Digital Commons Network^™

Parent Cultural Wealth: How Beliefs, Curriculum Familiarity, And Community Awareness Facilitate Home-Based Cs Conversations, Nicol Howard, Leiny Y. Garcia, Jean Ryoo, Julie Flapan, Michelle Choi, Paula Nazario, Wei Wei

Mathematics, Physics, and Computer Science Faculty Articles and Research

Parents are critical influences on children's computer science (CS) pathways, yet families’ involvement remains understudied. Guided by parental involvement and community cultural wealth frameworks, this work-in-progress examined survey data from 53 parents representing families historically underrepresented in computing within California (71% women; 54% Hispanic/Latin/x/e). Findings revealed that CS school curriculum familiarity and awareness of CS community events significantly predicted parent-child CS conversations, while parent CS confidence and attitudes did not, suggesting that curriculum transparency and community connections can promote equitable involvement. Future work will integrate qualitative interviews to examine how families draw upon their cultural wealth.

A Comprehensive Review Of Optimization Techniques For Healthcare Using Cancer Datasets, Dina Tbaishat, Mohammad Tubishat, Sharif Naser Makhadmeh, Mohammed Azmi Al-Betar

All Works

This study presents a systematic review of metaheuristic optimization techniques applied to healthcare problems using cancer datasets. A structured search of recently published peer-reviewed literature was carried out, focusing on five major application areas: feature selection, classification, image segmentation, hyperparameter tuning, and early detection. For each eligible study, the optimization strategy, dataset characteristics, data modality, learning model, validation protocol, and reported outcomes are provided. The reviewed works were organized into a taxonomy of original, modified, and hybridized algorithms, and a descriptive analysis was performed to assess algorithm prevalence and dataset utilization. The findings highlight that feature selection remains the most …

Flip It And Reverse It: Probing Consistency Of Llm World Knowledge Under Logical Transformations, Rhianna Smith

Computer Science Senior Theses

When assessing the extent to which someone truly understands the topics they discuss, you might ask them a series of questions that examine their knowledge from different angles. As Large Language Models (LLMs) are increasingly integrated into real-world tasks, they are often required to reason through ambiguous and uncertain scenarios that rarely have a single correct answer. In these settings, their usefulness is contingent on more than accuracy alone. Rather, it depends on the strength of the reasoning that underlies their responses. This begs the question: how might we extend these types of tests to LLMs? We formalize this informal …

Mimic: A Multimodal Dataset For Affective Incongruity In Video, Ethan M. Baird

Computer Science Senior Theses

This thesis investigates the computational challenges of constructing a database for affective incongruity, exploring the difficulties in automating the collection of contradictory affective states. Capturing these incongruities is essential for advancing Vision-Language Models (VLMs) and sentiment analysis, which struggle to interpret signals deviating from basic emotional archetypes. Curating non-congruent affect provides the data necessary for models to navigate complex social contexts, which is critical for applications like automated Audio Description (AD) for the visually impaired and nuanced Human-Computer Interaction (HCI). To capture these signals, two distinct methodologies were employed. The first utilized a tripartite decomposition of video data, isolating textual, …

Representing Lean Proofs: Tactics, Trajectories, Search, Elisaveta Samoylov

Computer Science Senior Theses

In neural theorem proving for interactive proof assistants such as Lean, tactic prediction models score proof steps by surface likelihood, missing whether a move actually advances the proof. This thesis proposes representing each step by the symbolic edit it induces on the proof state rather than by its token-level surface form, and shows that this effect-grounded view yields better tactic embeddings, reveals structured geometry in complete proofs, and enables a practical proof search prior.

We introduce Delta Tokens — token-level state edits augmented with structural indicators — and show they outperform surface-only representations on tactic retrieval and operator-analogy benchmarks. Embedding …

Ai, Translation, And Telling The Truth, David I. Smith

University Faculty Publications and Creative Works

I am working on a large translation project this year. I have been surprised to find several conversation partners voicing the assumption that I am getting AI to do the translating for me. I’ve been wondering how to respond.

