explore how AI can address these challenges, we designed a system providing personalized, AI-generated response suggestions tailored to creators' writing styles to maintain their media personas. In a comparison study with 13 creators that examines how the system affects response behaviors, creators valued the system's ability to soften potential impoliteness and anticipated higher response rates, helping them maintain a professional image. However, creators reported additional labor in verifying the authenticity of AI outputs to mitigate risks caused by AI misalignment. Our results indicate that while AI suggestions enhance efficiency, they can undermine a creator's engagement in communication by shifting their attention toward monitoring AI.
Source attribution shapes public responses to digital testimony on social media
Source Attribution and Audience Uptake in Digital Witnessing: Evidence from a Chinese Diaspora Account on X
Abstract: Digital-witnessing research explains how citizen testimony becomes publicly visible. Less is known about audience uptake after publication and whether the attributed source corresponds to the form of response. This study analyzes 47,635 posts (April 2022 to May 2026) from Teacher Li Is Not Your Teacher, a diaspora-based account that curates and publishes reports and testimony from inside China for transnational audiences. Among 25,093 witness-relevant posts, 3,601 explicitly attribute their material to ordinary submitters. These posts receive more replies than other witness-relevant posts across adjusted analyses and show a more reply-oriented profile. The primary doubly robust retweet estimate is near zero. Among audited replies to November 2022 user-submitted posts, 91.4% substantively engage the focal material. The pattern is consistent with ordinary-person source attribution presenting testimony as situated and open to public response. The study extends digital witnessing beyond visibility by separating redistribution from the public response that gathers around testimony.
Social reinforcement boosts word-of-mouth influence in marketing networks
How Social Reinforcement Shapes Word-of-Mouth: A Conviction-Based Model for Marketing Diffusion and Network Targeting
Abstract: Organizations increasingly rely on social networks to drive product adoption and word-of-mouth diffusion, yet a central challenge is that consumers differ not only in whether they adopt but also in how strongly they subsequently advocate. In particular, individuals who adopt under strong social reinforcement often develop greater conviction and become more persuasive. Existing diffusion models largely abstract away from this mechanism by treating influence strength as static. We introduce a conviction-based diffusion framework, the Pressure Threshold (PT) model, that captures how social reinforcement shapes both adoption and subsequent advocacy. In this model, an individual's influence increases with the cumulative social pressure experienced at adoption, providing a tractable representation of conviction formation while preserving the interpretability of classical threshold-based models. We study the managerial implications of this mechanism through marketing targeting and seeding. We show that accounting for reinforcement fundamentally alters optimal targeting strategies, shifting emphasis toward consumers in cohesive network regions where repeated exposure can amplify downstream word-of-mouth. While reinforcement introduces nonstandard features into the optimization problem, we demonstrate that scalable heuristics remain effective and provide data-dependent performance guarantees. Across synthetic and real-world networks, we find that reinforcement often substantially increases the effectiveness of seeding interventions, particularly in dense and clustered environments where opportunities for social validation are abundant. Using time-stamped social media cascades, we further show that the PT model improves predictive accuracy relative to standard diffusion models, highlighting the empirical relevance of conviction-driven amplification.
Long lived agents evaluated over weeks of real world replayed data
ReLiveGym: Evaluating Long-Lived Agents over Weeks of Replayed Reality
Abstract: As large language model (LLM) agents become widely adopted, they are increasingly deployed for tasks that require persistent monitoring or recurring actions (e.g., market analysis). These agents are expected to operate unattended for days or weeks, act at the right timing, and adapt to the dynamic environment over time. These challenges are not fully captured in the existing long-horizon agent work, as they often consider a static environment that is not temporally changing. We introduce ReLiveGym, a diagnostic evaluation environment of long-lived tasks in which agents act sparsely over simulated weeks of chronologically replayed real-world news, market, and social-media streams. The tasks span diverse levels of time sensitivity, reasoning intensity, and recurrence. Across eight base language models, we investigate how model choice and harness design affect agent performance on such long-lived tasks. Our results show that how agents determine when to act arises as an important harness-design axis for long-lived tasks; and that the optimal design varies across tasks and sometimes model choices as well. We also evaluate how continuous learning from hindsight feedback affects performance and addresses failure modes observed in these long-lived tasks. These findings indicate model choice, action timing mechanism, and use of feedback as important considerations in the design of long-lived agents. Code: https://github.com/SaharaLabsAI/ReLiveGym
Public reactions show mixed views on ai memory wearables
"Black Mirror?": Public Sensemaking of AI-Powered Lifelogging
Abstract: AI-powered lifelogging wearables are emerging as a new class of consumer devices that transform everyday experience into searchable, AI-curated memory archives. We study early public sensemaking around these systems at the moment of their market entry, using the Looki L1 as an empirical lens. Analysing large-scale Chinese-language and English-language social media discourse (N = 5,053 comments), we combine topic clustering with inductive thematic analysis to examine how users interpret the social, moral, and political implications of AI-mediated memory. Across contexts, users reference dystopian surveillance imaginaries, express privacy resignation and bystander concerns, and debate assistive value alongside consumer logics. English-language comments more often framed these devices through interpersonal power, evidentiary use, and hacking anxieties, while Chinese-language comments more often foregrounded labour exploitation, governance surveillance, and technological inevitability.
