Digital Commons Network^™

Memotune: A Measure And Moment-Driven Fine-Tuning Framework For Quantized Large Language Models, Yun Zhang, Xue Geng, Lizi Liao, Jintong Sun, Minghe Yu, Ge Yu

Research Collection School Of Computing and Information Systems

Quantizing large language models (LLMs) is essential for reducing memory and computational costs in natural language processing. Existing methods combine quantization with parameter-efficient fine-tuning but often fail to meet practical performance requirements. This paper introduces MeMoTune, a novel fine-tuning framework for quantized LLMs. By employing a measure and moment approach within a low-rank approximation framework in probability measure space, MeMoTune optimizes the objective function for superior fine-tuning results. The update process is further refined through scaled gradient, enhancing convergence efficiency and noise robustness. Experiments on tasks like text generation, summarization, and understanding show MeMoTune significantly outperforms state-of-the-art methods, e.g. fine-tuning …

R2dqg: A Quality Meets Diversity Framework For Question Generation Over Knowledge Bases, Yimeng Ren, Yanhua Yu, Lizi Liao, Yuhu Shang, Kangkang Lu, Mingliang Yan

Research Collection School Of Computing and Information Systems

The task of Knowledge-Based Question Generation (KBQG) involves generating natural language questions from structured knowledge sources, posing unique challenges in balancing linguistic diversity and semantic relevance. Existing models often focus on maximizing surface-level similarity to ground-truth questions, neglecting the need for diverse syntactic forms and leading to semantic drift during generation. To overcome these challenges, we propose Refine-Reinforced Diverse Question Generation (R2DQG), a two-phase framework leveraging a generation-then-refinement paradigm. The Generator first constructs a diverse set of expressive templates using dependency parse tree similarity, capturing a wide range of syntactic patterns and styles. These templates guide the creation of question …

Human-Ai-Collaboration-For-Coding, Siddhardha Ravi

Theses, Dissertations and Culminating Projects

AI-generated code, while rapidly producing functional solutions, often falls short in aspects like comprehensive error handling, robust documentation, and optimal architectural design, areas where human expertise excels. Conversely, humans can greatly benefit from AI's rapid code generation capabilities. This project proposes and evaluates "A Framework to Improve Code Quality by Utilizing Generative AI Coding Along With Human-Written Code", designed to create a synergy between AI and human intelligence for enhanced software development. Conducted over four weeks, the research leverages BigCodeBench as its core dataset to rigorously investigate how human intervention can improve AI-generated code quality, identify the most effective human-AI …

Tetd: Trusted Execution In Trust Domains, Zhanbo Wang, Jiaxin Zhan, Xuhua Ding, Fengwei Zhang, Ning Hu

Research Collection School Of Computing and Information Systems

Intel TDX empowers cloud service providers to construct confidential virtual machines called trust domains (TDs) on x86 platforms. Similar to its counterparts from AMD and Arm, TDX's hardware based protection over integrity and secrecy of virtual machine memory and vCPU states inevitably hinders legitimate virtual machine management such as introspection. At the presence of compromised high-privileged software (e.g., the guest kernel), neither the cloud service provider nor the TD owner can securely carry out a task within the TD. To tackle this problem, we propose TETD, an in-TD trusted execution technique without trusting any TD system software. Our design does …

Oblivious Digital Tokens, Mihael Liskij, Xuhua Ding, Gene Tsudik, David A. Basin

Research Collection School Of Computing and Information Systems

A computing device typically identifies itself by exhibiting unique measurable behavior or by proving its knowledge of a secret. In both cases, the identifying device must reveal information to a verifier. Considerable research has focused on protecting identifying entities (provers) and reducing the amount of leaked data. However, little has been done to conceal the fact that the verification occurred.We show how this problem naturally arises in the context of digital emblems, which were recently proposed by the International Committee of the Red Cross to protect digital resources during cyber-conflicts. To address this new and important open problem, we define …

Machine Learning Based Medical Ultrasound Image Classification And Grad-Cam Interpretation, Victoria C. Hemphill

All Theses

This work takes a step in creating a diagnostic tool for the classification decision process of Achilles tendinopathy using ultrasound images. An attention-based multiple instance learning model is developed to classify the images. Typically, doctors capture multiple ultrasound images of the Achilles tendon during a study to determine a complete diagnosis. Multiple instance models adopt this behavior by providing a single label for a set of instances (images). The images are grouped into ”bags” at the study level and passed into the model. The MIL model then uses its attention property to assign an importance score to each image to …

Practical Keyword Private Information Retrieval From Key-To-Index Mappings, Meng Hao, Weiran Liu, Liqiang Peng, Cong Zhang, Pengfei Wu, Lei Zhang, Hongwei Li, Deng, Robert H.

