Understanding the prevalence of misinformation in health topics online can inform public health policies and interventions.However, measuring such misinformation at scale remains a challenge, particularly for high-stakes but understudied topics like opioid-use disorder (OUD)-a leading cause of death in the U.S. We present the first large-scale study of OUDrelated myths on YouTube, a widely-used platform for health information.With clinical experts, we validate 8 pervasive myths and release an expert-labeled video dataset.To scale labeling, we introduce MYTHTRIAGE, an efficient triage pipeline that uses a lightweight model for routine cases and defers harder ones to a high-performing, but costlier, large language model (LLM).MYTHTRIAGE achieves up to 0.86 macro F1-score while estimated to reduce annotation time and financial cost by over 76% compared to experts and full LLM labeling.We analyze 2.9K search results and 343K recommendations, uncovering how myths persist on YouTube and offering actionable insights for public health and platform moderation. 1 Warning: Some content of this paper, included to contextualize our data, is misleading.

Online communities on Reddit are a popular choice among people with opioid use disorder (OUD) to seek information on drug use, withdrawal symptoms, and recovery. LLM-powered chatbots (e.g., ChatGPT) are widely being adopted as question-answer systems for health-related queries. However, such online health information seeking could potentially be hindered by myths and misinformation on OUD, misleading or causing genuine harm to people with OUD. In this work, we examine the prevalence of 5 OUD-related myths, on treatment models and patient characteristics, within human- (taken from Reddit) and LLM-generated responses to queries on OUD. We further explore the framing strategies used within responses (both human- and LLM-generated) promoting and countering the myths. We found that all 5 myths were more widespread within human-generated responses. In addition, myth-promoting responses adopted trustworthy and authoritative framings, compared to knowledge-imparting linguistic cues within those countering the myths. Our work offers recommendations to reduce online OUD misinformation.

With the advent of cloud platforms that offer GPU-as-a-Service (GPUaaS), such as Amazon EC2 and Microsoft Azure, researchers increasingly rely on virtual GPU instances for training deep learning (DL) workloads. These GPU instances vary in configuration attributes, including but not limited to the number of GPUs, the number of vCPUs, and per-hour usage cost. Identifying the appropriate GPU instance for training a DL workload becomes extremely difficult due to the huge GPU instance selection space offered by the cloud platforms and the corresponding variation in training performance or computational needs of different DL workloads. In this paper, we propose a GPU instance recommendation system called GPURank, which provides a recommended list of GPU instances to choose from for DL workloads. GPURank predicts and leverages two metrics: epoch training cost and average GPU utilization to make this choice. We curated a new benchmark dataset by profiling diverse DL workloads to train the regression models in GPURank’s prediction framework. We demonstrate that GPURank beats baselines on two pertinent problem settings: (1) unseen workloads and (2) unseen GPU instances, with a 25.89% and 20.10% higher average ranking performance on these respectively.

Discussions of Violence Against Women (VAW) in publicly accessible forums like online news media can influence the perceptions of people and organizations. Language reinforcing stigma around VAW can result in negative consequences such as unethical representation of survivors and trivialization of the act of violence. In this work, we study the presence of stigmatized framings in news media and how it differs based on media attributes like regionality, political leaning, veracity, and latent communities of news sources. We also investigate the interactions between VAW-based stigma and 14 issue-generic policies used to describe political communications. We found that articles from national, right-leaning, and conspiratorial news sources contain more stigma compared to their counterparts. Furthermore, alignment of articles to the issue-generic policies offers the highest explanation for the presence of stigma in news articles. We discuss implications for institutions to improve safe reporting guidelines on VAW.

Gender-based violence (GBV) is an ongoing public health issue. Prevention practice and research on GBV contend with an incomplete understanding of its public health burden due to gender-exclusive definitions of GBV and inhibited survivor disclosure. Prior work in CSCW and HCI has explored sensitive disclosures and GBV conversations but excludes examination of conversation dynamics in surfacing knowledge of GBV. We used a mixed-methods approach to understand the phenomenon characterized by Muted Group Theory as a mechanism inhibiting disclosure in the context of GBV discussions on Reddit. Using an iterative process informed by literature of GBV research on cis-gender women, girls, trans, and non-binary populations, we developed comprehensive keywords to obtain, annotate, and analyze 298 posts and 10,369 comments about GBV across 7 subreddits. We found that muting faced by survivors offline, precluding reporting, is replicated on the Reddit platform. This study surfaced 5 categories of muting in discussions of GBV situated by variations in communication norms between dominant and non-dominant groups. These findings were supported by analysis of linguistic attributes that inform an Ensemble Classifier's detection of muting in conversations of GBV. The results offer that muting is a harmful occurrence in online disclosures of GBV that is mediated by existing moderation practices. This work contributes an expanded understanding of GBV conversations online, muting as a feature of those conversations, and an initial foray into detection to inform muting prevention.

