arXiv subjects

Jevin D. West

Publications and source records attributed to Jevin D. West.

At least 19 recordsLinked to original sources

arXiv2026

Are Widely Known Findings Easier to Retract?

Failures of retraction are common in science. Why do they occur? And what determines whether a retraction is successful? We use data from citation records and Altmetrics to test proposed answers to these questions. LaCroix et al. employ network models to argue the social spread of information helps explain failures of retraction. One prediction is that widely known results, surprisingly, should be easier to retract, since their retraction is more relevant. Our results support this conclusion. We find highly cited papers show more significant reductions in citation after retraction and garner more attention to their retractions as they occur.

cs.DL↗

arXiv2026

Epistemic diversity across language models mitigates knowledge collapse

Artificial intelligence (AI) increasingly generates the very content used to train future AI systems. This feedback loop can degrade model quality, reduce informational diversity, and ultimately drive knowledge collapse, i.e. a degradation to a narrow and inaccurate set of ideas. We ask: to mitigate collapse, is it better to concentrate the internet's knowledge into a handful of dominant models (referred to as an AI monoculture), or to distribute it across a diverse ecosystem of models? To study the effect of diversity on model performance, we randomly segment the fixed training data across an increasing number of language models and evaluate the resulting ecosystems of models over ten self-training iterations. Our results show that diversity improves long-term performance of models, while monoculture accelerates collapse. Specifically, we observe that the optimal diversity level (i.e., the level that maximizes performance) increases monotonically with the number of self-training iterations. The observed effect is robust across various experimental settings, including different model families, parameter sizes, mixing human- and model-generated data, and temperature sampling methods, demonstrating the significance of ecosystem diversity for mitigating collapse. Moreover, our experiments with increased model and dataset sizes indicate that scaling up the system can amplify collapse in homogeneous ecosystems, thereby increasing the diversity benefits. In the presence of AI monoculture, our results suggest considering information environments with specialized AI models that maintain and enhance diversity in knowledge production, akin to the benefits of ecological diversity in biology and social systems.

cs.LG↗

arXiv2026

Deep Graph Learning will stall without Network Science

Deep graph learning focuses on flexible and generalizable models that learn patterns in an automated fashion. Network science focuses on models and measures revealing the organizational principles of complex systems with explicit assumptions. Both fields share the same goal: to better model and understand patterns in graph-structured data. However, deep graph learning prioritizes empirical performance but ignores fundamental insights from network science. Our position is that deep graph learning will stall without insights from network science. In this position paper, we formulate six Calls for Action to leverage untapped insights from network science to address current issues in deep graph learning, ensuring the field continues to make progress.

cs.LG↗

arXiv2025

RIP Twitter API: A eulogy to its vast research contributions

Since 2006, Twitter's APIs have been rich sources of data for researchers studying social phenomena such as misinformation, public communication, crisis response, and political behavior. However, in 2023, Twitter began heavily restricting data access, dismantling its academic access program, and setting the Enterprise API price at $42,000 per month. Lacking funds to pay this fee, academics are scrambling to continue their research. This study systematically tabulates the number of studies, citations, publication dates, disciplines, and major topics of research using Twitter data between 2006 and 2024. While we cannot know exactly what will be lost now that Twitter data is cost-prohibitive, we can illustrate its research value during the years it was available. A search of eight databases found that between 2006 and 2024, a total of 33,306 studies were published in 8,914 venues, with 610,738 citations across 16 disciplines. Major disciplines include social science, engineering, data science, and public health. Major topics include information dissemination, tweet credibility, research methodologies, event detection, and human behavior. Twitter-based studies increased by a median of 25% annually from 2006 to 2023, but following Twitter's decision to charge for data, the number of studies dropped by 13%. Much of the 2024 research likely used data collected before the API shutdown, suggesting further decline ahead. This trend highlights a growing loss of empirical insight and access to real-time, public communication-raising concerns about the long-term consequences for studying society, technology, and global events in an era increasingly connected by social media.

cs.CY↗

arXiv2025

The Role of Follow Networks and Twitter's Content Recommender on Partisan Skew and Rumor Exposure during the 2022 U.S. Midterm Election