Research Collection School Of Computing and Information Systems

This paper introduces practical schemes for keyword Private Information Retrieval (keyword PIR), enabling private queries on public databases using keywords. Unlike standard indexbased PIR, keyword PIR presents greater challenges, since the query’s position within the database is unknown and the domain of keywords is vast. Our key insight is to construct an efficient and compact key-to-index mapping, thereby reducing the keyword PIR problem to standard PIR. To achieve this, we propose three constructions incorporating several new techniques. The high-level approach involves (1) encoding the server’s key-value database into an indexable database with a key-to-index mapping and (2) invoking standard PIR …

Think Both Ways: Teacher-Student Bidirectional Reasoning Enhances Mcq Generation And Distractor Quality, Yimiao Qiu, Yang Deng, Quanming Yao, Zhimeng Zhang, Zhiang Dong, Chang Yao, Jingyuan Chen

Research Collection School Of Computing and Information Systems

Generating high-quality Multiple Choice Questions (MCQs) remains challenging for educational tools due to the need for contextual relevance and plausible distractors. Existing methods still struggle with these dual requirements, leading to questions that lack depth and distractors that are either too obvious or irrelevant. In this paper, we propose BiFlow, a novel framework that integrates bidirectional reasoning perspectives: teacher reasoning generates contextually relevant questions and plausible distractors, while student reasoning evaluates question clarity and the misleading nature of the distractors. To further enhance reasoning, we introduce PathFinder, a mechanism that employs breadth-first search and Chainof-Thought (CoT) strategies to explore diverse …

Classifying Advanced Persistent Threat Stages And Techniques Via Graph-Enhanced Network Flow Representations, Md Taef Uddin Nadim

Graduate Theses and Dissertations

Advanced Persistent Threats (APTs) are complex, stealthy attacks that involve multiple stages and many attack techniques used in each stage, making them difficult to defend against. Although many solutions can detect APTs, most of them only detect the existence of attack, but cannot produce fine-grained classification over the stage of the APT and the specific attack technique used. Some existing solutions can classify the stages of APT, but few of them provide attack technique classification, and existing work do not provide interpretability for the classification or countermeasures for the attack. In this thesis work, we propose a solution named CAPTure, …

Study Of Ai Applications In Biomedical Data Acquisition, Communication, And Analysis: Cest Mri Acceleration And Ecg Transmissions, Adarsha Bhattarai

Dissertations and Doctoral Documents, University of Nebraska-Lincoln, 2023–

This dissertation investigates the application of artificial intelligence in biomedical data acquisition, communication, and analysis to advance neurological research and to enable the early detection of cardiovascular conditions. Despite significant advances in imaging and physiological modalities, challenges persist. Imaging modalities, such as the chemical exchange saturation transfer magnetic resonance imaging (CEST MRI) technique are challenged by a prolonged data acquisition time and high operational costs. In addition, physiological modalities such as electrocardiogram (ECG) sensors face constraints in providing uninterrupted signal monitoring which is crucial for the timely detection of premature cardiac abnormalities. The primary goal of this work is to …

Online Prediction Of Streaming Data, Aleena Chanda

Dissertations and Doctoral Documents, University of Nebraska-Lincoln, 2023–

We present two new approaches for point prediction with streaming data based on a) the Count-Min sketch and b) Gaussian Process Priors with random bias. The methods are intended for the most general case where no true model can be usefully formulated for the data stream. In statistical contexts, this is often called the M open problem class. For the Count Min Sketch method we show that the predicted distribution function ^F converges to F under the assumption that the data consists of i.i.d samples from a fixed distribution function F. To implement the Gaussian Process Prior methods, we used …

How Developers Use Type-System Related Programming Language Features, Samuel W. Flint

Dissertations and Doctoral Documents, University of Nebraska-Lincoln, 2023–

Optional type annotations are a popular feature of programming languages that allow developers to omit explicit type information in code while, in some cases, retaining many of the benefits of static typing, such as in-code documentation, improved detection of type errors, or enforcement of code properties. However, how developers use and understand optional type annotations is not clear. The focus of this dissertation is to understand the use and comprehension of optional type annotations.

