Digital Commons Network^™

Bias Testing And Mitigation In Llm-Based Code Generation, Dong Huang, Jie M. Zhang, Qingwen Bu, Xiaofei Xie, Junjie Chen, Heming Cui

Research Collection School Of Computing and Information Systems

As the adoption of LLMs becomes more widespread in software coding ecosystems, a pressing issue has emerged: does the generated code contain social bias and unfairness, such as those related to age, gender, and race? This issue concerns the integrity, fairness, and ethical foundation of software applications that depend on the code generated by these models but are underexplored in the literature. This paper presents a novel bias testing framework that is specifically designed for code generation tasks. Based on this framework, we conduct an extensive empirical study on the biases in code generated by five widely studied LLMs (i.e., …

A Rate-Dependent Coreset Selector For Continual Learning On Time-Varying Data Distributions, Zilin Luo, Zichen Tian, Yaoyao Liu, Qianru Sun

Research Collection School Of Computing and Information Systems

In this paper we review the concept of “phase” defined in Class-Incremental Learning (CIL), i.e., learning new classes while not forgetting old ones. Due to this design, classic CIL algorithms are mostly offline or can handle only intensive data distribution shifts across the phases. However, real-world data streams are often online, usually with uncertain or untraceable changes in their data distributions. To this end, we design the per-step distribution shifts by modeling the class sampling weights using bell-shaped curves. Such a design respects the rise-and-fall nature and presents realistic but underexplored challenges for CIL: 1) The data non-stationarity across steps …

Genscore: Agent-Based Short-Answer Question Generation And Scoring In Software Engineering Courses, Nguyen Binh Duong Ta, Lwin Khin Shar

Research Collection School Of Computing and Information Systems

Short-answer questions are commonly used in educational assessments, as they are often viewed as a more effective way than multiple-choice questions to determine whether students have achieved the intended learning outcomes. However, manually creating appropriate questions targeting different cognitive levels such as those defined by the Bloom’s Taxonomy, and grading text answers from students are not trivial tasks for instructors. Existing work on auto-question generation and scoring in computing education typically targets coding-based questions. However, in software engineering courses, assessments can extend beyond coding to understanding of processes, DevOps methodologies, system design, etc. This work aims to address the dual …

Reliable-Data-Split (Rds): Maximizing Model Potential With Reinforced Selection Strategy, Hoang D. Nguyen, Xuan-Son Vu, Quoc Tuan Truong, Duc-Trong Le

Research Collection School Of Computing and Information Systems

The nexus between data characteristics and parametric models is fundamental for developing effective and reliable artificial intelligence (AI) systems. Mismatches in data properties for model development may lead to deleterious effects on AI model performance in machine learning practice. This paper proposes a Reliable Data Split (RDS) procedure to learn how to select data points that will generalise the target domain adequately by employing prior knowledge of the data generative process. We introduce a reinforced selection strategy using deep reinforcement learning with diverse black box predictors in maximising ensemble rewards as the proxy of model performance potential while maintaining an …

Immersion In Virtual Cultural Heritage: Framework, Research Landscape, Future Directions, And Ongoing Study, Xuenan Huo, Keng Siau

Research Collection School Of Computing and Information Systems

This study proposes that a nuanced understanding of immersion’sdistinct dimensions and their mediating role between technical de-sign features and cultural learning outcomes is critical for advancingVirtual Cultural Heritage (VCH) research and practice. To advancethe immersion designs in the field of VCH, the study provides anoverview of the immersion nuances and research state-of-the-artof VCH studies. A tripartite framework of spatial, emotional, andsocial immersion in virtual environments is articulated. A litera-ture review was conducted based on this framework. The reviewreveals multiple research gaps and opportunities, including absenceof immersion, predominant focus on spatial immersion, lagged AIintegration, lack of accessibility and inclusive designs, and …

Development Of A Handy Tool For The Selection Of Urban Stormwater Best Management Practices, Aaron T. Kenny

Civil & Environmental Engineering Theses & Dissertations

The City of Norfolk, Virginia faces substantial stormwater management challenges due to shallow groundwater, tidal influence, dense urban development, and limited right-of-way. These constraints limit the applicability of many Best Management Practices (BMPs) and require the early identification of feasible practices before detailed hydrologic modeling. This thesis introduces a decision-support tool that quickly and systematically identifies and prioritizes BMPs that are both feasible and well-suited to Norfolk’s Municipal Separate Storm Sewer System (MS4) program, streamlining early-stage selection and saving time and resources.