A short, but in the end inadequate answer is that, impressive as the current variations on machine translation are, they still get things wrong. Neural machine translation services such as Google Translate and DeepL still produce oddities fairly regularly. I have been working lately with seventeenth-century Czech texts, an area in which I would expect machine translation to struggle a little …

Olaf: Towards Robust Llm-Based Annotation Framework In Empirical Software Engineering, Mia Mohammad Imran, Tarannum Shaila Zaman

Computer Science Faculty Research & Creative Works

Large Language Models (LLMs) are increasingly used in empirical software engineering (ESE) to automate or assist annotation tasks such as labeling commits, issues, and qualitative artifacts. Yet the reliability and reproducibility of such annotations remain underexplored. Existing studies often lack standardized measures for reliability, calibration, and drift, and frequently omit essential configuration details. We argue that LLM-based annotation should be treated as a measurement process rather than a purely automated activity. In this position paper, we outline the Operationalization for LLM-based Annotation Framework (OLAF), a conceptual framework that organizes key constructs: reliability, calibration, drift, consensus, aggregation, and transparency. The paper …

Scientific Crosstalk: Natural Language Processing, Praveshika Bhandari

Theses and Dissertations

While sentiment analysis has made significant strides in domains such as social media and personal correspondence, its application to formal scientific writings remains under-explored. The crosstalk between emotional expressions in personal and professional communications has also received limited attention despite its potential to reveal insights into the emotional drivers of scientific creativity. Our research introduces a computational framework designed to detect and quantify emotional expressions across various documents over time. Leveraging state-of-the-art transformer models fine-tuned on domain-specific corpora, the framework models emotional tone distribution. Integrating emotion analysis with knowledge graph modeling enables the exploration of emotional trends alongside key scientific …

The Security Of Llm-Generated Code, Christopher Brian Gonzalez Ayala

Student Theses

The rapid adoption of Large Language Models (LLMs) in software development has transformed coding practices by enabling automated code generation, completion, and optimization. Despite these advantages, concerns persist regarding the security and reliability of LLM-generated code. This study presents a comprehensive evaluation of both the functional correctness and security of code produced by three prominent LLMs as of early 2026. A total of 4,800 code snippets were generated using 100 security-focused programming prompts derived from the OWASP Top 10:2025, translated across eight natural languages and two phrasing styles (literal and natural developer-oriented prompts). To assess performance, a multi-stage experimental framework …

When Saying "No" Is Not Enough: Cognitive-Action Decoupling And The Illusion Of Safety In Llm Agents, Shasha Yu

School of Professional Studies

Current safety evaluations of large language models (LLMs) predominantly rely on textual compliance, implicitly assuming that refusal-style responses correspond to safe behavior. This assumption becomes fragile when LLMs are embedded in agentic systems with the ability to execute state-changing actions. In this paper, we present an empirical critique of text-centric safety evaluation through an action-aware study of LLM agents under controlled conditions. Across multiple state-of-the-art models, we observe a recurring cognitive-action decoupling: agents generate policy-aligned refusal language while still producing unsafe tool-mediated action proposals. This produces an illusion of safety, where conversational audits indicate compliance even as operational risk persists. …

Trace: Temporal Rhetorical Analysis And Consistency Evaluation For Legislative Speech, David Hernandez

Master's Theses

Legislators frequently discuss the same policy issues across multiple hearings and legislative sessions, sometimes maintaining consistent positions and other times modifying or reframing their stance over time. Understanding how these positions evolve is important for analyzing political discourse and democratic accountability, yet identifying such shifts at scale remains difficult.

We introduce TRACE (Temporal Rhetorical Analysis and Consistency Evaluation), a system built on the Digital Democracy Database (DDDB) for detecting rhetorical inconsistency in California legislative hearing testimony. TRACE organizes utterances into speaker-anchored timelines indexed by bill and session, then applies hybrid semantic retrieval — combining dense BGE embeddings with BM25 lexical …

Sam3-Litetext: An Anatomical Study Of The Sam3 Text Encoder For Efficient Vision-Language Segmentation, Chengxi Zeng, Yuxuan Jiang, Ge Gao, Shuai Wang, Duolikun Danier, Bin Zhu, Stevan Rudinac, David Bull, Fan Zhang

Research Collection School Of Computing and Information Systems

Vision-language segmentation models such as SAM3 enable flexible, prompt-driven visual grounding, but inherit large, general-purpose text encoders originally designed for open-ended language understanding. In practice, segmentation prompts are short, structured, and semantically constrained, leading to substantial over-provisioning in text encoder capacity and persistent computational and memory overhead. In this paper, we perform a large-scale anatomical analysis of text prompting in vision–language segmentation, covering 404,796 real prompts across multiple benchmarks. Our analysis reveals severe redundancy: most context windows are underutilized, vocabulary usage is highly sparse, and text embeddings lie on a low-dimensional manifold despite high-dimensional representations. Motivated by these findings, we …

Frozen Lvlms For Micro-Video Recommendation: A Systematic Study Of Feature Extraction And Fusion, Huatuan Sun, Yunshan Ma, Changguang Wu, Yanxin Zhang, Pengfei Wang, Xiaoyu Du