Optional type annotations are examined through four lenses: first, by examining the evolution of usage in a statically typed programming language (Kotlin, the default language for …

Use Of A Preliminary Artificial Intelligence-Based Laryngeal Cancer Screening Framework For Low-Resource Settings: Development And Validation Study, Shao Wei Sean Lam, Min Hun Lee, Michael Dorosan, Samuel Altonji, Hiang Khoon Tan, Walter T. Lee

Research Collection School Of Computing and Information Systems

Background: Early-stage diagnosis of laryngeal cancer significantly improves patient survival and quality of life. However, the scarcity of specialists in low-resource settings hinders the timely review of flexible nasopharyngoscopy (FNS) videos, which are essential for accurate triage of at-risk patients.Objective: We introduce a preliminary AI-based screening framework to address this challenge for the triaging of at-risk patients in low-resource settings. This formative research addresses multiple challenges common in high-dimensional FNS videos: (1) selecting clear, informative images; (2) deriving regions within frames that show an anatomical landmark of interest; and (3) classifying patients into referral grades based on the FNS video …

Wildfires Classification In Canadian Boreal Forest: A Comparative Study Of Logistic Regression And Xgboost Models, Brandon Tran, Elijah James Duran, Mike Luu, Hesham Morgan, Surendra Maharjan, Wenzhao Li, Hesham El-Askary

Mathematics, Physics, and Computer Science Faculty Articles and Research

In recent years, Canada has faced a growing number of wildfires. These events have devastated ecosystems, displaced communities, and posed severe health risks. To minimize the damage caused by such disasters, this study aims to develop an early warning system that predicts wildfire occurrences. Two machine learning models for binary classification of wildfire occurrence in Canadian wild forests, Logistic regression and XGBoost, will be compared and evaluated. The models are used to predict the likelihood of wildfire events based on various environmental and climatic factors. The models are evaluated using a 70-30 split validation approach and their performance is assessed …

Robustness Investigation, Detection, And Defense Of Deep Learning Models Against False Data Injection, Amirhossein Nazeri

All Dissertations

This dissertation addresses the critical challenge of adversarial robustness in deep learning systems, focusing on two fundamental domains: time-series prediction and object detection. As these AI systems become increasingly deployed in safety-critical applications from power grid management to autonomous vehicles their vulnerability to adversarial attacks poses significant risks to infrastructure and human safety.

The first contribution introduces a novel stealthy black-box False Data Injection (FDI) attack specifically designed for quasi-periodic time-series data. Unlike existing attacks that produce easily detectable anomalies, our method generates adversarial perturbations that preserve the underlying periodicity and statistical properties of the data, effectively bypassing traditional anomaly …

How Developers Use Type-System Related Programming Language Features, Samuel W. Flint

School of Computing: Dissertations, Theses, and Student Research

Optional type annotations are a popular feature of programming languages that allow developers to omit explicit type information in code while, in some cases, retaining many of the benefits of static typing, such as in-code documentation, improved detection of type errors, or enforcement of code properties. However, how developers use and understand optional type annotations is not clear. The focus of this dissertation is to understand the use and comprehension of optional type annotations.

Optional type annotations are examined through four lenses: first, by examining the evolution of usage in a statically typed programming language (Kotlin, the default language for …

Evaluating Immersion And Agency In Ai-Assisted Live Murder Mystery Games, Seraphina C. Courtney

LSU New Orleans Theses and Dissertations

This thesis explores the impact of AI-assisted narrative generation on player immersion and agency in a live-action roleplaying (LARP) experience. A live-action murder mystery game was designed and run in two formats: a static version with GPT-4 generated characters, dialogue, and stage directions, and a dynamic version where players created their own characters and improvised freely, guided by AI-generated narrative beats, a story element that moves the plot forward [21], that provided a narrative scaffolding for the players. The dynamic version employed a distributed computer vision system that tracks the movement of key items in the play-space so their relevance …

Introduction To C++ (Volume I), Hussam Ghunaim Ph.D.

All Open Educational Resources

This book is written as an Open Education Resource (OER) to replace expensive commercial materials currently used at the Department of Computer Science at Fort Hays State University. It has two volumes corresponding to the CSCI 121 and CSCI 221 courses. These courses are developed to introduce college freshmen students to Object-Oriented Programming utilizing C++. The author tried to bridge the gap in the current programming textbooks by avoiding lengthy and, on many occasions, unnecessary details. This book’s main feature is to present the discussed principles in the least wording possible while providing adequate examples and exercises to reinforce students’ …

Optimal Hypergraph Connectivity With Cut Queries, Hang Liao

Dartmouth College Ph.D Dissertations

Finding connected components in undirected hypergraphs—hypergraph connectivity—is a fundamental problem in computer science. It can be framed as a special case of Symmetric Submodular Function Minimization (SSFM), where the objective is to determine if the non-trivial minimizer is zero. This thesis develops an optimal algorithm for hypergraph connectivity within the $\CUT$ query model, where an algorithm probes a subset of vertices to learn the weight of the hyperedges ``cut" by that partition.