The tool implements a two-stage methodology. First, feasibility gates are applied using catalog attributes derived from the Virginia …

Marginalized Pre-Service Teachers’ Digital Competence: An Exploration Of Epistemic Beliefs And Lived Experiences, Jerolyn Rose Brown

STEMPS Theses & Dissertations

Amid growing calls for equity in teacher education, this dissertation investigates how digital competence and epistemological beliefs interact within the lived experiences of marginalized pre-service teachers (mPSTs), also referenced as teacher candidates, whose racial and ethnic identities often position them at the periphery of dominant educational narratives. Anchored in critical narrative inquiry (Pino Gavidia & Adu, 2022) and framed by the DigCompEdu digital competence model (Redecker & Punie, 2017), the study employs a multimethod approach, integrating quantitative data from the SELFIEforTEACHERS self-assessment tool (Economou, 2023) with qualitative insights drawn from in-depth narrative interviews. The participant cohort consists of 50 marginalized …

The Role Women Play As Facilitators And Barriers In The Career Trajectory Of Female Leaders, Anna Marie Smith

School of Public Service Theses & Dissertations

As gender diversity in the American workplace increases, the complexities surrounding the challenges faced by women continue to evolve. In recent years, this shift has led to an increase in women entering the workforce in record numbers, providing them with opportunities for successful careers, particularly with respect to leadership, all while balancing work and family life. However, despite the significant gains women have made in the employment arena, they continue to lag behind their male counterparts when it comes to leadership opportunities. While research is abundant on the characteristics of female leaders, what motivates them, and how they compare to …

Identifying Alcohol Intervention-Relevant Constructs For Greek Life: Examining Across Greek Membership And Gender Among College Students, Emily Susan Renzoni

Psychology Theses & Dissertations

Among college populations, students who are members of social Greek life organizations (i.e., social fraternities and sororities) drink more and experience higher rates of negative alcohol-related consequences than their non-affiliated peers. Although there has been success of alcohol interventions in general college populations, research has shown that college drinking interventions that are delivered to students in Greek life are not efficacious in reducing alcohol consumption or related consequences. It is meaningful to investigate if the content commonly addressed in these interventions, although well-documented as being linked to drinking in the general college population, has the same associations for students who …

Exploring The Impact Of Identity-Based Bullying On Suicidal Thoughts And Behaviors Among 6th-12th Grade Youth In Virginia, Camille N. Johnson

Psychology Theses & Dissertations

Suicidal thoughts and behaviors (STBs) are a growing public health concern among youth. Bullying victimization is a robust risk factor for STBs and identity-based bullying (IBB) or bullying that occurs due to a person’s actual or perceived social-identity/identities that is ingrained in discrimination, prejudice, or bias, has been found to be more strongly associated with STBs compared to bullying that was unrelated to identity. The current study used data from the Virginia School Survey of Climate and Working Conditions (N=399,493) to examine associations between general bullying, adult-perpetrated bullying, and IBB with depression and STBs among 6th-12th grade youth. This study …

Identity Salience Of Organizational Culture: The Intersection Of Racial And Occupational Identities In High-Risk Professions, Sarah Espinosa

