Information Security Commons^™

Metarag: Identifying Website Owner Using Meta-Path-Guided Dynamic Graph Retrieval-Augmented Generation, Cheng Tu, Yunshan Ma, Bingyang Guo, Qianyu Li, Yang Li, Min Zhang, Fan Shi, Xiang Wang

Research Collection School Of Computing and Information Systems

Website owner identification aims to link websites to their real-world owners, which is crucial for credibility assessment and information provenance in information retrieval and vital for applications in cybersecurity, Internet governance, and digital regulation. Existing approaches for website owner identification primarily rely on querying infrastructure registration records or analyzing webpage content. However, these methods often fail due to incomplete or outdated registration records and sparse webpage content. We observe that inter-website relationships, derived from shared infrastructure data such as primary domains, IP blocks, and geolocations, can provide valuable but underutilized ownership cues. To exploit this insight, we propose MetaRAG, a …

From Dissertation To Deployment: A Unified Software Platform Operationalizing Clinical-Prediction And Sequential-Security Ai For Healthcare, Olsi Shehu, Damiana Teliti, Jasmin Kevrić, Bekir Karlik

Communications of the IIMA

Advances in machine learning for healthcare are abundant, yet most validated models remain confined to research notebooks and never reach secure, usable clinical software. This paper addresses that deployment gap by presenting a unified, security-hardened software platform that operationalizes two complementary streams of doctoral research inside a single, role-based hospital information system. The first stream contributes a clinical-prediction capability: an ultra-hybrid ensemble that couples a quantum-inspired feature transformation, particle-swarm feature selection, and calibrated soft voting for cancer-outcome prediction (96.41% accuracy, AUC-ROC 0.983 on TCGA-BRCA), survival stratification, multi-cancer generalization, and pharmacogenomic drug-response classification (89.31% mean accuracy across 25 compounds). The second …

Navigating Enhanced Exploration Assistance (Nexa), Azhari Abbas, Caleb Fakunle, Ryan Powell, Donovan Livingston

Discovery Day - Daytona Beach

NEXA is an artificial intelligence software platform developed to enhance residential security and property monitoring through seamless integration with autonomous drone systems. This research application of advanced AI in surveillance aims to create a standalone solution capable of real-time threat detection and intelligent alert management. By processing visual and sensory data, NEXA facilitates autonomous drone operation with minimal human intervention. Secure communication channels ensure that instant alerts are delivered to property owners and, potentially, law enforcement, improving response times in security incidents, search-and-rescue operations, and perimeter surveillance. Additionally, NEXA is capable of interfacing with commercially available drone platforms and presents …

A Survey On Machine Learning Applications For Operating System Fingerprinting, Siri Siqveland

Discovery Day - Daytona Beach

In the modern age of computers and interconnected networks, cybersecurity and cyber-attackers are evolving in tandem to exploit each other’s vulnerabilities. One technique used by both parties is Operating System Fingerprinting (OSF): with the knowledge of what Operating System a target system is running, innate vulnerabilities can be identified and patched or exploited. Historically, OSF utilizes two main methods: passive and active—the former trades accuracy with undetectability while the latter is generally more detectable but more accurate. However, recent work has combined OSF with Machine Learning (ML) to improve accurate identification. The work presented here is a survey for the …

Efficient And Universal Watermarking For Llm-Generated Code Detection, Boquan Li, Zirui Fu, Mengdi Zhang, Peixin Zhang, Jun Sun, Xingmei Wang

Research Collection School Of Computing and Information Systems

Large language models (LLMs) have significantly enhanced the usability of AI-generated code, providing effective assistance to programmers. This advancement also raises ethical and legal concerns, such as academic dishonesty and the generation of malicious code. For accountability, it is imperative to detect whether a piece of code is AI-generated. Watermarking is broadly considered a promising solution and has been successfully applied to identify LLM-generated text. However, existing efforts on code are far from ideal, suffering from limited universality and excessive time and memory consumption. In this work, we propose a plugand- play watermarking approach for AI-generated code detection, named ACW …

Evaluating Machine Learning Models On Classification Of Novel Cyber Attacks In The Healthcare Domain, Promise Ehimen

Dissertations, Theses, and Projects

The increasing adoption of the Internet of Medical Things (IoMT) has improved healthcare delivery through connected medical devices while simultaneously expanding the cybersecurity risks facing healthcare organizations. Although machine learning based intrusion detection systems have demonstrated high detection accuracy, their ability to respond reliably to previously unseen cyberattacks remains uncertain. This study investigated how a Neural Network model and a Logistic Regression model classified novel cyberattacks within the IoMT environment. The Neural Network and Logistic Regression models were both trained and tested using a subset of the CICIoMT2024 benchmark dataset. The Neural Network achieved 99.82% test accuracy and a 0.94 …