Our approach is constructive, culminating in an optimal algorithm for the general problem by first developing the necessary tools for two foundational subproblems. The main contributions of this …

Low-Cost Monitoring And Fingerprinting Of High-Powered Electric Systems, Kwabena Buamono Aboagye-Otchere

Theses and Dissertations

Electric motors are vital to industry, transport, and energy, yet their maintenance challenges persist. While traditional reactive maintenance leads to costly downtime and safety risks, predictive maintenance, especially through IoT and machine learning offers early fault detection and operational efficiency. However, this shift introduces security concerns due to unintended magnetic emissions from motors. These emissions, though useful for non-intrusive monitoring, can be exploited to eavesdrop on sensitive industrial processes. This dissertation explores the dual nature of magnetic emissions: their value in motor diagnostics and their potential as a security vulnerability. It demonstrates how emissions can identify motors, monitor health, and …

Fact-Checker: A Web Application For Leveraging Large Language Models For Fact-Checking Youtube Videos, Andrew R. Craig

Electronic Theses, Projects, and Dissertations

Fact-Checker is a web application that allows users to fact-check YouTube videos. It feeds YouTube’s closed captioning transcript to a large language model (LLM) to extract claims. It then uses multiple LLMs, such as Gemini, Llama, and Claude, to verify these claims. The modular design makes it easy to change to a different LLM or model if needed. The application is built using Python for access to Application Programming Interfaces (APIs) and Streamlit as the front-end framework. The utilization of Docker and Dockerfiles enables easy distribution and deployment. It enables the application to be deployed on almost any hardware platform …

L3net: Localized And Layered Reparameterization For Incremental Learning, Xuandi Luo, Huaidong Zhang, Yi Xie, Hongrui Zhang, Xuemiao Xu, Shengfeng He

Research Collection School Of Computing and Information Systems

Model-based class incremental learning (CIL) methods aim to address the challenge of catastrophic forgetting by retaining certain parameters and expanding the model architecture. However, retaining too many parameters can lead to an overly complex model, increasing inference overhead. Additionally, compressing these parameters to reduce the model size can result in performance degradation. To tackle these challenges, we propose a novel three-stage CIL framework called Localized and Layered Reparameterization for Incremental Learning (L3Net). The rationale behind our approach is to balance model complexity and performance by selectively expanding and optimizing critical components. Specifically, the framework introduces a Localized Dual-path Expansion structure, …

Focus: Evaluating Pre-Trained Vision-Language Models On Underspecification Reasoning, Kankan Zhou, Yibin Lai, Kyriakos Mouratidis, Jing Jiang

Research Collection School Of Computing and Information Systems

Humans possess a remarkable ability to interpret underspecified ambiguous statements by inferring their meanings from contexts such as visual inputs. This ability, however, may not be as developed in recent pre-trained visionlanguage models (VLMs). In this paper, we introduce a novel probing dataset called FOCUS to evaluate whether state-of-the-art VLMs have this ability. FOCUS consists of underspecified sentences paired with image contexts and carefully designed probing questions. Our experiments reveal that VLMs still fall short in handling underspecification even when visual inputs that can help resolve the ambiguities are available. To further support research in underspecification, FOCUS will be released …

A Comprehensive Analysis Of Evolving Permission Usage In Android Apps: Trends, Threats, And Ecosystem Insights, Ali Alkinoon, Trung Cuong Dang, Ahod Alghuried, Abdulaziz Alghamdi, Soohyeon Choi, Manar Mohaisen, An Wang, Saeed Salem, David Mohaisen

Research Collection School Of Computing and Information Systems

The proper use of Android app permissions is crucial to the success and security of these apps. Users must agree to permission requests when installing or running their apps. Despite official Android platform documentation on proper permission usage, there are still many cases of permission abuse. This study provides a comprehensive analysis of the Android permission landscape, highlighting trends and patterns in permission requests across various applications from the Google Play Store. By distinguishing between benign and malicious applications, we uncover developers’ evolving strategies, with malicious apps increasingly requesting fewer permissions to evade detection, while benign apps request more to …

Explainable Multimodal Sentiment Analysis Of Social Media Visual Content For Child Safety, Yee Sen Tan, Zhaoxia Wang