Social media platforms shape users' experiences through the algorithmic systems they deploy. In this study, we examine to what extent Twitter's content recommender, in conjunction with a user's social network, impacts the topic, political skew, and reliability of information served on the platform during a high-stakes election. We utilize automated accounts to document Twitter's algorithmically curated and reverse chronological timelines throughout the U.S. 2022 midterm election. We find that the algorithmic timeline measurably influences exposure to election content, partisan skew, and the prevalence of low-quality information and election rumors. Critically, these impacts are mediated by the partisan makeup of one's personal social network, which often exerts greater influence than the algorithm alone. We find that the algorithmic feed decreases the proportion of election content shown to left-leaning accounts, and that it skews content toward right-leaning sources when compared to the reverse chronological feed. We additionally find evidence that the algorithmic system increases the prevalence of election-related rumors for right-leaning accounts, and has mixed effects on the prevalence of low-quality information sources. Our work provides insight into the outcomes of Twitter's complex recommender system at a crucial time period before controversial changes to the platform and in the midst of nationwide elections and highlights the need for ongoing study of algorithmic systems and their role in democratic processes.

cs.SI↗

arXiv2025

From job titles to jawlines: Using context voids to study generative AI systems

In this paper, we introduce a speculative design methodology for studying the behavior of generative AI systems, framing design as a mode of inquiry. We propose bridging seemingly unrelated domains to generate intentional context voids, using these tasks as probes to elicit AI model behavior. We demonstrate this through a case study: probing the ChatGPT system (GPT-4 and DALL-E) to generate headshots from professional Curricula Vitae (CVs). In contrast to traditional ways, our approach assesses system behavior under conditions of radical uncertainty -- when forced to invent entire swaths of missing context -- revealing subtle stereotypes and value-laden assumptions. We qualitatively analyze how the system interprets identity and competence markers from CVs, translating them into visual portraits despite the missing context (i.e. physical descriptors). We show that within this context void, the AI system generates biased representations, potentially relying on stereotypical associations or blatant hallucinations.

cs.CY↗

arXiv2024

Echo Chambers in the Age of Algorithms: An Audit of Twitter's Friend Recommender System

The presence of political misinformation and ideological echo chambers on social media platforms is concerning given the important role that these sites play in the public's exposure to news and current events. Algorithmic systems employed on these platforms are presumed to play a role in these phenomena, but little is known about their mechanisms and effects. In this work, we conduct an algorithmic audit of Twitter's Who-To-Follow friend recommendation system, the first empirical audit that investigates the impact of this algorithm in-situ. We create automated Twitter accounts that initially follow left and right affiliated U.S. politicians during the 2022 U.S. midterm elections and then grow their information networks using the platform's recommender system. We pair the experiment with an observational study of Twitter users who already follow the same politicians. Broadly, we find that while following the recommendation algorithm leads accounts into dense and reciprocal neighborhoods that structurally resemble echo chambers, the recommender also results in less political homogeneity of a user's network compared to accounts growing their networks through social endorsement. Furthermore, accounts that exclusively followed users recommended by the algorithm had fewer opportunities to encounter content centered on false or misleading election narratives compared to choosing friends based on social endorsement.

cs.SI↗

arXiv2024

Search Engines Post-ChatGPT: How Generative Artificial Intelligence Could Make Search Less Reliable

In this commentary, we discuss the evolving nature of search engines, as they begin to generate, index, and distribute content created by generative artificial intelligence (GenAI). Our discussion highlights challenges in the early stages of GenAI integration, particularly around factual inconsistencies and biases. We discuss how output from GenAI carries an unwarranted sense of credibility, while decreasing transparency and sourcing ability. Furthermore, search engines are already answering queries with error-laden, generated content, further blurring the provenance of information and impacting the integrity of the information ecosystem. We argue how all these factors could reduce the reliability of search engines. Finally, we summarize some of the active research directions and open questions.

cs.IR↗

arXiv2023

How should the advent of large language models affect the practice of science?

