arXiv subjects

David A. Broniatowski

Publications and source records attributed to David A. Broniatowski.

14 recordsLinked to original sources

arXiv2026

Why AI Governance Frameworks Are Hard to Adopt: A Role-Based Stress Test of the NIST AI RMF

AI governance frameworks can be known, used, and implemented in form without becoming governance in practice. This paper examines that problem through a role-based stress test of the NIST Artificial Intelligence Risk Management Framework (AI RMF) in consumer lending. We treat framework adoption as a governance translation problem: whether RMF language can become role-usable, cross-level, authority-connected governance over the AI system-in-use, rather than producing governance-looking artifacts. The study uses LLM-based role simulation as a structured analytic probe. We apply a 4 $\times$ 2 $\times$ 3 design across four organizational roles, two AI deployments, and three governance hard cases, producing 120 scored responses. Results show that local translation was not the main problem. Simulated actors generally understood their assigned roles and translated the RMF into local activity. The harder problem was whether that activity became governance value. Actor role was strongly associated with Cross-Level Governance Value, Authority Connection, Governance Translatability, and governance value. Deployment was strongly associated with Structural Fit: the RMF fit a bounded ML underwriting model more cleanly than a workflow-embedded LLM underwriting copilot. Risk reduction was harder still. It appeared only when governance value was present and Structural Fit was full, but neither condition was sufficient by itself. The paper contributes a diagnostic account of framework-based AI governance. Frameworks create value when they help organizations see, interpret, escalate, authorize, and correct risk in the AI system-in-use. They also create value when they reveal limits of governability under existing evidence paths, authority structures, and system boundaries.

cs.CY↗

arXiv2026

Platform architecture determines whether recommendation algorithms can shape information quality on social media

Social media platforms shape public discourse through two fundamental design choices that naturally co-occur in any field investigation: platform architecture, which defines what types of actors exist and how they interact, and recommendation algorithm, which determines what content is surfaced to users. Using agent-based simulation, we orthogonally manipulate both factors, exploring four prototypical architectures -- tree (e.g., Reddit), layered hierarchy (e.g., Facebook), network (e.g., Twitter), and complete graph (e.g., TikTok) -- and two algorithms: chronological (LIFO) and popularity-based (Hot). Drawing on prior theory that identifies and ranks canonical system architectures in terms of their flexibility we hypothesize that algorithmic effects on information spread and quality should be largest on the most flexible platforms and smallest on the most constrained ones. We find strong confirmation of this prediction. On tree-like platforms like Reddit, the algorithm has no detectable effect on information spread and quality. On layered hierarchies and networks like Facebook and Twitter, respectively, the Hot algorithm has modest positive effects on both the spread of information and its quality. On complete structures like TikTok, the Hot algorithm leads to a winner-take-all dynamics that has strong negative effects on both information spread and quality, making the relation between content quality and popularity unpredictable. These findings imply that architectural considerations are more powerful levers than algorithmic interventions for the design of healthy online spaces and public discourse. Platform reform efforts focused exclusively on algorithm choice may be insufficient on architecturally unconstrained platforms and unnecessary on architecturally constrained ones.

cs.SI↗

arXiv2023

Inductive detection of Influence Operations via Graph Learning

Influence operations are large-scale efforts to manipulate public opinion. The rapid detection and disruption of these operations is critical for healthy public discourse. Emergent AI technologies may enable novel operations which evade current detection methods and influence public discourse on social media with greater scale, reach, and specificity. New methods with inductive learning capacity will be needed to identify these novel operations before they indelibly alter public opinion and events. We develop an inductive learning framework which: 1) determines content- and graph-based indicators that are not specific to any operation; 2) uses graph learning to encode abstract signatures of coordinated manipulation; and 3) evaluates generalization capacity by training and testing models across operations originating from Russia, China, and Iran. We find that this framework enables strong cross-operation generalization while also revealing salient indicators$\unicode{x2013}$illustrating a generic approach which directly complements transductive methodologies, thereby enhancing detection coverage.

cs.LG↗

arXiv2022

Facebook's Architecture Undermines Vaccine Misinformation Removal Efforts

Misinformation promotes distrust in science, undermines public health, and may drive civil unrest. Vaccine misinformation, in particular, has stalled efforts to overcome the COVID-19 pandemic, prompting social media platforms' attempts to reduce it. Some have questioned whether "soft" content moderation remedies -- e.g., flagging and downranking misinformation -- were successful, suggesting that the addition of "hard" content remedies -- e.g., deplatforming and content bans -- is necessary. We therefore examined whether Facebook's vaccine misinformation content removal policies were effective. Here, we show that Facebook's policies reduced the number of anti-vaccine posts but also caused several perverse effects: pro-vaccine content was also removed, engagement with remaining anti-vaccine content repeatedly recovered to pre-policy levels, and this content became more misinformative, more politically polarised, and more likely to be seen in users' newsfeeds. We explain these results as an unintended consequence of Facebook's design goal: promoting community formation. Members of communities dedicated to vaccine refusal appear to seek out misinformation from multiple sources. Community administrators make use of several channels afforded by the Facebook platform to disseminate misinformation. Our findings suggest the need to address how social media platform architecture enables community formation and mobilisation around misinformative topics when managing the spread of online content.