Research Collection School Of Computing and Information Systems

Ensuring the safety and well-being of children is increasingly important, especially in a world where visual content is pervasive. This paper proposes a novel multimodal, multilingual, and multiclass sentiment analysis method for social media content, aimed at improving content moderation for child safety. Our approach integrates textual, visual, and audio data from videos, categorizing sentiment into four levels: positive, slightly negative, negative, and strongly negative, enabling granular detection of harmful content. To enhance explainability and trust, we also leverage interpretable mechanisms to analyze the contributions of each modality. Evaluation of our method demonstrates strong generalization across diverse video types, and …

Xfinbench: Benchmarking Llms In Complex Financial Problem Solving And Reasoning, Zhihan Zhang, Yixin Cao, Lizi Liao

Research Collection School Of Computing and Information Systems

Solving financial problems demands complex reasoning, multimodal data processing, and a broad technical understanding, presenting unique challenges for current large language models (LLMs). We introduce **XFinBench**, a novel benchmark with 4,235 examples designed to evaluate LLM’s ability in solving comple**X**, knowledge-intensive **Fin**ancial problems across diverse graduate-level finance topics with multi-modal context. We identify five core capabilities of LLMs using XFinBench, i.e., _terminology understanding_, _temporal reasoning_, _future forecasting_, _scenario planning_, and _numerical modelling_. Upon XFinBench, we conduct extensive experiments on 18 leading models. The result shows that o1 is the best-performing text-only model with an overall accuracy of 67.3%, but still …

Taclr: A Scalable And Efficient Retrieval-Based Method For Industrial Product Attribute Value Identification, Yindu Su, Huike Zou, Lin Sun, Ting Zhang, Haiyang Yang, Chen Li Yu, David Lo, Qingheng Zhang, Shuguang Han, Jufeng Chen

Research Collection School Of Computing and Information Systems

Product Attribute Value Identification (PAVI) involves identifying attribute values from product profiles, a key task for improving product search, recommendation, and business analytics on e-commerce platforms. However, existing PAVI methods face critical challenges, such as inferring implicit values, handling outof-distribution (OOD) values, and producing normalized outputs. To address these limitations, we introduce Taxonomy-Aware Contrastive Learning Retrieval (TACLR), the first retrieval-based method for PAVI. TACLR formulates PAVI as an information retrieval task by encoding product profiles and candidate values into embeddings and retrieving values based on their similarity. It leverages contrastive training with taxonomy-aware hard negative sampling and employs adaptive inference …

Fine‑Tuning Multimodal Large Language Models For Product Bundling, Xiaohao Liu, Jie Wu, Zhulin Tao, Yunshan Ma, Yinwei Wei, Tat-Seng Chua

Research Collection School Of Computing and Information Systems

Recent advances in product bundling have leveraged multimodal information through sophisticated encoders, but remain constrained by limited semantic understanding and a narrow scope of knowledge. Therefore, some attempts employ In-context Learning (ICL) to explore the potential of large language models (LLMs) for their extensive knowledge and complex reasoning abilities. However, these efforts are inadequate in understanding mulitmodal data and exploiting LLMs' knowledge for product bundling. To bridge the gap, we introduce Bundle-MLLM, a novel framework that fine-tunes LLMs through a hybrid item tokenization approach within a well-designed optimization strategy. Specifically, we integrate textual, media, and relational data into a unified …

Collisionrepair: First‑Aid And Automated Patching For Storage Collision Vulnerabilities In Smart Contracts, Yu Pan, Wanjing Han, Yue Duan, Mu Zhang

Research Collection School Of Computing and Information Systems

Storage collision vulnerabilities, a significant security risk in upgradeable smart contracts, often arise when a user-facing proxy contract and a backend logic contract share storage space. While static analysis techniques can detect such issues, they often over-approximate program states, leading to false positives and requiring developers to manually verify each issue, giving attackers time to exploit any overlooked vulnerabilities. To address this, we propose COLLISIONREPAIR, an automated patching technique for mitigating storage collision risks. COLLISIONREPAIR monitors storage access sequences between proxy and logic contracts by defining an "ownership" property for storage locations. It then replays historical transactions to recover existing …

Prism: To Fortify Widget Based User‑App Data Exchanges Using Android Virtualization Framework, Yingtat Ng, Zhe Chen, Haiqing Qiu, Xuhua Ding

Research Collection School Of Computing and Information Systems

We present Prism, an UI hardening technique for an Android app to safeguard its widgets against a corrupted kernel. Prism ensures secure interface rendering and allows for visual authentication, which developers could use to enable user intent confidentiality protection. Our design leverages the recent Android Virtualization Framework with minimal changes to the existing UI framework and graphics subsystem. It is much easier to deploy and use Prism on Android phones than TrustZone-based secure UI schemes, because the apps are not admitted to the Secure World and retain their full rights to manage and control their own interfaces. We have implemented …