arXiv subjects

Sabina Tomkins

Publications and source records attributed to Sabina Tomkins.

16 recordsLinked to original sources

arXiv2026

AnnotateThis: Analyzing a human-LLM system for annotating social media data with the concept of climate change mitigation pessimism

Large language models (LLMs) are increasingly being integrated into research workflows. However, LLMs have been shown to struggle with difficult and nuanced concepts such as those found in computational social science (CSS) research. Within the CSS community, there has been a call for new systems to be developed which center humans in LLM-supported scientific workflows. We develop AnnotateThis, a human-centered system for inspecting and improving LLM annotations, a process we refer to as LLM grounding for a target concept. AnnotateThis is developed with both computational and social scientists to reflect existing workflows for data annotation. It includes a range of information features for users to interrogate the quality and reliability of LLM annotations. We evaluate our system in two settings. In the first, we assume a researcher may not have access to ground truth data and that users of AnnotateThis have limited prior knowledge of the concept they would like an LLM to annotate. That is, they may be conducting concept specification and LLM grounding simultaneously. In the second setting, we assume access to ground truth labels and that the concept is specified for a given annotation task; here, the task of LLM grounding is more straightforward. We find that in both settings users can improve the quality of LLM annotations with AnnotateThis and that their final annotations far surpass those created without human intervention. For example, when we evaluate with ground truth labels, we see an absolute improvement of 0.15 in F-Measure and 0.23 in accuracy over a fully automated state-of-the-art method for prompt refinement.

cs.CY↗

arXiv2026

Anti-establishment sentiment on TikTok: Implications for understanding influence(rs) and expertise on social media

Distrust of public serving institutions and anti-establishment views are on the rise (especially in the U.S.). As people turn to social media for information, it is imperative to understand whether and how social media environments may be contributing to distrust of institutions. In social media, content creators, influencers, and other opinion leaders often position themselves as having expertise and authority on a range of topics from health to politics, and in many cases devalue and dismiss institutional expertise to build a following and increase their own visibility. However, the extent to which this content appears and whether such content increases engagement is unclear. This study analyzes the prevalence of anti-establishment sentiment (AES) on the social media platform TikTok. Despite its popularity as a source of information, TikTok remains relatively understudied and may provide important insights into how people form attitudes towards institutions. We employ a computational approach to label TikTok posts as containing AES or not across topical domains where content creators tend to frame themselves as experts: finance and wellness. As a comparison, we also consider the topic of conspiracy theories, where AES is expected to be common. We find that AES is most prevalent in conspiracy theory content, and relatively rare in content related to the other two topics. However, we find that engagement patterns with such content varies by area, and that there may be platform incentives for users to post content that expresses anti-establishment sentiment.

cs.SI↗

arXiv2025

How candidates evoke identity and issues on TikTok

Social media platforms are increasingly central to campaign communication, with both paid (advertising) and earned (organic) posts used for fundraising, mobilization, and persuasion. TikTok, and other short-form video platforms, with its short-video format and content-driven algorithms, demand unique content. We examine the final six months before the 2024 US Presidential Election to understand how major campaigns used TikTok. We frame our analysis around two political science theories. The first is the expressive (identity) model, where voters are motivated by their group memberships and candidates appeal to those identities. Alternatively, the instrumental (issues) model argues voters align with politicians advocating their key issues. We also examine how often candidates attacked opponents, reflecting literature showing attacks are common in politics. We combine two datasets: posts from the Harris and Trump campaigns on TikTok (July-November 2024) and a two-wave 2022 survey of around 1,000 respondents. Results show Trump more often disparaged Harris and emphasized identities and issues distinguishing Republicans, while Harris more often highlighted Democratic identities and valued issues. Although issues predict party ID, both candidates referenced identities more (34 percent of posts) than issues (25 percent), with most posts mentioning neither (55 percent).

cs.CY↗

arXiv2025

What does the public want their local government to hear? A data-driven case study of public comments across the state of Michigan

