University of Illinois at Chicago
University of Illinois at Chicago: UIC INDIGO (INtellectual property in DIGital form available online in an Open environment)Not a member yet
21439 research outputs found
Sort by
Partisan or Principled? Explaining Political Differences in Attitudes About Democratic Norm Violations
No full textDemocracy in the United States faces significant challenges, with public tolerance for undemocratic actions raising concerns about partisan interests undermining democratic norms. This research examined the psychological mechanisms underlying support for such actions and whether they differ between Republicans and Democrats. Three studies (N = 3,149) tested competing explanations for partisan differences by presenting scenarios where violations of democratic norms either benefited, harmed, or had no impact on respondents’ political party. Across studies, partisans rationalized politically advantageous violations as more democratic and opposed them as less than neutral or harmful actions. Republicans supported vote-by-mail restrictions more than Democrats, even without partisan benefit. That said, both parties strongly opposed a different violation of voter access. These findings suggest that although Americans broadly support democratic principles, partisan interests and differing interpretations of specific democratic practices shape their differences in opposition to undemocratic actions
Low-resource Multi-grained Natural Language Understanding: English and Beyond
No full textLow-resource settings, in which intelligent systems constantly face emerging knowledge beyond their initial learning, are inevitable when developing intelligent systems. Only by extracting the true semantic understanding of the linguistic inputs can these systems prevail when little knowledge is provided. This persistent challenge hampers the ability of intelligent systems to excel in the emerging essential tasks. In reality, low-resource can occur on different granularities of the textual understanding: (1) coarse-grained on the sentence-level, (2) fine-grained on the token-level, or both.
Throughout this manuscript, we address the issues of low-resource settings in Natural Language Understanding (NLU) across multiple granularities. First, we tackle the challenges of low-resource coarse-grained annotations by introducing dynamic semantic extraction together with multi-perspective matching and aggregation networks. Secondly, we address the concerns of unavailable fine-grained annotations and explore the potentials of inducing such information without the need of token-level supervised training by extracting and refining the preserved knowledge existent in generic-purpose language models with additional multi-level contrastive learning objectives. Third, we overcome the challenges of low-resource multi-grained annotations by reinforcing the interconnections of different granularities via coarse-to-fine chain-of-thought reasoning and structured knowledge from Abstract Meaning Representation Graph. Finally, we broaden the scope of low-resource NLU challenges beyond English, focusing on the cross-lingual transfer towards low-resource languages through the novel phonemic transcription integration beyond the textual scripts. Our work leverages publicly available datasets catering for both Task-oriented Dialogue Systems (SNIPS, NLUE, ATIS, MTOP, MASSIVE) in conjunction with the open-source comprehensive generic-purpose multilingual NLU benchmark datasets such as XTREME
Novel Antenna Designs for Wireless Information and Power Transfer Systems
No full textThe global machine-to-machine (M2M) connections market is expected to reach 5.3 billion units by 2029, growing at a compound annual growth rate of 12.7% during the forecast period. Due to the high demand for massive-device systems, rectifying antenna (rectenna) technology has emerged as an effective and sustainable solution for harvesting energy from ambient electromagnetic waves. This technology powers low-power ubiquitous IoT devices and wireless sensors connected via 5G and beyond 5G (B5G) wireless ecosystems, especially in environments where it is difficult or impossible to change batteries and where the exact location of the devices may be unknown. The receiving terminals are powered through an antenna that receives incident electromagnetic waves in the gigahertz frequency range, couples the energy to a rectifier circuit, which charges a storage device through an efficient power management circuit, and powers the entire low-power terminal platform. For low incident power density levels, co-design of the RF powering and power management circuits is required for optimal performance. In this thesis, we propose new paradigms for low-power far-field wireless power and information transfer devices for IoT and wireless sensor systems. In these devices, the wireless data and power transfer modules can be integrated into a single device. We will design, model, fabricate, and characterize various wireless data and power transfer devices, including but not limited to: (i) compact, low-profile transparent antennas for multiband and multirange wireless power transfer, (ii) wideband simultaneous wireless information and power transfer devices in the IMS band, (iii) compact, wide-angle, high-gain, and bandwidth-enhanced rectennas for radiative energy harvesting systems, and (iv) travelling-wave-antenna-enabled wide angular and multi-polarization wireless information and power transfer. This research is expected to pave the way for next-generation energy harvesting systems that can be connected via IoT and wireless sensor networks