Spatiotemporal Sycophancy: Negation-Based Gaslighting In Video Large Language Models, Ziyao Tang, Pengkun Jiao, Bin Zhu, Huiyan Qi, Jingjing Chen, Yu-Gang Jiang

Research Collection School Of Computing and Information Systems

Video Large Language Models (Vid-LLMs) have demonstrated remarkable performance in video understanding tasks, yet their robustness under conversational interaction remains largely underexplored. In this paper, we identify spatiotemporal sycophancy, a failure mode in which Vid-LLMs retract initially correct, visually grounded judgments and conform to misleading user feedback under negation-based gaslighting. Rather than merely changing their answers, the models often fabricate unsupported temporal or spatial explanations to justify incorrect revisions. To systematically investigate this phenomenon, we propose a negation-based gaslighting evaluation framework and introduce GasVideo-1000, a curated benchmark designed to probe spatiotemporal sycophancy with clear visual grounding and temporal reasoning requirements. …

Rendering Data Unlearnable By Exploiting Llm Alignment Mechanisms, Ruihan Zhang, Jun Sun

Research Collection School Of Computing and Information Systems

Large language models (LLMs) are increasingly trained on massive, heterogeneous text corpora, raising serious concerns about the unauthorised use of proprietary or personal data during model training. In this work, we address the problem of data protection against unwanted model learning in a realistic blackbox setting. We propose Disclaimer Injection, a novel data-level defence that renders text unlearnable to LLMs. Rather than relying on model-side controls or explicit data removal, our approach exploits the models’ own alignment mechanisms: injecting carefully designed alignment-triggers to prevent effective learning. Through layer-wise analysis, we find that finetuning on such protected data induces persistent activation …

Air: Improving Agent Safety Through Incident Response, Zibo Xiao, Jun Sun, Junjie Chen

Research Collection School Of Computing and Information Systems

Large Language Model (LLM) agents are increasingly deployed in practice across a wide range of autonomous applications. Yet current safety mechanisms for LLM agents focus almost exclusively on preventing failures in advance, providing limited capabilities for responding to, containing, or recovering from incidents after they inevitably arise. In this work, we introduce AIR, the first incident response framework for LLM agent systems. AIR defines a domain-specific language for managing the incident response lifecycle autonomously in LLM agent systems, and integrates it into the agent's execution loop to (1) detect incidents via semantic checks grounded in the current environment state and …

Data Governance Maturity, Ai Integration, And Equity In Colorado K-12 Public Schools, John R. Curtin

Electronic Theses and Dissertations

Colorado's 179 K-12 public school districts operate as autonomous governance units, each responsible for securing and managing student data assets that span health, financial, residential, and academic records. The accelerating integration of artificial intelligence (AI) and machine learning (ML) tools into administrative workflows, productivity software, and instructional platforms has fundamentally altered the risk landscape for student data, yet governance frameworks at the state, district, and school levels have not kept pace. This dissertation investigates whether Colorado's decentralized educational governance structure is institutionally capable of producing equitable, secure, and sustainable data governance outcomes in the AI era.

Drawing on Institutional Theory …

A Systematic Review Of Intrusion Detection Systems For Internet Of Medical Things: Performance, Efficiency, Explainability, And Generalization, Oswald Adohinzin, Youssef Harrath

Research & Publications

The Internet of Medical Things (IoMT) has transformed health care delivery through medical devices, remote patient monitoring, and real-time clinical decision support. However, the proliferation of IoMT devices introduces security vulnerabilities that put patient safety and data privacy at risk. Intrusion Detection Systems (IDS) have emerged as essential components for protecting IoMT networks from cyberattacks. This article presents a systematic review of IoMT-IDS research, analyzing 53 high-quality papers published between 2020 and 2025, identified through database searches spanning 2016–2025 across IEEE Xplore, Springer, ScienceDirect, and ACM Digital Library. We organize the literature through a comprehensive taxonomy spanning classical machine learning …

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 …

Security Assessment Of A Machine Learning Approach To Generate And Validate Digital Signatures, Juan Ortiz Couder

Doctoral Dissertations and Master's Theses

Cybersecurity has become a global concern as cyber-attacks have become more common, and the cost of the damage caused by them continues to increase. There are several approaches to improve the cyber security of systems such as Digital Signatures, hashing, watermarking, and encryption among others. Digital Signatures are a cryptographic technique used to verify the authenticity and integrity of digital messages or documents. Digital Signatures use a combination of hashing and public-private key encryption to verify the authenticity and integrity of videos, just as they are used for documents and messages. As a result of using a combination of other …