cs.SI↗

arXiv2022

Knowledge-Augmented Language Models for Cause-Effect Relation Classification

Previous studies have shown the efficacy of knowledge augmentation methods in pretrained language models. However, these methods behave differently across domains and downstream tasks. In this work, we investigate the augmentation of pretrained language models with commonsense knowledge in the cause-effect relation classification and commonsense causal reasoning tasks. After automatically verbalizing ATOMIC2020, a wide coverage commonsense reasoning knowledge graph, and GLUCOSE, a dataset of implicit commonsense causal knowledge, we continually pretrain BERT and RoBERTa with the verbalized data. Then we evaluate the resulting models on cause-effect pair classification and answering commonsense causal reasoning questions. Our results show that continually pretrained language models augmented with commonsense knowledge outperform our baselines on two commonsense causal reasoning benchmarks, COPA and BCOPA-CE, and the Temporal and Causal Reasoning (TCR) dataset, without additional improvement in model architecture or using quality-enhanced data for fine-tuning.

cs.CL↗

arXiv2022

GisPy: A Tool for Measuring Gist Inference Score in Text

Decision making theories such as Fuzzy-Trace Theory (FTT) suggest that individuals tend to rely on gist, or bottom-line meaning, in the text when making decisions. In this work, we delineate the process of developing GisPy, an open-source tool in Python for measuring the Gist Inference Score (GIS) in text. Evaluation of GisPy on documents in three benchmarks from the news and scientific text domains demonstrates that scores generated by our tool significantly distinguish low vs. high gist documents. Our tool is publicly available to use at: https://github.com/phosseini/GisPy.

cs.CL↗

arXiv2022

Applying Word Embeddings to Measure Valence in Information Operations Targeting Journalists in Brazil

Among the goals of information operations are to change the overall information environment vis-á-vis specific actors. For example, "trolling campaigns" seek to undermine the credibility of specific public figures, leading others to distrust them and intimidating these figures into silence. To accomplish these aims, information operations frequently make use of "trolls" -- malicious online actors who target verbal abuse at these figures. In Brazil, in particular, allies of Brazil's current president have been accused of operating a "hate cabinet" -- a trolling operation that targets journalists who have alleged corruption by this politician and other members of his regime. Leading approaches to detecting harmful speech, such as Google's Perspective API, seek to identify specific messages with harmful content. While this approach is helpful in identifying content to downrank, flag, or remove, it is known to be brittle, and may miss attempts to introduce more subtle biases into the discourse. Here, we aim to develop a measure that might be used to assess how targeted information operations seek to change the overall valence, or appraisal, of specific actors. Preliminary results suggest known campaigns target female journalists more so than male journalists, and that these campaigns may leave detectable traces in overall Twitter discourse.

cs.CL↗

arXiv2021

Twitter and Facebook posts about COVID-19 are less likely to spread false and low-credibility content compared to other health topics

On February 2, 2020, the World Health Organization declared a COVID-19 social media "infodemic", with special attention to misinformation -- frequently understood as false claims. To understand the infodemic's scope and scale, we analyzed over 500 million posts from Twitter and Facebook about COVID-19 and other health topics, between March 8 and May 1, 2020. Following prior work, we assumed URL source credibility is a proxy for false content, but we also tested this assumption. Contrary to expectations, we found that messages about COVID-19 were more likely to contain links to more credible sources. Additionally, messages linking to government sources, and to news with intermediate credibility, were shared more often, on average, than links to non-credible sources. These results suggest that more ambiguous forms of misinformation about COVID-19 may be more likely to be disseminated through credible sources when compared to other health topics. Furthermore, the assumption that credibility is an adequate proxy for false content may overestimate the prevalence of false content online: less than 25% of posts linking to the least credible sources contained false content. Our results emphasize the importance of distinguishing between explicit falsehoods and more ambiguous forms of misinformation due to the search for meaning in an environment of scientific uncertainty.

cs.SI↗

arXiv2021

Predicting Directionality in Causal Relations in Text

In this work, we test the performance of two bidirectional transformer-based language models, BERT and SpanBERT, on predicting directionality in causal pairs in the textual content. Our preliminary results show that predicting direction for inter-sentence and implicit causal relations is more challenging. And, SpanBERT performs better than BERT on causal samples with longer span length. We also introduce CREST which is a framework for unifying a collection of scattered datasets of causal relations.