Co-design of technology encourages participation and decision-making input of end-users. In the case of technologies for individuals with Intellectual and Developmental Disabilities (IDD), the end-users are historically left out of the design process. Further deepening the disconnect between this group and technology, they are also excluded from formal technology design knowledge sharing, such as college courses. To address this, our study investigates the efficacy of a formal classroom adaptation of co-design activities to encourage learning and participation. Through collaboration between educators and designers, we adopted user-centered co-design activities to facilitate knowledge and application of technological design methods within a class of 13 students with IDD. Findings uncovered factors contributing to co-teaching collaboration planning and reflection between educators and designers, and ways that activities can provide accessible collaborative learning environments for students with IDD by supporting collaboration, cognitive engagement, and meta-cognition. We discuss how these factors can support successful co-teacher collaborations that promote student empowerment. Finally, we contribute collaborative co-teaching strategies for educational co-design activities for individuals with IDD.

Mental health discussions on public forums influence the perceptions of people. Negative consequences may result from hostile and “othering” portrayals of people with mental disorders. Adopting the lens of Moral Foundation Theory (MFT), we study framings of mental health discourse on Twitter and News, and how moral underpinnings abate or exacerbate stigma. We adopted a large language model based representation framework to score 13,277,115 public tweets and 21,167 news articles against MFT’s five foundations. We found discussions on Twitter to demonstrate compassion, justice and equity-centered moral values for those suffering from mental illness, in contrast to those on News. That said, stigmatized discussions appeared on both Twitter and News, with news articles being more stigmatizing than tweets. We discuss implications for public health authorities to refine measures for safe reporting of mental health, and for social media platforms to design affordances that enable empathetic discourse.

The behavior of information cascades (such as retweets) has been modeled extensively. While point process-based generative models have long been in use for estimating cascade growths, deep learning has greatly enhanced the integration of diverse features and signals. We observe two significant temporal signals in cascade data that have not been reported or exploited to our knowledge. First, the popularity of the cascade root is known to influence cascade size strongly; but we find that the effect falls off rapidly with time. Second, we find a measurable positive correlation between the novelty of the root content (with respect to a streaming external corpus) and the relative size of the resulting cascade. Responding to these observations, we proposeGammaCas, a new cascade growth model as a parametric function of time, which combines deep influence signals from content (e.g., tweet text), network features (e.g., followers of the root user), and exogenous event sources (e.g., online news). Specifically, our model processes these signals through a customized recurrent network, whose states then provide the parameters of the cascade rate function, which is integrated over time to predict the cascade size. The network parameters are trained end-to-end using observed cascades.GammaCasoutperforms seven recent and diverse baselines significantly on a large-scale dataset of retweet cascades coupled with time-aligned online news — it beats the best baseline with 18.98% increase in terms of Kendall's$\tau$correlation and a reduction of 19.2 in Mean Absolute Percentage Error. Extensive ablation and case studies unearth interesting insights regarding retweet cascade dynamics.

Community affiliation of a node plays an important role in determining its contextual position in the network, which may raise privacy concerns when a sensitive node wants to hide its identity in a network. Oftentimes, a target community seeks to protect itself from adversaries so that its constituent members remain hidden inside the network. The current study focuses on hiding such sensitive communities so that the community affiliation of the targeted nodes can be concealed. This leads to the problem of community deception, which investigates the avenues of minimally rewiring nodes in a network so that a given target community maximally hides from a community detection algorithm (CDA). We formalize the problem of community deception and introduce Network deception using permanence loss (NEURAL), a novel method that greedily optimizes a node-centric objective function to determine the rewiring strategy. Theoretical settings pose a restriction on the number of strategies that can be employed to optimize the objective function, which in turn reduces the overhead of choosing the best strategy from multiple options. We also show that our objective function is submodular and monotone. When tested on both synthetic and seven real-world networks, NEURAL is able to deceive six widely used CDAs. We benchmark its performance with respect to four state-of-the-art methods on four evaluation metrics. In addition, our qualitative analysis on three other attributed real-world networks reveals that NEURAL, quite strikingly, captures important metainformation about edges that otherwise could not be inferred by observing only their topological structures.