Benchmarking Gaslighting Attacks Against Speech Large Language Models, Jinyang Wu, Bin Zhu, Xiandong Zou, Qiquan Zhang

PhD Student’s Publications Collection

As Speech Large Language Models (Speech LLMs) become increasingly integrated into voice-based applications, ensuring their robustness against manipulative or adversarial input becomes critical. Although prior work has studied adversarial attacks in text-based LLMs and vision-language models, the unique cognitive and perceptual challenges of speech-based interaction remain underexplored. In contrast, speech presents inherent ambiguity, continuity, and perceptual diversity, which make adversarial attacks more difficult to detect. In this paper, we introduce gaslighting attacks, strategically crafted prompts designed to mislead, override, or distort model reasoning as a means to evaluate the vulnerability of Speech LLMs. Specifically, we construct five manipulation strategies: Anger, …

Federated Neuromorphic Intelligence: Advancing Robustness, Efficiency, And Continual Adaptation In Edge Environments, Manh V. Nguyen

Dissertations

This dissertation investigates how spiking neural networks (SNNs) can improve federated edge intelligence by advancing three interconnected goals: communication efficiency, adversarial robustness, and continual adaptation. As edge computing deployments expand across Internet of Things (IoT), sensing, and privacy-sensitive applications, conventional federated learning approaches built around artificial neural networks (ANNs) face growing limitations in power consumption, bandwidth demand, and resilience to real-world uncertainty. SNNs offer an alternative computational paradigm based on event-driven, sparse, and temporally structured processing that is naturally suited to constrained edge environments. However, their behavior in practical federated settings remains insufficiently understood.

To address this gap, this dissertation …

Hijacking The Prompt: A Survey Of Prompt Injection Attacks, Detection, And Defense In Large Language Models, Edward J. Griggs

Cybersecurity Undergraduate Research Showcase

Prompt injection attacks, ranked the number-one vulnerability in AI systems by OWASP's 2025 Top 10 for Large Language Model Applications, remain largely unsolved, and this survey examines why. As large language models (LLMs) are deployed across enterprise workflows, agentic systems, and consumer tools, their fundamental inability to distinguish trusted instructions from untrusted user data has created a persistent and expanding attack surface. This paper presents a structured taxonomy of prompt injection attack vectors, including direct injection, indirect injection, multimodal attacks, tool and agent exploitation, hybrid chained techniques, and autonomous propagating threats. These vectors are mapped across five impact categories (data …

A.I.R.E. - Ai-Assisted Reverse Engineering, Laurene Robinson

Posters - 2026

Reverse engineering plays a vital role in cybersecurity by helping analysts examine unknown binaries, investigate malware, identify vulnerabilities, and better protect sensitive systems. However, once a program is compiled and stripped, the meaningful names that describe its behavior are lost, leaving behind generic function labels like FUN_00401a30. Analysts must then manually interpret decompiled code, trace call chains, and infer program behavior function by function, which is slow and mentally demanding on large binaries. To address this challenge, this project introduces A.I.R.E., a local Ghidra extension that extracts contextual evidence from stripped functions and uses a locally hosted language model to …

The Psychology Behind Ai-Generated Phishing And Social Engineering Attacks, A’Shya Reynolds

School of Cybersecurity Master's Level Projects and Papers

Cybercrime has evolved significantly with the integration of artificial intelligence (AI), transforming traditional phishing and social engineering attacks into highly sophisticated and personalized threats. While early phishing attempts relied on generic messaging and low success rates, modern AI-driven attacks leverage advanced data analytics, natural language processing, and behavioral prediction to manipulate victims more effectively.

This research examines how cybercriminals utilize AI to enhance psychological manipulation techniques in phishing and social engineering attacks, increasing victim susceptibility. Drawing from interdisciplinary literature in cybersecurity and psychology, this study explores key psychological mechanisms, including cognitive biases, emotional triggers, and decision-making processes that influence victim …

Be Responsible In Your Answers! Monitoring Out-Of-Domain Behaviors In Domain-Specific Llms, Boquan Li, Chenzhe Lou, Zhe Ren, Peixin Zhang, Zirui Fu, Jun Sun, Yaowen Zheng