High-Order Spectral Simulation of Dispersive Two-Dimensional Materials
No full textOver the past twenty years, the field of plasmonics has been revolutionized with the isolation and utilization of two--dimensional materials, particularly graphene. Consequently there is significant interest in rapid, robust, and highly accurate computational schemes which can incorporate such materials. Standard volumetric approaches can be contemplated, but these require huge computational resources. Here we describe an algorithm which addresses this issue for nonlocal models of the electromagnetic response of graphene. Our methodology not only approximates the graphene layer with a surface current, but also reformulates the governing volumetric equations in terms of surface quantities using Dirichlet--Neumann Operators. We have recently shown how these surface equations can be numerically simulated in an efficient, stable, and accurate fashion using a High--Order Perturbation of Envelopes methodology. We extend these results to the nonlocal model mentioned above, and using an implementation of this algorithm, we study absorbance spectra of TM polarized plane--waves scattered by a periodic grid of graphene ribbons
Monitoring Salivary Immune Profiles in Response to Nonsurgical Periodontal Therapy
No full textSaliva is increasingly recognized as a valuable diagnostic fluid for monitoring oral health status due to its ease of non-invasive collection and rich biomarker content. Limited evidence exists regarding saliva’s ability to aid in assessing the immune activity in periodontal disease (PD) and its resolution. Precise immune subset identification associated with PD pathology may facilitate development of novel diagnostic or prognostic cellular markers. This study examined the prognostic value of saliva in periodontal disease progression in response to non-surgical therapy, by monitoring Myeloid (Macrophages/Monocytes) and Lymphoid (B Cells) immune subsets. This ex-vivo, case-control study focused on salivary immune profiling in human subjects before and after NSPT (n=29/group). Periodontally healthy (n=12) samples from age- and gender-matched controls were collected from subjects presenting for crown lengthening and soft tissue augmentation procedures. Clinical parameters including pocket probing depth (PPD), bleeding on probing (BOP), & plaque index (PI) were measured pre- & post-NSPT. Salivary immune cells were evaluated for flow cytometry and GCF cytokines were quantified by multiplex. Our results show significant reductions in the PPD (6.67x10-5), BOP (1.24x10-8), PI (3.05x10-5), and clinical parameters post-NSPT in PD subjects compared to healthy controls. Assessment of the cellular immune mediators revealed decreased levels of the M1 (CD11b+CD14+HLA-DR+IFNγ+) macrophages and IFNγ+CD19+ B cell populations in PD subjects following NSPT. On the contrary, higher levels of M2 macrophage (CD11b+CD14+HLA-DR+IL-10+) and CD19+IL-10 B regulatory cells was observed in post-NSPT. This correlates with the clinical parameters observed following NSPT. Overall, the reduction in the M1 macrophage and B cell subsets and the converse increase in the regulatory M2 macrophage and B cells correlates with the clinical improvement in periodontal disease. Therefore, saliva can provide relevant clinical information regarding the response of periodontal therapy
Ergonomic Redesign of the Needle Driver for Quality Compliance and Comfortable Surgical Performance
No full textThe needle driver is a critical surgical instrument used in suturing procedures. However, traditional designs present ergonomic challenges that can lead to discomfort, inefficiency and potential musculoskeletal disorders(MSDs) among surgeons. This study focuses on identifying these limitations and proposing a redesigned needle driver that enhances usability while maintaining quality compliance.
Through qualitative research conducted at the University of Illinois Chicago, feedback from medical and surgical residence was collected via surveys, focus groups and interviews. The findings highlighted several key issues, including proper tool fit, hand fatigue, difficulty in tool handling and safety concerns related to needle slippage and improper grip. Additionally, left-handed surgeons face challenges using right-handed instruments and variations in hand sizes or often overlooked in existing designs.
To address these concerns, the study proposes an ergonomic redesign featuring adjustable ring handles and rod lengths to accommodate different hands lengths and improve grip stability. The redesign incorporates bayonet mount mechanism, allowing for secure and easy adjustments without compromising the stability of the instrument. This new design aims to reduce hand strain, enhance precision and improve overall surgical efficiency.