Opposing comments boost engagement and shift attitudes on social media
The Influence of the Vocal Few: Evidence from Social Media Comments
Abstract: Online comment sections let a small number of vocal individuals reach far beyond their own networks. We conduct a large-scale field experiment on Facebook that randomizes the presence and stance of comments beneath posts for a racial justice organization, reaching around one million U.S. users. Opposing comments increase reactions, comments, and link clicks by 15-43 percent relative to no comments, whereas supportive comments have little effect. A complementary survey experiment shows that similar opposing comments make attitudes less progressive and reduce donations to the organization. Through a common feature of online platforms, the vocal few can exert outsized influence.
Digital platforms reshape use of formal and informal Arabic online
Digital diglossia: Arabic between X and Facebook
Abstract: This study highlights the distribution of Standard Arabic (SA; H(igh) variety) and Colloquial Arabic (CA; L(ow) variety) across X and Facebook. 16754 public posts were collected via Python, with 10000 retained as the net dataset. Posts were classified into 7 discourse categories: *politics, technology, science, business, culture, fun,* and *sports*. Bivariate analyses, including Chi-square tests and Cramer's V (CV), examined associations among platform, discourse category, and diglossic choice, while binary logistic regression with Platform x Discourse Category interactions tested whether these associations varied across platforms. Findings reveal that there are significant associations between discourse category and diglossic choice on X, chi-square(6, *N* = 5000) = 600.35, p < .001, CV = .347, and Facebook, chi-square(6, N = 5000) = 1249.52, p < .001, CV = .500. Across platforms, platform was also associated with diglossic choice, chi-square(1, N = 10000) = 262.16, p < .001, CV = .162. Binary logistic regression further shows higher odds of SA use on X than Facebook in the political reference category (*OR* = 1.31, p = .0028), with significant platform-by-domain interactions for Culture (OR = 2.65), Fun (*OR* = 6.34), Sports (*OR* = 26.71), Science (OR = 0.41), and Technology (OR = 0.71). The study concludes that the diglossic use of SA and CA contributes to the growing body of research on digital discourse, unveiling that the digital age reshapes but does not erode diglossic boundaries, giving rise instead to a reconfigured digital diglossia.
Behavior aware method improves social media influencer role playing
From Memory to Behavior: A Behavior-Aware Role-Playing Framework for Social Media Influencers
Abstract: Large language models have shown strong potential as role-playing agents for real individuals, yet faithful impersonating remains challenging. Existing in-context learning-based methods fail to capture how individuals react under different situations. In addition, LLM-based evaluation is difficult for obscure individuals. To address these challenges, we propose Situation--Internal state--Behavior Persona method to incorporate situation-dependent behavioral strategies. We further design an evaluation protocol that provides LLM evaluators with references about the impersonated individual. We evaluate our approach on a newly constructed dataset for the task of generating replies on social media. Experimental results show that our proposed method outperforms state-of-the-art ICL-based baselines, while our evaluation protocol achieves moderate correlation with human judgment. Besides, experiments on fictional-character benchmarks demonstrate that our proposed method is applicable beyond the social media setting. These findings suggest that incorporating behavioral information broadly improves the fidelity of role-playing for real individuals on social media or fictional characters.