Research Collection School Of Computing and Information Systems

Large Language Models (LLMs) have accelerated the rapid development of chatbot web applications in various domains, such as coding, biomedicine and psychology. Compared to general LLMs like ChatGPT, domain-specific LLMs require a greater sense of responsibility. For instance, if a programming LLM casually answers medical or psychological questions, it not only misleads the public but also poses legal risks. This highlights new demands for monitoring and preventing such irresponsible behaviors. Existing efforts attempt to monitor LLMs from multiple aspects, such as lying, jailbreaks, and toxic content, while overlooking out-of-domain behaviors. In this work, we propose an innovative LLM domain monitoring …

Penforge: On-The-Fly Expert Agent Construction For Automated Penetration Testing, Huihui Huang, Jieke Shi, Junkai Chen, Ting Zhang, Yikun Li, Chengran Yang, Eng Lieh Ouh, Lwin Khin Shar, David Lo

Research Collection School Of Computing and Information Systems

Penetration testing is essential for identifying vulnerabilities in web applications before real adversaries can exploit them. Recent work has explored automating this process with Large Language Model (LLM)-powered agents, but existing approaches either rely on a single generic agent that struggles in complex scenarios or narrowly specialized agents that cannot adapt to diverse vulnerability types. We therefore introduce PenForge, a framework that dynamically constructs expert agents during testing rather than relying on those prepared beforehand. By integrating automated reconnaissance of potential attack surfaces with agents instantiated on the fly for context-aware exploitation, PenForge achieves a 30.0% exploit success rate (12/40) …

Where Did It Go Wrong? Attributing Undesirable Llm Behaviors Via Representation Gradient Tracing, Zhe Li, Wei Zhao, Yige Li, Jun Sun

Research Collection School Of Computing and Information Systems

Large Language Models (LLMs) have demonstrated remarkable capabilities, yet their deployment is frequently undermined by undesirable behaviors such as generating harmful content, factual inaccuracies, and societal biases. Diagnosing the root causes of these failures poses a critical challenge for AI safety. Existing attribution methods, particularly those based on parameter gradients, often fall short due to prohibitive noisy signals and computational complexity. In this work, we introduce a novel and efficient framework that diagnoses a range of undesirable LLM behaviors by analyzing representation and its gradients, which operates directly in the model's activation space to provide a semantically meaningful signal linking …

Propaganda Ai: An Analysis Of Semantic Divergence In Large Language Models, Nay Myat Min, Long H. Pham, Yige Li, Jun Sun

Research Collection School Of Computing and Information Systems

Large language models (LLMs) can exhibit concept-conditioned semantic divergence: common high-level cues (e.g., ideologies, public figures) elicit unusually uniform, stance-like responses that evade token-trigger audits. This behavior falls in a blind spot of current safety evaluations, yet carries major societal stakes, as such concept cues can steer content exposure at scale. We formalize this phenomenon and present RAVEN (Response Anomaly Vigilance), a black-box audit that flags cases where a model is simultaneously highly certain and atypical among peers by coupling semantic entropy over paraphrastic samples with cross-model disagreement. In a controlled LoRA fine-tuning study, we implant a concept-conditioned stance using …

Llmqua: Practical Backdoor Injection On Large Language Model Quantization, Xiangxiang Chen, Peixin Zhang, Jun Sun, Jin Song Dong, Wenhai Wang, Jingyi Wang

Research Collection School Of Computing and Information Systems

Quantization is widely used to enable local deployment of large language models (LLMs) on resource-constrained devices. Recent work (e.g., QuRA) shows quantization can be exploited via rounding manipulation to implant backdoors. However, such an attack has been evaluated only on small models and does not directly apply to LLMs due to three key constraints: (1) limited poisoning data from small, task-agnostic calibration sets; (2) layer-wise quantization restricting adversarial access to global representations; and (3) lack of gradient access in quantization pipelines, blocking gradient-based attacks.We propose LLMQuA, a practical quantization-phase backdoor attack tailored to the LLM setting. LLMQuA (i) injects backdoors …

Large-Scale File Fragment Classification Via Multi-View Learning, Samuel Hildebrand

LSU Master's Theses

File reassembly is one of the most fundamental tasks in digital forensics, enabling recovery of data from potentially damaged storage media even when file system metadata is unavailable. This thesis reviews more than two decades of work in the realm of file carving, with a particular focus on fragmented file carving, which remains a focus of research, and file fragment classification, a principal component of fragmented file carving. This thesis serves a literature review of both file carving and fragmented file carving, surveys the massive amounts of data needed for the task of fragment classification and the datasets that serve …

Anomaly Detection For Multi-System Bug Triage, Gibran Miguel Zavala Gamero, Hayoung Cheon, Mustafa Iqbal