School of Public Service Theses & Dissertations

Despite decades of effort, issues of race and color in organizations remain unresolved. Representative Bureaucracy (RB) has been a necessary but insufficient response to these challenges. While RB promotes diversity and inclusion, it struggles to address deeper issues of equity and justice, which are seen in the persistence of police brutality, even with increased representation of officers of color. This dissertation will propose a new theoretical approach to address the shortcomings of RB, a framework for the Identity Salience of Organizational Culture (ISOC). The ISOC framework seeks to build beyond RB to explain how the culture of high-risk professions with …

The Vampire In Video Games: Narrative Structures, Game Mechanics, And Cultural Significance, Ricardo Longo Minervino

Theses

This thesis examines how vampires are represented in video games by analyzing narrative structures, game mechanics, and cultural meaning. Using the existing games of Vampyr, V Rising, Vampire: The Masquerade – Bloodlines, The Sims 4: Vampires, Skyrim: Dawnguard, Baldur’s Gate 3, and Bloodlines 2, the study identifies which design strategies successfully create an engaging and believable vampire experience for the player, as well as determining which development decisions could hurt the experience for players. Using formalism, symbolism, feminist methodology, and player reception theory, this thesis argues that vampire games function best when players confront moral dilemmas, experience consequences for feeding …

Examining The Relations Between Parenting Stress, Harsh Parenting Practices, And Children’S Internalizing And Externalizing Symptoms, Caitlin A. Wharton

Individual, Family, and Community Education ETDs

This study examines: relations between maternal parenting stress in infancy and preschool-aged children’s internalizing and externalizing symptoms; whether verbal and physical harsh parenting practices mediate these relations; and whether these relations are moderated by child gender. Research links maternal parenting stress to harsh parenting, which is a risk factor for children’s internalizing and externalizing symptoms. But limited research has examined how maternal stress in infancy affects preschool-aged children’s outcomes, or how verbal and physical harsh parenting practices independently affect these outcomes. The study data came from the Future of Families & Child Wellbeing Study. Results indicate maternal parenting stress in …

Mando-Llm: Heterogeneous Graph Transformers With Large Language Models For Smart Contract Vulnerability Detection, Nhat Minh Nguyen, Huu Hoang Nguyen, Long Le Thanh, Zahra Ahmadi, Thanh Nam Doan, Daoyuan Wu, Lingxiao Jiang

Research Collection School Of Computing and Information Systems

Detecting vulnerabilities in smart contracts is vital for the security and reliability of decentralized apps. To facilitate vulnerability detection, contract codes, including bug patterns, are represented as heterogeneous graphs with various nodes and edges, like control-flow and function-call graphs. However, existing graph learning techniques struggle with large, complex graphs. This paper presents MANDO-LLM, a novel framework that combines heterogeneous graph transformers (HGTs) with large language models (LLMs) for detecting vulnerabilities in smart contracts represented as heterogeneous contract graphs built upon control-flow and call graphs. MANDO-LLM uses LLMs to capture code features from control-flow and call data, customizes HGTs to learn …

Evaluating Defi Vulnerabilities: The Role Of Bug Bounty Programs On Defi Software Supply Chain, Ping Fan Ke, Yi Meng Lau, Lingxiao Jiang

Research Collection School Of Computing and Information Systems

Decentralized finance (DeFi), powered by blockchain technology, enables peer-to-peer financial transactions without intermediaries. Despite rapid adoption, DeFi attracts malicious actors exploiting vulnerabilities. To mitigate risks, we propose a framework assessing entry points in the DeFi software supply chain: smart contracts, oracles/third-party feeds, user interfaces, off-chain storage, and crypto wallets. Applying this framework, we evaluate whether industry solutions—particularly bug bounty programs—adequately address these gaps. Our preliminary analysis indicates that most programs cover smart contract vulnerabilities (85.7%), followed by user interface issues (21.3%) and crypto wallet loopholes (11.9%). However, third-party risks, such as oracle feeds, are frequently deemed out of scope. This …

A Socio-Technical Analysis Of Market Reactions On Meme Coins: Trump’S Presidential Effect, Ping Fan Ke, Yi Meng Lau