City council meetings are vital sites for civic participation where the public can speak directly to their local government. By addressing city officials and calling on them to take action, public commenters can potentially influence policy decisions spanning a broad range of concerns, from housing, to sustainability, to social justice. Yet studies of these meetings have often been limited by the availability of large-scale, geographically-diverse data. Relying on local governments' increasing use of YouTube and other technologies to archive their public meetings, we propose a framework that characterizes comments along two dimensions: the local concerns where concerns are situated (e.g., housing, election administration), and the societal concerns raised (e.g., functional democracy, anti-racism). Based on a large record of public comments we collect from 15 cities in Michigan, we produce data-driven taxonomies of the local concerns and societal concerns that these comments cover, and employ machine learning methods to scalably apply our taxonomies across the entire dataset. We then demonstrate how our framework allows us to examine the salient local concerns and societal concerns that arise in our data, as well as how these aspects interact.

cs.CY↗

arXiv2025

Identity isn't everything -- how far do demographics take us towards self-identified party ID?

How well do demographics explain party identification? Demographics are related to party identification in political polls, news articles, and academic publications. Yet, there is a diversity of party identification even within demographic groups which have historically been attached to one party. And some groups lack a clear connection to either party. It may be that demographics on their own fail to account for the fact that people generally belong to a variety of groups. They must select the groups which are most important to them when shaping a political identity, and may choose to construct an identity relatively unattached to any specific demographic group to which they belong. This prompts the question, do we need to consider measures of identity strength when using demographics to explain party identification? We utilize a predictive framework to address these questions and find that demographics are highly predictive for some groups (e.g., Black Democrats), while others benefit from the inclusion of identity strength (e.g., Hispanic Republicans).

cs.CY↗

arXiv2025

Counterfactual inference in sequential experiments

We consider after-study statistical inference for sequentially designed experiments wherein multiple units are assigned treatments for multiple time points using treatment policies that adapt over time. Our goal is to provide inference guarantees for the counterfactual mean at the smallest possible scale -- mean outcome under different treatments for each unit and each time -- with minimal assumptions on the adaptive treatment policy. Without any structural assumptions on the counterfactual means, this challenging task is infeasible due to more unknowns than observed data points. To make progress, we introduce a latent factor model over the counterfactual means that serves as a non-parametric generalization of the non-linear mixed effects model and the bilinear latent factor model considered in prior works. For estimation, we use a non-parametric method, namely a variant of nearest neighbors, and establish a non-asymptotic high probability error bound for the counterfactual mean for each unit and each time. Under regularity conditions, this bound leads to asymptotically valid confidence intervals for the counterfactual mean as the number of units and time points grows to $\infty$ together at suitable rates. We illustrate our theory via several simulations and a case study involving data from a mobile health clinical trial HeartSteps.

stat.ML↗

arXiv2025

Algorithms in the Stacks: Investigating automated, for-profit diversity audits in public libraries

Algorithmic systems are increasingly being adopted by cultural heritage institutions like libraries. In this study, we investigate U.S. public libraries' adoption of one specific automated tool -- automated collection diversity audits -- which we see as an illuminating case study for broader trends. Typically developed and sold by commercial book distributors, automated diversity audits aim to evaluate how well library collections reflect demographic and thematic diversity. We investigate how these audits function, whether library workers find them useful, and what is at stake when sensitive, normative decisions about representation are outsourced to automated commercial systems. Our analysis draws on an anonymous survey of U.S. public librarians (n=99), interviews with 14 librarians, a sample of purchasing records, and vendor documentation. We find that many library workers view these tools as convenient, time-saving solutions for assessing and diversifying collections under real and increasing constraints. Yet at the same time, the audits often flatten complex identities into standardized categories, fail to reflect local community needs, and further entrench libraries' infrastructural dependence on vendors. We conclude with recommendations for improving collection diversity audits and reflect on the broader implications for public libraries operating at the intersection of AI adoption, escalating anti-DEI backlash, and politically motivated defunding.

cs.CY↗

arXiv2025

PUBLICSPEAK: Hearing the Public with a Probabilistic Framework in Local Government