SMU Data Science Review

Large-scale software systems produce vast volumes of logs and telemetry, making manual incident triage slow and error prone. This study presents an unsupervised anomaly detection pipeline that fuses logs, metrics, and traces through late fusion. Using Hybrid Ensemble modeling with Isolation Forest, and Long Short-Term Memory (LSTM) Deep Learning model, the system detects cross-service anomalies producing and assigning a composite triage score reflecting severity and impact. Ranked alerts are categorized into Critical, High, or Medium priorities for review. A retrieval-augmented generation (RAG) layer enriches results with contextual summaries for explainable triage. Evaluated on synthetic multi-service datasets, the pipeline …

Deep Learning Approaches For Anti-Money Laundering On Mobile Transactions: Review, Framework, And Directions, Jiani Fan, Lwin Khin Shar, Ruichen Zhang, Ziyao Liu, Wenzhuo Yang, Dusit Niyato, Kwok-Yan Lam

Research Collection School Of Computing and Information Systems

Money laundering is a financial crime that obscures the origin of illicit funds, necessitating the development and enforcement of anti-money laundering (AML) policies by governments and organizations. The proliferation of mobile payment platforms and smart IoT devices has significantly complicated AML investigations. As payment networks become more interconnected, there is an increasing need for efficient real-time detection to process large volumes of transaction data on heterogeneous payment systems by different operators such as digital currencies, cryptocurrencies, and account-based payments. Most of these mobile payment networks are supported by connected devices, many of which are considered loT devices in the FinTech …

Mm-Attackg: A Multimodal Approach To Attack Graph Construction With Large Language Models, Yongheng Zhang, Xinyun Zhao, Yunshan Ma, Haokai Ma, Yingxiao Guan, Guozheng Yang, Yuliang Lu, Xiang Wang

Research Collection School Of Computing and Information Systems

Cyber Threat Intelligence (CTI) parsing aims to extract key threat information from massive data, transform it into actionable intelligence, enhance threat detection and defense efficiency, including attack graph construction, intelligence fusion, and indicator extraction. Among these research topics, Attack Graph Construction (AGC) is essential for visualizing and understanding the potential attack paths of threat events from CTI reports. Existing approaches primarily construct the attack graphs purely from the textual data to reveal the logical threat relationships between entities within the attack behavioral sequence. However, they typically overlook the specific threat information inherent in visual modalities, which preserves key threat details …

Systematic Approaches To Characterizing Vulnerabilities And Enhancing Robustness Of Text And Vision-Language Models, Poojitha Thota

Computer Science and Engineering Dissertations

The proliferation of artificial intelligence (AI) across critical domains, including news summarization, privacy-policy analysis, and medical decision support, has raised growing concerns about the security and robustness of these systems against adversarial manipulation. This dissertation investigates adversarial robustness in generative AI by addressing three key research goals: (1) characterizing adversarial vulnerabilities across generative models, (2) developing systematic defenses to improve the robustness of generative models, and (3) designing deployment-time safeguards for securing LLM interactions.

Towards the first goal, we characterize adversarial vulnerabilities across text-based and multimodal systems. In abstractive text summarization, we show that inference-time perturbations can exploit lead bias …

Ordered Mini-Batch Training For Differentially Private And Encrypted Logistic Regression, Ryan Leone

Theses, Dissertations and Culminating Projects

Logistic regression has found extensive use as a supervised machine learning algorithm due to its simplicity and efficiency in binary and multivariate classification tasks. As data sharing grows across connected devices, safeguarding sensitive personal and industrial information is of increased importance. Privacy-preserving machine learning techniques such as differential privacy and homomorphic encryption offer mathematically rigorous security guarantees, but introduce difficult accuracy, privacy loss, and computational overhead issues. This thesis investigates PPML for logistic regression through a collaborative mini-batch training framework. I propose and implement an ordered mini-batch strategy, compare it to standard shuffled methods, then integrate differential privacy noise injection …

Tackling The Societal And Regulatory Challenges Of Emerging Technologies: A Case Study Of Deepfake, Jingyao Li

2026

Governing emerging technologies such as Artificial Intelligence (AI) poses enduring challenges for policymakers, industries, and societies. Early-stage governance is often hindered by limited understanding of technological implications, rapid innovation cycles, and resistance from powerful industry actors who favor minimal oversight. Yet, timely and effective governance is essential, as new technologies are most malleable in their formative stages. This dissertation examines how emerging technologies can be governed effectively by using deepfakes technology as a focal case. This dissertation comprises three interrelated studies.

The first paper reviews the literature on deepfakes and emerging technology governance, identifying the distinct characteristics of deepfake technology …