Large language models (LLMs) are being increasingly incorporated into scientific workflows. However, we have yet to fully grasp the implications of this integration. How should the advent of large language models affect the practice of science? For this opinion piece, we have invited four diverse groups of scientists to reflect on this query, sharing their perspectives and engaging in debate. Schulz et al. make the argument that working with LLMs is not fundamentally different from working with human collaborators, while Bender et al. argue that LLMs are often misused and over-hyped, and that their limitations warrant a focus on more specialized, easily interpretable tools. Marelli et al. emphasize the importance of transparent attribution and responsible use of LLMs. Finally, Botvinick and Gershman advocate that humans should retain responsibility for determining the scientific roadmap. To facilitate the discussion, the four perspectives are complemented with a response from each group. By putting these different perspectives in conversation, we aim to bring attention to important considerations within the academic community regarding the adoption of LLMs and their impact on both current and future scientific practices.

cs.CL↗

arXiv2022

Gender-based homophily in collaborations across a heterogeneous scholarly landscape

In this article, we investigate the role of gender in collaboration patterns by analyzing gender-based homophily -- the tendency for researchers to co-author with individuals of the same gender. We develop and apply novel methodology to the corpus of JSTOR articles, a broad scholarly landscape, which we analyze at various levels of granularity. Most notably, for a precise analysis of gender homophily, we develop methodology which explicitly accounts for the fact that the data comprises heterogeneous intellectual communities and that not all authorships are exchangeable. In particular, we distinguish three phenomena which may affect the distribution of observed gender homophily in collaborations: a structural component that is due to demographics and non-gendered authorship norms of a scholarly community, a compositional component which is driven by varying gender representation across sub-disciplines and time, and a behavioral component which we define as the remainder of observed gender homophily after its structural and compositional components have been taken into account. Using minimal modeling assumptions, the methodology we develop allows us to test for behavioral homophily. We find that statistically significant behavioral homophily can be detected across the JSTOR corpus and show that this finding is robust to missing gender indicators in our data. In a secondary analysis, we show that the proportion of women representation in a field is positively associated with the probability of finding statistically significant behavioral homophily.

stat.AP↗

arXiv2019

Delineating Knowledge Domains in the Scientific Literature Using Visual Information

Figures are an important channel for scientific communication, used to express complex ideas, models and data in ways that words cannot. However, this visual information is mostly ignored in analyses of the scientific literature. In this paper, we demonstrate the utility of using scientific figures as markers of knowledge domains in science, which can be used for classification, recommender systems, and studies of scientific information exchange. We encode sets of images into a visual signature, then use distances between these signatures to understand how patterns of visual communication compare with patterns of jargon and citation structures. We find that figures can be as effective for differentiating communities of practice as text or citation patterns. We then consider where these metrics disagree to understand how different disciplines use visualization to express ideas. Finally, we further consider how specific figure types propagate through the literature, suggesting a new mechanism for understanding the flow of ideas apart from conventional channels of text and citations. Our ultimate aim is to better leverage these information-dense objects to improve scientific communication across disciplinary boundaries.

cs.DL↗

arXiv2018

Why scatter plots suggest causality, and what we can do about it

Scatter plots carry an implicit if subtle message about causality. Whether we look at functions of one variable in pure mathematics, plots of experimental measurements as a function of the experimental conditions, or scatter plots of predictor and response variables, the value plotted on the vertical axis is by convention assumed to be determined or influenced by the value on the horizontal axis. This is a problem for the public understanding of scientific results and perhaps also for professional scientists' interpretations of scatter plots. To avoid suggesting a causal relationship between the x and y values in a scatter plot, we propose a new type of data visualization, the diamond plot. Diamond plots are essentially 45 degree rotations of ordinary scatter plots; by visually jarring the viewer they clearly indicate that she should not draw the usual distinction between independent/predictor variable and dependent/response variable. Instead, she should see the relationship as purely correlative.

cs.HC↗

arXiv2017

Men Set Their Own Cites High: Gender and Self-citation across Fields and over Time

How common is self-citation in scholarly publication, and does the practice vary by gender? Using novel methods and a data set of 1.5 million research papers in the scholarly database JSTOR published between 1779 and 2011, the authors find that nearly 10 percent of references are self-citations by a paper's authors. The findings also show that between 1779 and 2011, men cited their own papers 56 percent more than did women. In the last two decades of data, men self-cited 70 percent more than women. Women are also more than 10 percentage points more likely than men to not cite their own previous work at all. While these patterns could result from differences in the number of papers that men and women authors have published rather than gender-specific patterns of self-citation behavior, this gender gap in self-citation rates has remained stable over the last 50 years, despite increased representation of women in academia. The authors break down self-citation patterns by academic field and number of authors and comment on potential mechanisms behind these observations. These findings have important implications for scholarly visibility and cumulative advantage in academic careers.