Research Collection School Of Computing and Information Systems

Meme coins are a unique type of cryptocurrency whose value is shaped by internet culture and viral trends. This study introduces a socio-technical research model to examine key factors influencing meme coin dynamics and applies it to analyze market reactions to Donald Trump’s 2024 U.S. presidential election victory and inauguration, focusing on the $TRUMP meme coin and other politics-related meme coins, known as PolitiFi. Using a mixed-methods approach, we analyze publicly available news, social media activity, and marketplace data to investigate the interaction between social engagement and technical infrastructure. Econometric analysis shows that Trump-related events triggered short-term price surges, increased …

Jury-And-Judge Chain-Of-Thought For Uncovering Toxic Data In 3d Visual Grounding, Kaixiang Huang, Qifeng Zhang, Jin Wang, Jingru Yang, Yang Zhou, Huan Yu, Guodong Lu, Shengfeng He

Research Collection School Of Computing and Information Systems

3D Visual Grounding (3DVG) faces persistent challenges due to coarse scene-level observations and logically inconsistent annotations, which introduce ambiguities that compromise data quality and hinder effective model supervision. To address these challenges, we introduce Refer-Judge, a novel framework that harnesses the reasoning capabilities of Multimodal Large Language Models (MLLMs) to identify and mitigate toxic data. At the core of Refer-Judge is a Jury-and-Judge Chain-of-Thought paradigm, inspired by the deliberative process of the judicial system. This framework targets the root causes of annotation noise: jurors collaboratively assess 3DVG samples from diverse perspectives, providing structured, multi-faceted evaluations. Judges then consolidate these insights …

Misodice: Multi-Agent Imitation From Mixed-Quality Demonstrations, The Viet Bui, Tien Mai, Hong Thanh Nguyen

Research Collection School Of Computing and Information Systems

We study offline imitation learning (IL) in cooperative multi-agent settings, where demonstrations have unlabeled mixed quality - containing both expert and suboptimal trajectories. Our proposed solution is structured in two stages: trajectory labeling and multi-agent imitation learning, designed jointly to enable effective learning from heterogeneous, unlabeled data. In the first stage, we combine advances in large language models and preference-based reinforcement learning to construct a progressive labeling pipeline that distinguishes expert-quality trajectories. In the second stage, we introduce MisoDICE, a novel multi-agent IL algorithm that leverages these labels to learn robust policies while addressing the computational complexity of large joint …

No Experts, No Problem: Avoidance Learning From Bad Demonstrations, Minh Huy Hoang, Tien Mai, Pradeep Varakantham

Research Collection School Of Computing and Information Systems

This paper addresses the problem of learning avoidance behavior within the context of offline imitation learning. In contrast to conventional methodologies that prioritize the replication of expert or near-expert demonstrations, our work investigates a setting where expert (or desirable) data is absent, and the objective is to learn to eschew undesirable actions by leveraging demonstrations of such behavior (i.e., learning from negative examples).To address this challenge, we propose a novel training objective grounded in the maximum entropy principle. We further characterize the fundamental properties of this objective function, reformulating the learning process as a cooperative inverse Q-learning task. Moreover, we …

A Partition Cover Approach To Tokenization, Jia Peng Lim, Shawn Tan, Davin Choo, Hady Wirawan Lauw

Research Collection School Of Computing and Information Systems

Tokenization is the process of encoding strings into tokens of a fixed vocabulary size, and is widely utilized in Natural Language Processing applications. The leading tokenization algorithm today is Byte Pair Encoding (BPE), which formulates the tokenization problem as a compression problem and tackles it by performing sequences of merges. In this work, we formulate tokenization as an optimization objective, show that it is NP-hard via a simple reduction from vertex cover, and propose a polynomial-time greedy algorithm GreedTok. Our formulation naturally relaxes to the well-studied weighted maximum coverage problem which has a simple -approximation algorithm GreedWMC. Through empirical evaluations …