Despite its advantages, the redesigned needle driver presents challenges such as potential mechanical failures in threaded components and contamination risks exposed areas. Future work will focus on validating the design using biomechanics simulation software and 3D hand scans to ensure effectiveness in reducing hand fatigue and enhancing surgical performance. The proposed innovation aims to set a new standard for ergonomic and high-performance surgical tools
Perceptions of the Cardiovascular Physical Examination Among Academic Clinician Educators in Cardiology
No full textThe cardiovascular physical examination (CPE) is a symbolic and valuable component of clinical cardiology practice. Cardiologists working at academic hospitals work alongside trainees, who observe their preceptors as they form their own professional identities. Prior publications describe a waning emphasis on the CPE as imaging technology improves and time at the
patient’s bedside decreases. As academic cardiologists are uniquely positioned within this professional evolution, our aim was to investigate their behaviors and philosophies related to the CPE in an effort to understand the resulting impact on their cardiology practice, teaching, and learner assessment.
Fourteen cardiologists from academic centers in Chicago (4 women, 8 men) participated in virtual interviews. Their responses were transcribed and organized through an open and focused coding process.
Participants’ responses were focused in two areas, professional conflict and individual emotions. They believed that the technique necessary to perform the CPE requires repetition and feedback, which they strived to emphasize in their workplace culture. Subjects also practice using a Hypothesis-Driven Exam approach, where a patient’s symptoms prime the examiner to look for
specific findings. Lastly, their professional conflict manifested as a struggle between recognition of the changing environment and the responsibility to uphold the CPE as an endeavor worth teaching and promoting
Evaluating and Mitigating Bias in Large Language Models and Retrieval-Augmented Generation Systems
No full textSocial biases in large language models (LLMs) raise critical fairness concerns; this work addresses them through two complementary projects. The first project addresses the challenge of mitigating gender bias without compromising the language modeling capabilities of LLMs. Counterfactual Data Augmentation (CDA), a widely used method for fine-tuning, generates synthetic data that may poorly align with real-world distributions or ignore the social context of altered sensitive attributes (e.g., gender) in the pretraining corpus. To overcome this, we propose Context-CDA, which leverages LLMs to produce contextually relevant and diverse counterfactual data for fine-tuning. By minimizing distributional discrepancies between the debiasing corpus and the pre-training data, this approach enhances alignment with real-world usage. To further improve data quality, we implement semantic entropy filtering to remove uncertain text samples. Evaluations on bias benchmarks demonstrate that Context-CDA significantly reduces bias while preserving model performance.
The second project focuses on bias propagation in Retrieval-Augmented Generation (RAG) Systems. We investigate how the addition of retrieved contexts influences the bias behavior of LLMs. Our findings reveal a reduction in bias after incorporating the RAG pipeline, indicating that the inclusion of external context often helps counteract stereotype driven predictions. We also delve deeper into understanding the model’s reasoning process by integrating Chain-of-Thought (CoT) prompting into the RAG system while assessing faithfulness of the model’s CoT. Our experiments reveal that the model’s bias inclination shifts between stereotype and anti-stereotype responses as more contextual information is incorporated. Also, contrary to the bias reduction observed with standard RAG, we find that applying CoT with RAG increases overall bias across datasets. This counterintuitive result can be attributed to the bias-accuracy trade-off. While CoT improves accuracy by encouraging more deliberate reasoning, this often comes at the expense of fairness, thereby advocating for the design of bias-aware reasoning frameworks to mitigate this trade-off.
Public datasets used: Statistical and neural machine translation news commentary, StereoSet, CrowS-Pairs, Multi-Genre Natural Language Inference, Natural Language Inference Bias, Semantic textual similarity benchmark, BiasBios, Question-answering Natural Language Inference, Recognizing Textual Entailment, Stanford Sentiment Treebank v2, Winogender, WinoBias, WikiText-103, Colossal Clean Crawled Corpus, Bias in Open-ended Language Generation, Holistic Bia
Understanding How Disabled Persons Negotiate in the Practice of Informal Family Support
No full textThis study filled a gap in disability theory related to persons with physical disability (herein PWPD)’s negotiations for informal support in the complex, power-imbalanced relationship of family. To summarize the problem under study:
•For some PWPD, support is both formal and informal;
•This means PWPD negotiate either as consumer, as ‘care recipient’ in hands of family who define ability, or both;
•The unequal power and conflict in the relationship between PWPD and unpaid family support person is rarely investigated, and there’s a theory gap;
•Disabled people know: “When you ask people to help, you know there’s kind of a limit.” – Alice Wong, disability advocate; and
•Does this relationship build social capital that is becoming the replacement trend for formal support?