physics.soc-ph↗

arXiv2016

Leveraging Citation Networks to Visualize Scholarly Influence Over Time

Assessing the influence of a scholar's work is an important task for funding organizations, academic departments, and researchers. Common methods, such as measures of citation counts, can ignore much of the nuance and multidimensionality of scholarly influence. We present an approach for generating dynamic visualizations of scholars' careers. This approach uses an animated node-link diagram showing the citation network accumulated around the researcher over the course of the career in concert with key indicators, highlighting influence both within and across fields. We developed our design in collaboration with one funding organization---the Pew Biomedical Scholars program---but the methods are generalizable to visualizations of scholarly influence. We applied the design method to the Microsoft Academic Graph, which includes more than 120 million publications. We validate our abstractions throughout the process through collaboration with the Pew Biomedical Scholars program officers and summative evaluations with their scholars.

cs.HC↗

arXiv2016

Static Ranking of Scholarly Papers using Article-Level Eigenfactor (ALEF)

Microsoft Research hosted the 2016 WSDM Cup Challenge based on the Microsoft Academic Graph. The goal was to provide static rankings for the articles that make up the graph, with the rankings to be evaluated against those of human judges. While the Microsoft Academic Graph provided metadata about many aspects of each scholarly document, we focused more narrowly on citation data and used this contest as an opportunity to test the Article Level Eigenfactor (ALEF), a novel citation-based ranking algorithm, and evaluate its performance against competing algorithms that drew upon multiple facets of the data from a large, real world dataset (122M papers and 757M citations). Our final submission to this contest was scored at 0.676, earning second place.

cs.DL↗

arXiv2016

Viziometrics: Analyzing Visual Information in the Scientific Literature

Scientific results are communicated visually in the literature through diagrams, visualizations, and photographs. These information-dense objects have been largely ignored in bibliometrics and scientometrics studies when compared to citations and text. In this paper, we use techniques from computer vision and machine learning to classify more than 8 million figures from PubMed into 5 figure types and study the resulting patterns of visual information as they relate to impact. We find that the distribution of figures and figure types in the literature has remained relatively constant over time, but can vary widely across field and topic. Remarkably, we find a significant correlation between scientific impact and the use of visual information, where higher impact papers tend to include more diagrams, and to a lesser extent more plots and photographs. To explore these results and other ways of extracting this visual information, we have built a visual browser to illustrate the concept and explore design alternatives for supporting viziometric analysis and organizing visual information. We use these results to articulate a new research agenda -- viziometrics -- to study the organization and presentation of visual information in the scientific literature.

cs.SI↗

arXiv2014

Memory in network flows and its effects on spreading dynamics and community detection

Random walks on networks is the standard tool for modelling spreading processes in social and biological systems. This first-order Markov approach is used in conventional community detection, ranking, and spreading analysis although it ignores a potentially important feature of the dynamics: where flow moves to may depend on where it comes from. Here we analyse pathways from different systems, and while we only observe marginal consequences for disease spreading, we show that ignoring the effects of second-order Markov dynamics has important consequences for community detection, ranking, and information spreading. For example, capturing dynamics with a second-order Markov model allows us to reveal actual travel patterns in air traffic and to uncover multidisciplinary journals in scientific communication. These findings were achieved only by using more available data and making no additional assumptions, and therefore suggest that accounting for higher-order memory in network flows can help us better understand how real systems are organized and function.

physics.soc-ph↗

arXiv2012

The role of gender in scholarly authorship

Gender disparities appear to be decreasing in academia according to a number of metrics, such as grant funding, hiring, acceptance at scholarly journals, and productivity, and it might be tempting to think that gender inequity will soon be a problem of the past. However, a large-scale analysis based on over eight million papers across the natural sciences, social sciences, and humanities re- reveals a number of understated and persistent ways in which gender inequities remain. For instance, even where raw publication counts seem to be equal between genders, close inspection reveals that, in certain fields, men predominate in the prestigious first and last author positions. Moreover, women are significantly underrepresented as authors of single-authored papers. Academics should be aware of the subtle ways that gender disparities can appear in scholarly authorship.

physics.soc-ph↗