Research Collection School Of Computing and Information Systems

Frozen Large Video Language Models (LVLMs) are increasingly employed in micro-video recommendation (MVR) due to their strong multimodal understanding. However, existing apporches typically deploy LVLMs as fixed black-box feature extractors without systematically comparing alternative representation strategies. To address this gap, we present the first systematic empirical study on various feature extraction paradigms and integration strategies, along with hierarchical representations from frozen LVLMs for MVR. Extensive experiments on representative LVLMs reveal that hidden states from multiple decoder layers provide richer and more effective representations for MVR. Guided by this insight, we propose the Dual Feature Fusion (DFF) Framework, a lightweight approach …

Co-Designing With Autistic Livestreamers: Care, Constraints, And Trade-Offs In Livestreaming, Terrance Mok, Anthony Tang, Lora Oehlberg

Research Collection School Of Computing and Information Systems

Autistic livestreamers use platforms like Twitch for social connection, self-expression, and community, but these spaces also impose ongoing social and emotional demands. Prior work has documented these experiences, but less is known about what autistic creators themselves envision for the tools and platforms they use. We address this gap through a Research through Design (RtD) co-design study with three autistic Twitch streamers, using speculative artefacts as discussion prompts to explore how participants reasoned about potential livestreaming technologies. Across three co-design activities, we identify three overarching tensions shaping autistic streaming practice: Expression versus Misinterpretation and Harm; Public Participation versus Control and …

Towards Auto-Evaluation For Large Language Models, Jiahao Ying

Dissertations and Theses Collection (Open Access)

The rapid advancement of large language models (LLMs) has created an urgent need for evaluation methodologies that are timely, scalable, reliable, and informative. Conventional evaluation benchmarks, although essential for measuring model capabilities and guiding model development, are often constructed and maintained through labor-intensive human annotation. As LLMs continue to improve through increases in model scale, training data, and computational resources, static benchmarks may quickly lose discriminative power. Moreover, the growing use of large and diverse training corpora increases the risk of benchmark leakage, which can inflate evaluation results and obscure the true capabilities of models. These challenges call for a …

History To Future: Evolving Agent With Experience And Thought For Zero-Shot Vision-And-Language Navigation, Guangzhao Dai, Shuo Wang, Zihan Wang, Guo-Sen Xie, Yang Yang, Jinshan Pan, Qianru Sun, Xiangbo Shu

Research Collection School Of Computing and Information Systems

Vision-and-Language Navigation in Continuous Environment (VLN-CE) requires an agent to follow language instructions to navigate the target destination. With the advancement of large language models (LLMs), recent efforts have explored adapting them for zero-shot VLN-CE, offering a promising solution in addressing the drawbacks of poor generalization in the training-based paradigm. However, existing LLM-based works primarily perform naive reasoning for decision-making and lack feedback, e.g., reviewing historical errors and predicting future potentials. Consequently, it may suffer from continuous failure for those initial error tasks. In this paper, we rethink LLM-based zero-shot VLN-CE and propose a new paradigm, named EvoNav, to improve …

Task Complexity Matters: An Empirical Study Of Reasoning In Llms For Sentiment Analysis, Donghao Huang, Zhaoxia Wang

Research Collection School Of Computing and Information Systems

Large language models (LLMs) with reasoning capabilities have fueled a compelling narrative that reasoning universally improves performance across language tasks. We test this claim through a comprehensive evaluation of 504 configurations across seven model families—including adaptive, conditional, and reinforcement learning-based reasoning architectures—on sentiment analysis datasets of varying granularity (binary, five-class, and 27-class emotion). Our findings reveal that reasoning effectiveness is strongly task-dependent, challenging prevailing assumptions: (1) Reasoning shows task-complexity dependence—binary classification degrades up to -19.9 F1% points (pp), while 27-class emotion recognition gains up to  +16.0 pp; (2) Distilled reasoning variants underperform base models by 3–18 pp on simpler tasks, …

A Pruning-Based Question-Answering For Interactive Video Search: A Simple Baseline, Yu Tong Cheng, Phuong Anh Nguyen, Chong-Wah Ngo

Research Collection School Of Computing and Information Systems

There are various factors affecting the performance of video search. An imprecise query will enlarge search space and reduce the discriminative power of ranking functions. This problem is further exacerbated by the presence of numerous visually or semantically similar videos in large datasets. Consequently, users need to painstakingly browse through many highly similar candidates to locate the search target, leading to increased cognitive load and inefficient searching. Ideally, engaging users through interactive questioning to resolve uncertainties in the search process is an effective strategy for progressively narrowing down the search space. However, despite rapid advances in deep learning, generating informative …