When Less Language Is More: Language-Reasoning Disentanglement Makes Llms Better Multilingual Reasoners, Weixiang Zhao, Jiahe Guo, Yang Deng, Tongtong Wu, Wenxuan Zhang, Yulin Hu, Xingyu Sui, Yanyan Zhao, Wanxiang Che, Bing Qin, Tat-Seng Chua, Ting Liu

Research Collection School Of Computing and Information Systems

Multilingual reasoning remains a significant challenge for large language models (LLMs), with performance disproportionately favoring high-resource languages. Drawing inspiration from cognitive neuroscience, which suggests that human reasoning functions largely independently of language processing, we hypothesize that LLMs similarly encode reasoning and language as separable components that can be disentangled to enhance multilingual reasoning. To evaluate this, we perform a causal intervention by ablating language-specific representations at inference time. Experiments on 10 open-weight LLMs spanning 11 typologically diverse languages show that this language-specific ablation consistently boosts multilingual reasoning performance. Layer-wise analyses further confirm that language and reasoning representations can be effectively …

Sheetpedia: A 300k-Spreadsheet Corpus For Spreadsheet Intelligence And Llm Fine-Tuning, Zailong Tian, Zhuoheng Han, Houfeng Wang, Lizi Liao

Research Collection School Of Computing and Information Systems

Spreadsheets are widely used for data analysis and reporting, yet their complex structure and formula logic pose significant challenges for AI systems. We introduce Sheetpedia, a large-scale corpus of over 290,000 diverse spreadsheets (from 324,000+ workbooks) compiled from enterprise email archives and online forums. We detail a rigorous collection and preprocessing pipeline (integrating the Enron email spreadsheet archive and the Fuse web corpus, plus a new crawl of Excel forums) to standardize formats, filter languages, and remove duplicates. Sheetpedia provides extensive coverage of real formulas and annotations – addressing a gap left by prior table datasets (e.g. web tables used …

Llm-Driven Semantic Explanations For Soil Moisture Prediction Models, Bamory Ahmed Toru Koné, Khouloud Boukadi, Rima Grati, Emna Ben Abdallah, Massimo Mecella

All Works

Efficient soil moisture prediction is crucial for sustainable agricultural practices, especially in the face of climate change and increasing water scarcity. However, the adoption of machine learning (ML) models in this context is frequently limited by their lack of interpretability, particularly among non-expert users such as farmers. This study proposes a novel approach to soil moisture prediction that combines high predictive performance with enhanced explainability. We propose a framework that leverages large language models (LLMs) to generate textual explanations based on a proposed irrigation and soil moisture ontology, thus making the model's predictions more understandable to farmers. The ontology formalizes …

Artificial Intelligence In Waste Management Systems: Applications, Challenges, And Prospects, Imane Belyamani

All Works

Despite global recognition of the climate crisis, greenhouse gas emissions are projected to rise by 8.8 % by 2030, primarily due to inadequate planning, poor implementation, and insufficient financial support. While international initiatives such as the ’Waste to Zero’ coalition launched at the 28th Conference of the Parties to the UNFCCC (COP 28) highlight the urgency of advancing decarbonization and the circularity of waste systems, this review focuses on how artificial intelligence (AI) can accelerate that transformation. It systematically explores the role of AI in advancing waste management practices, with a focus on predictive analytics, route optimization, and machine learning-based …

Introducing Statistical And Machine Learning-Based Methods Of Enhancing The Resiliency And Security Of Electrical-Based Critical Infrastructure, Joshua Tyler

Masters Theses and Doctoral Dissertations

Since its introduction into society in the late 1800’s, electricity has become a critical component in how society has functioned. High-voltage electricity provides power for appliances that have become integral to daily living, such as lights, refrigeration, and Heating, Ventilation, And Cooling (HVAC). Low-voltage electricity is used to process and transmit information on and between computers. This information may pertain to medical, financial, and defense-related activities. Over the past hundred years, significant research has been performed to improve electrical-based technology’s capability, scale, resiliency, and security. This study focuses solely on resiliency and security, both of which require efficient data collection, …