Local governments around the world are making consequential decisions on behalf of their constituents, and these constituents are responding with requests, advice, and assessments of their officials at public meetings. So many small meetings cannot be covered by traditional newsrooms at scale. We propose PUBLICSPEAK, a probabilistic framework which can utilize meeting structure, domain knowledge, and linguistic information to discover public remarks in local government meetings. We then use our approach to inspect the issues raised by constituents in 7 cities across the United States. We evaluate our approach on a novel dataset of local government meetings and find that PUBLICSPEAK improves over state-of-the-art by 10% on average, and by up to 40%.

cs.AI↗

arXiv2023

The Disparate Impacts of College Admissions Policies on Asian American Applicants

There is debate over whether Asian American students are admitted to selective colleges and universities at lower rates than white students with similar academic qualifications. However, there have been few empirical investigations of this issue, in large part due to a dearth of data. Here we present the results from analyzing 685,709 applications from Asian American and white students to a subset of selective U.S. institutions over five application cycles, beginning with the 2015-2016 cycle. The dataset does not include admissions decisions, and so we construct a proxy based in part on enrollment choices. Based on this proxy, we estimate the odds that Asian American applicants were admitted to at least one of the schools we consider were 28% lower than the odds for white students with similar test scores, grade-point averages, and extracurricular activities. The gap was particularly pronounced for students of South Asian descent (49% lower odds). We trace this pattern in part to two factors. First, many selective colleges openly give preference to the children of alumni, and we find that white applicants were substantially more likely to have such legacy status than Asian applicants, especially South Asian applicants. Second, after adjusting for observed student characteristics, the institutions we consider appear less likely to admit students from geographic regions with relatively high shares of applicants who are Asian. We hope these results inform ongoing discussions on the equity of college admissions policies.

cs.CY↗

arXiv2022

Doubly robust nearest neighbors in factor models

We introduce and analyze an improved variant of nearest neighbors (NN) for estimation with missing data in latent factor models. We consider a matrix completion problem with missing data, where the $(i, t)$-th entry, when observed, is given by its mean $f(u_i, v_t)$ plus mean-zero noise for an unknown function $f$ and latent factors $u_i$ and $v_t$. Prior NN strategies, like unit-unit NN, for estimating the mean $f(u_i, v_t)$ relies on existence of other rows $j$ with $u_j \approx u_i$. Similarly, time-time NN strategy relies on existence of columns $t'$ with $v_{t'} \approx v_t$. These strategies provide poor performance respectively when similar rows or similar columns are not available. Our estimate is doubly robust to this deficit in two ways: (1) As long as there exist either good row or good column neighbors, our estimate provides a consistent estimate. (2) Furthermore, if both good row and good column neighbors exist, it provides a (near-)quadratic improvement in the non-asymptotic error and admits a significantly narrower asymptotic confidence interval when compared to both unit-unit or time-time NN.

stat.ML↗

arXiv2021

Blocks as geographic discontinuities: The effect of polling place assignment on voting

A potential voter must incur a number of costs in order to successfully cast an in-person ballot, including the costs associated with identifying and traveling to a polling place. In order to investigate how these costs affect voting behavior, we introduce two quasi-experimental designs that can be used to study how the political participation of registered voters is affected by differences in the relative distance that registrants must travel to their assigned Election Day polling place and whether their polling place remains at the same location as in a previous election. Our designs make comparisons of registrants who live on the same residential block, but are assigned to vote at different polling places. We find that living farther from a polling place and being assigned to a new polling place reduce in-person Election Day voting, but that registrants largely offset for this by casting more early in-person and mail ballots.

stat.AP↗

arXiv2020

Fast Physical Activity Suggestions: Efficient Hyperparameter Learning in Mobile Health

Users can be supported to adopt healthy behaviors, such as regular physical activity, via relevant and timely suggestions on their mobile devices. Recently, reinforcement learning algorithms have been found to be effective for learning the optimal context under which to provide suggestions. However, these algorithms are not necessarily designed for the constraints posed by mobile health (mHealth) settings, that they be efficient, domain-informed and computationally affordable. We propose an algorithm for providing physical activity suggestions in mHealth settings. Using domain-science, we formulate a contextual bandit algorithm which makes use of a linear mixed effects model. We then introduce a procedure to efficiently perform hyper-parameter updating, using far less computational resources than competing approaches. Not only is our approach computationally efficient, it is also easily implemented with closed form matrix algebraic updates and we show improvements over state of the art approaches both in speed and accuracy of up to 99% and 56% respectively.