A Novel Hierarchical Multi-Agent System For Payments Using Llms, Donghao Huang, Joon Kiat Chua, Zhaoxia Wang

Research Collection School Of Computing and Information Systems

Large language model (LLM) agents, such as OpenAI’s Operator and Claude’s Computer Use, can automate workflows but unable to handle payment tasks. Existing agentic solutions have gained significant attention; however, even the latest approaches face challenges in implementing end-to-end agentic payment workflows. To address this gap, this research proposes the Hierarchical Multi-Agent System for Payments (HMASP), which provides an end-to-end agentic method for completing payment workflows. The proposed HMASP leverages either open-weight or proprietary LLMs and employs a modular architecture consisting of the Conversational Payment Agent (CPA - first agent level), Supervisor agents (second agent level), Routing agents (third agent …

Rode: Linear Rectified Mixture Of Diverse Experts For Food Large Multi-Modal Models, Pengkun Jiao, Xinlan Wu, Bin Zhu, Jingjing Chen, Chong-Wah Ngo, Yu-Gang

Research Collection School Of Computing and Information Systems

Large Multi-modal Models (LMMs) have significantly advanced a variety of vision-language tasks. The scalability and availability of high-quality training data play a pivotal role in the success of LMMs. In the realm of food, while comprehensive food datasets such as Recipe1M offer an abundance of ingredient and recipe information, they often fall short of providing ample data for nutritional analysis. The Recipe1M+ dataset, despite offering a subset for nutritional evaluation, is limited in the scale and accuracy of nutrition information. To bridge this gap, we introduce Uni-Food, a unified food dataset that comprises over 100,000 images with various food labels, …

Benchmarking Gaslighting Negation Attacks Against Multimodal Large Language Models, Bin Zhu, Yinxuan Gui, Huiyan Qi, Jingjing Chen, Chong-Wah Ngo, Ee-Peng Lim

Research Collection School Of Computing and Information Systems

Multimodal Large Language Models (MLLMs) have exhibited remarkable advancements in integrating different modalities, excelling in complex understanding and generation tasks. Despite their success, MLLMs remain vulnerable to conversational adversarial inputs. In this paper, we systematically study gaslighting negation attacks—a phenomenon where models, despite initially providing correct answers, are persuaded by user-provided negations to reverse their outputs, often fabricating justifications. We conduct extensive evaluations of state-of-the-art MLLMs across diverse benchmarks and observe substantial performance drops when negation is introduced. Notably, we introduce the first benchmark GaslightingBench, specifically designed to evaluate the vulnerability of MLLMs to negation arguments. GaslightingBench consists of multiple-choice …

A Hybrid Approach To Phishing Email Detection: Leveraging Machine Learning And Large Language Models, Hessa Shamal Biri

Theses

Phishing attacks have reached a new level of sophistication through the deployment of large language models by attackers. The current AI-generated threats defeat existing detection systems which base their operation on past data. The thesis presents a hybrid system for phishing email detection which combines real email data with synthetic LLM-created samples to enhance traditional machine learning classifiers performance. The study created a hybrid dataset of 20,627 emails by combining 18,631 real messages from the Kaggle Email Classification Dataset with 1,996 synthetic emails. The synthetic emails were generated using four large language models LLaMA-3, Falcon, LLaVA, and Mistral to capture …

Principles Of Privacy And Security In Artificial Intelligence And Applications, Khang Tran

Dissertations

Modern artificial intelligence (AI) systems have transformed critical domains such as healthcare, software engineering, finance, and the legal system. Despite their broad impact, concerns about trustworthiness, especially regarding privacy and security, remain major obstacles to wider adoption. Addressing these concerns requires both a systematic understanding of the privacy and security risks inherent in AI systems and the development of principled foundations for trustworthy AI that safeguard client privacy and security. This goal is particularly challenging because of the complexity of modern large-scale AI systems, the trade-offs between privacy and model utility, and the need to simultaneously ensure other important properties …

Bridging Images And Language In Radiology: A Comprehensive Prisma Systematic Review Of Transformer Vision-Language Models And Clinical Readiness, Mohammad T. Khasawneh Dr., Sadaf Tabatabaee