While disability studies has advanced theory accounting for support as ‘a gift,’ there is a ‘social debt’ narrated by disabled persons when asking for support. Disability studies has yet to critically examine and theorize ‘social debt,’ or conditions under which reciprocity in family relationships presumed to be capital-creating act as debt-relieving. As policy trends seek to ‘operationalize’ disabled persons’ social capital and networks in place of formal supports, it is important to theorize how social networks can reproduce unequal relations of power and capital. This study centers the voices of PWPD negotiating for informal supports in family relationships to understand the process of negotiating in order to advance theory on this social process. This study aligns a critical research lens with Constructivist Grounded Theory (herein “CGT”), using qualitative methodology with participant-driven data collection via Flip-based video narrative diaries followed by in-person individual interview(s) of nine participants, born between 1942 and 1999, spanning life course stages. The study explores three main areas: (1) the content included by PWPD when constructing narratives about informal supports; (2) the principles that guide, techniques employed, and value sought by PWPD when negotiating for informal supports compared to formal support; and (3) the capital value of social relationships that have become the substitute for formal supports. Results indicate self-structuring a narrative diary about informal supports is valued and valuable, that PWPD’s negotiations for informal family support are a dynamic-dependent constellation of contingencies, steeped in feelings that include indebtedness, blurred lines between informal and formal acts, near preclusion of a ‘no-deal’ option, and deep gratitude. These aspects coincide with voiced preference for control through formal supports, but did not usually coincide with discontinuing the adult informal support relationship absent an alternative. An exception to seeking informal support is risk of harm to valued relationships, to aging parents, or the disabled person for violation, minimization, or interruption of their transition to adulthood. Finally, most acts of reciprocity served a debt-relieving intention. From a careful explanation of specific events and actions of PWPD emerges an interpretive, grounded theoretical construct “social debt” that cuts across the life course (Charmaz, 2014). Results of this research have implication for expanding research to use participant-led data gathering to uncover hard-to-see structures, to understanding underlying structures of PWPD’s access to satisfying supports, and to help impact policies implemented in the name of economic reason to remove rather than build structural inequality barriers to formal supports for persons with physical disabilities
Hip Hop DJs as Technocultural Signifiers: Participatory Culture, Labor, and the Affordances of Twitch.tv
No full textThis study explored how Hip Hop DJs creatively and communally adopted Twitch.tv as a venue in 2020, at the onset of the COVID-19 pandemic, to sustain their craft and to continue their work within the gig economy. The dissertation investigated the DJ’s transition from performing in physical venues to exploring and navigating the Twitch digital economy by creating their own channels, leveraging the platform’s affordances, and relying on the support of
the communities they formed. This dissertation positions DJs as technocultural cultural signifiers who reshaped their labor through strategic creativity.
The research requestions driving this study were:
1. How do Hip Hop DJs use Twitch to engage with audiences?
2. How do the platform specific affordances of Twitch enable and/or constrain forms
of DJs’ engagement and participation within the gig economy?
3. What connections, disruptions, and other movements exist between forms of
participation supported by Twitch and those supported by the many cultural
practices of Hip Hop?
Methodologically, this study employed a triangulation of critical technocultural discourse analysis, observations of 43 DJs, and interviews of 30 DJs who performed on Twitch. The relevant theoretical frameworks to guide this study included affordance theory and vernacular affordances. Additionally, this research engaged in literature regarding the gig economy, participatory culture, and emotional, relational, and visible labor.
The analysis of the platform and of the Hip Hop DJs who engaged on Twitch resulted in
several key findings including: Hip Hop DJs experienced a relational dynamic shift, Twitch affordances enabled DJs to engage in performative capitalism, and Hip Hop DJs participated in technocultural labor to establish themselves on a platform where their work was initially
prohibited.
This study concluded that although DJs were participating on a masculine-dominated
interface, their engagement, and presence disrupted hegemonic systems. The Hip Hop DJ
community navigated Twitch’s habitus as a counterculture, ultimately leading to the platform’s integration of the DJ category. This acknowledgment solidified the Hip Hop DJ community’s presence on Twitch and within its digital economy
Data sources:
Access Repository Dashboard
Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