cs.LG↗

arXiv2020

IntelligentPooling: Practical Thompson Sampling for mHealth

In mobile health (mHealth) smart devices deliver behavioral treatments repeatedly over time to a user with the goal of helping the user adopt and maintain healthy behaviors. Reinforcement learning appears ideal for learning how to optimally make these sequential treatment decisions. However, significant challenges must be overcome before reinforcement learning can be effectively deployed in a mobile healthcare setting. In this work we are concerned with the following challenges: 1) individuals who are in the same context can exhibit differential response to treatments 2) only a limited amount of data is available for learning on any one individual, and 3) non-stationary responses to treatment. To address these challenges we generalize Thompson-Sampling bandit algorithms to develop IntelligentPooling. IntelligentPooling learns personalized treatment policies thus addressing challenge one. To address the second challenge, IntelligentPooling updates each user's degree of personalization while making use of available data on other users to speed up learning. Lastly, IntelligentPooling allows responsivity to vary as a function of a user's time since beginning treatment, thus addressing challenge three. We show that IntelligentPooling achieves an average of 26% lower regret than state-of-the-art. We demonstrate the promise of this approach and its ability to learn from even a small group of users in a live clinical trial.

cs.LG↗

arXiv2020

Streamlined Empirical Bayes Fitting of Linear Mixed Models in Mobile Health

To effect behavior change a successful algorithm must make high-quality decisions in real-time. For example, a mobile health (mHealth) application designed to increase physical activity must make contextually relevant suggestions to motivate users. While machine learning offers solutions for certain stylized settings, such as when batch data can be processed offline, there is a dearth of approaches which can deliver high-quality solutions under the specific constraints of mHealth. We propose an algorithm which provides users with contextualized and personalized physical activity suggestions. This algorithm is able to overcome a challenge critical to mHealth that complex models be trained efficiently. We propose a tractable streamlined empirical Bayes procedure which fits linear mixed effects models in large-data settings. Our procedure takes advantage of sparsity introduced by hierarchical random effects to efficiently learn the posterior distribution of a linear mixed effects model. A key contribution of this work is that we provide explicit updates in order to learn both fixed effects, random effects and hyper-parameter values. We demonstrate the success of this approach in a mobile health (mHealth) reinforcement learning application, a domain in which fast computations are crucial for real time interventions. Not only is our approach computationally efficient, it is also easily implemented with closed form matrix algebraic updates and we show improvements over state of the art approaches both in speed and accuracy of up to 99% and 56% respectively.

stat.ML↗

arXiv2020

Rapidly Personalizing Mobile Health Treatment Policies with Limited Data

In mobile health (mHealth), reinforcement learning algorithms that adapt to one's context without learning personalized policies might fail to distinguish between the needs of individuals. Yet the high amount of noise due to the in situ delivery of mHealth interventions can cripple the ability of an algorithm to learn when given access to only a single user's data, making personalization challenging. We present IntelligentPooling, which learns personalized policies via an adaptive, principled use of other users' data. We show that IntelligentPooling achieves an average of 26% lower regret than state-of-the-art across all generative models. Additionally, we inspect the behavior of this approach in a live clinical trial, demonstrating its ability to learn from even a small group of users.

cs.LG↗

arXiv2018

Personalizing Intervention Probabilities By Pooling

In many mobile health interventions, treatments should only be delivered in a particular context, for example when a user is currently stressed, walking or sedentary. Even in an optimal context, concerns about user burden can restrict which treatments are sent. To diffuse the treatment delivery over times when a user is in a desired context, it is critical to predict the future number of times the context will occur. The focus of this paper is on whether personalization can improve predictions in these settings. Though the variance between individuals' behavioral patterns suggest that personalization should be useful, the amount of individual-level data limits its capabilities. Thus, we investigate several methods which pool data across users to overcome these deficiencies and find that pooling lowers the overall error rate relative to both personalized and batch approaches.

cs.LG↗