cs.CL↗

arXiv2020

A Multi-Modal Method for Satire Detection using Textual and Visual Cues

Satire is a form of humorous critique, but it is sometimes misinterpreted by readers as legitimate news, which can lead to harmful consequences. We observe that the images used in satirical news articles often contain absurd or ridiculous content and that image manipulation is used to create fictional scenarios. While previous work have studied text-based methods, in this work we propose a multi-modal approach based on state-of-the-art visiolinguistic model ViLBERT. To this end, we create a new dataset consisting of images and headlines of regular and satirical news for the task of satire detection. We fine-tune ViLBERT on the dataset and train a convolutional neural network that uses an image forensics technique. Evaluation on the dataset shows that our proposed multi-modal approach outperforms image-only, text-only, and simple fusion baselines.

cs.CL↗

arXiv2020

The Role of Individual User Differences in Interpretable and Explainable Machine Learning Systems

There is increased interest in assisting non-expert audiences to effectively interact with machine learning (ML) tools and understand the complex output such systems produce. Here, we describe user experiments designed to study how individual skills and personality traits predict interpretability, explainability, and knowledge discovery from ML generated model output. Our work relies on Fuzzy Trace Theory, a leading theory of how humans process numerical stimuli, to examine how different end users will interpret the output they receive while interacting with the ML system. While our sample was small, we found that interpretability -- being able to make sense of system output -- and explainability -- understanding how that output was generated -- were distinct aspects of user experience. Additionally, subjects were more able to interpret model output if they possessed individual traits that promote metacognitive monitoring and editing, associated with more detailed, verbatim, processing of ML output. Finally, subjects who are more familiar with ML systems felt better supported by them and more able to discover new patterns in data; however, this did not necessarily translate to meaningful insights. Our work motivates the design of systems that explicitly take users' mental representations into account during the design process to more effectively support end user requirements.

cs.CY↗

arXiv2020

The Twitter Social Mobility Index: Measuring Social Distancing Practices from Geolocated Tweets

Social distancing is an important component of the response to the novel Coronavirus (COVID-19) pandemic. Minimizing social interactions and travel reduces the rate at which the infection spreads, and "flattens the curve" such that the medical system can better treat infected individuals. However, it remains unclear how the public will respond to these policies. This paper presents the Twitter Social Mobility Index, a measure of social distancing and travel derived from Twitter data. We use public geolocated Twitter data to measure how much a user travels in a given week. We find a large reduction in travel in the United States after the implementation of social distancing policies, with larger reductions in states that were early adopters and smaller changes in states without policies. Our findings are presented on http://socialmobility.covid19dataresources.org and we will continue to update our analysis during the pandemic.

cs.SI↗

arXiv2020

Content analysis of Persian/Farsi Tweets during COVID-19 pandemic in Iran using NLP

Iran, along with China, South Korea, and Italy was among the countries that were hit hard in the first wave of the COVID-19 spread. Twitter is one of the widely-used online platforms by Iranians inside and abroad for sharing their opinion, thoughts, and feelings about a wide range of issues. In this study, using more than 530,000 original tweets in Persian/Farsi on COVID-19, we analyzed the topics discussed among users, who are mainly Iranians, to gauge and track the response to the pandemic and how it evolved over time. We applied a combination of manual annotation of a random sample of tweets and topic modeling tools to classify the contents and frequency of each category of topics. We identified the top 25 topics among which living experience under home quarantine emerged as a major talking point. We additionally categorized broader content of tweets that shows satire, followed by news, is the dominant tweet type among the Iranian users. While this framework and methodology can be used to track public response to ongoing developments related to COVID-19, a generalization of this framework can become a useful framework to gauge Iranian public reaction to ongoing policy measures or events locally and internationally.

cs.SI↗

arXiv2019

Identifying Nuances in Fake News vs. Satire: Using Semantic and Linguistic Cues

The blurry line between nefarious fake news and protected-speech satire has been a notorious struggle for social media platforms. Further to the efforts of reducing exposure to misinformation on social media, purveyors of fake news have begun to masquerade as satire sites to avoid being demoted. In this work, we address the challenge of automatically classifying fake news versus satire. Previous work have studied whether fake news and satire can be distinguished based on language differences. Contrary to fake news, satire stories are usually humorous and carry some political or social message. We hypothesize that these nuances could be identified using semantic and linguistic cues. Consequently, we train a machine learning method using semantic representation, with a state-of-the-art contextual language model, and with linguistic features based on textual coherence metrics. Empirical evaluation attests to the merits of our approach compared to the language-based baseline and sheds light on the nuances between fake news and satire. As avenues for future work, we consider studying additional linguistic features related to the humor aspect, and enriching the data with current news events, to help identify a political or social message.

cs.CL↗