Vaping Insights Perceptions And Experiences On A College Campus, Willard G. Jackson

Masters Theses and Doctoral Dissertations

The purpose of this study was to determine if relationships existed between student vaping habits in proximity to a college campus and current tobacco related bans implemented by the institution. Determining these relationships provided insights that can be utilized to maximize effectiveness of campus tobacco policy. Research outcomes represented a framework for educating current and potential tobacco consumers on campus regarding harm reduction, tobacco cessation, and resources available to prevent a lifetime of nicotine addiction or other tobacco related harm to health. The study addressed research questions in a mixed-methods design. Quantitative research questions included: Does a relationship exist between …

An Evaluation Of The Effectiveness Of Living Learning Communities On Student Success, Retention, And Belonging, Jason D. Harville

Masters Theses and Doctoral Dissertations

This project of practice evaluated the impact of Living Learning Communities (LLCs) on student success, retention, and sense of belonging at a regional public university. The program evaluation includes quantitative data comparing GPA and retention rates of LLC and non-LLC students, a sense of belonging scale, and qualitative data gathered from interviews and surveys with LLC participants and program partners. Findings indicate that LLC participation positively correlates with increased student retention and a stronger sense of community, though academic performance differences were not statistically significant. Thematic analysis of qualitative interviews revealed that intentional faculty-student interaction, tailored academic support, and early …

Extraction Of Cellulose From Paper Towels For Sustainable Bio-Based Polyurethane Adhesives, Yashkumar J. Patel

Electronic Theses & Dissertations

The increasing consumption of paper towels generates significant waste, necessitating sustainable recycling solutions. This study investigates the extraction of cellulose from waste paper towels and its application in bio-based polyurethane (PU) adhesives. The extraction process includes alkali treatment, bleaching, and acid hydrolysis to obtain recycled cellulose (rCL). The extracted cellulose is then used in PU synthesis alongside castor oil polyol (COP) and methylene diphenyl diisocyanate (MDI), with performance compared to industrial cellulose-based PU. The obtained rCL and synthesized PU adhesives have been characterized with FT-IR spectra for structural confirmation. In addition, the synthesized PU adhesives demonstrated improved mechanical and thermal …

Separating The Control Of Moving And Holding In Human Post-Stroke Arm Paresis, Alkis M. Hadjiosif, Kahori Kita, Scott T. Albert, Robert A. Scheidt, Reza Shadmehr, John W. Krakauer

Biomedical Engineering Faculty Research and Publications

Moving and holding-still (holding) have been posited to be separately controlled. The paretic arm after stroke exhibits different abnormalities during rest vs. movement, providing an opportunity to ask whether control of these behaviors is independently affected in stroke. We quantified resting postural abnormalities in human stroke patients by measuring their biases in force production as they held their hand still in various locations in a planar workspace and then assessed whether these resting force biases influenced reaching in the same workspace. Patients displayed marked resting force biases at each location, even when the arm was supported. However, these …

Effective Transfer Of Tumor Annotations From Hematoxylin And Eosin To Fluorescence Images Of Breast And Lung Tissues, Tianling Niu, Emi Ampo, Julie M. Jorns, Mollie Patton, Tongtong Lu, Dong Hye Ye, Tina Yen, Bing Yu

Biomedical Engineering Faculty Research and Publications

Significance: Accurate transfer of annotations from histological images to fluorescence images is essential in developing deep learning (DL)-based optical imaging systems for intraoperative assessment of tumor margins. Manual annotation is time-consuming, prone to interobserver variability, and impractical for large-scale datasets.

Aim: We present a semi-automated method that can effectively transfer tumor annotations from pathologist-annotated hematoxylin and eosin (H&E) images to fluorescence images captured using microscopy with ultraviolet surface excitation (MUSE). This method is not intended for intraoperative use but rather to facilitate the creation of annotated datasets for DL model development.

Approach: Our semi-automated method consists of nonrigid image registration, …