Systems Science and Industrial Engineering Student Scholarship

Transformer vision-language models (VLMs) promise end-to-end automation of radiology reporting and related multimodal tasks. However, evidence remains fragmented across datasets, architectures, evaluation practices, and levels of clinical validation, limiting fair comparison and safe translation into practice. Following PRISMA 2020/PRISMA-S, search engines including PubMed, IEEE Xplore, Web of Science, and Google Scholar were systematically searched for peer-reviewed, English-language studies published between 2019 and 2025 that used paired radiology images and free-text reports. Dual reviewers screened records and extracted data using a locked schema covering datasets, modalities, architectures, training objectives, evaluation metrics, and indicators of clinical readiness. Free-text model descriptions were normalized …

Frictional Intelligence, Posheng Cheng

Masters Theses

This is an experimental interaction design project that challenges anthropomorphism in human-computer interaction. In particular, the recent advancement of artificial intelligence technologies like Large Language Models has taken anthropomorphism to new heights. The conversational chatbot interface of AI prioritizes mimicking an inherently human communication medium to maximize human-likeness. However, anthropomorphism has several downsides. Conversational interfaces obscure the limitations and the tangible cost of the technology. They also imply fictional moral status and human-level cognitive capabilities, which means general public sentiment focuses on the ``overhyped'' excitement and fear rather than on other socio-ethical and capacity questions that are far more urgent …

Optimizing Gated Rnns, Joshua Paul Fechete

Honors Projects

Gated recurrent neural networks such as Long Short-Term Memory (LSTM) and Gated Recurrent Units (GRU) help fix instability present in normal recurrent neural networks. This allows them to be used for various real-world tasks, and due to their architecture, they are uniquely qualified to handle variable sized input such as text. However, even before training can begin on a machine learning model, various hyperparameters must be chosen to decide how the model will be architectured. Choosing good hyperparameters is vital for creating a model that performs well but is not larger and more computationally expensive to run than it needs …

What Is The Skeleton Of Cognition? A Structural Account Of World Reconstruction Through Processing Axes, Griselda Poe

Publications and Research

This paper describes how the placement of a single processing axis reorganizes human cognition and generates a reconstructed world.

Most existing psychological and social theories begin from emotion, desire, morality, or social behavior. In doing so, they have discussed what forms on top of the cognitive skeleton without first fixing the skeleton itself. When the skeleton is not fixed, entirely different explanations of the same phenomenon can coexist, and it becomes difficult to identify which constitutes a foundational account.

This paper fixes the skeleton first. That skeleton is the processing axis.

The question is: when a single processing axis organizes …

Iterative Silver-Label Refinement For Temporal Information Extraction In Biomedical Literature, Chan Lee

UNLV Theses, Dissertations, Professional Papers, and Capstones

Temporal information extraction plays a critical role in the biomedical domain, where the ability to identify events and their temporal relationships is central to interpreting research findings. However, annotated corpora for this task remain scarce and costly to produce and the existing models developed for clinical text do not transfer well. This work bridges that gap through iterative silver-label refinement. A temporal model originally trained on news-domain data is adapted to biomedical text through cycles of automatic labeling, targeted correction, and retraining without the need for comprehensive manual annotation.

Key contributions include a practical iterative refinement methodology demonstrating that the …

Beyond English: Auditing And Mitigating Cross-Lingual Data Contamination In Multimodal Large Language Models, Pavan Dharma Adapa

Theses and Dissertations

This thesis extends data contamination auditing for multimodal large language models to multilingual settings. Using LLaVA 1.5 and a high-fidelity French parallel dataset derived from ScienceQA, the study evaluates how performance changes when identical image-question pairs are translated from English to French. The resultsshow a substantial cross-lingual performance decline and frequent flips from correct English predictions to incorrect French predictions, indicating that benchmark performance can depend heavily on memorized English-specific patterns rather than stable multimodal reasoning. To address this weakness, the thesis introduces an inference-time mitigation strategy based on perturbation ensembling and cross-lingual consistency aggregation. The proposed method reduces instance-level …

Evaluation And Mitigation Of Bias And Toxicity In Open-Source Large Language Models Using Crows-Pairs And Bold, Sai Harika Gade

Theses and Dissertations

This thesis evaluates bias and harmful language generation in five open-source language models and tests practical mitigation methods that do not require retraining. Two masked models are assessed with a sentence-pair benchmark for stereotype preference, and three generative models are assessed with a prompt-based benchmark for harmful continuations across demographic domains. The study uses a unified experimental workflow to compare model behavior, summarize differences across bias categories, and measure changes after intervention. Results show that the masked models favor stereotypical content above a random baseline, while the generative models usually produce low average toxicity but still show uneven risk across …