Information Security Commons^™

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 …

Avadclip: Audio-Visual Collaboration For Robust Video Anomaly Detection, Peng Wu, Wanshun Su, Guansong Pang, Yujia Sun, Qingsen Yan, Peng Wang, Yanning Zhang

Research Collection School Of Computing and Information Systems

With the increasing adoption of video anomaly detection in intelligent surveillance domains, conventional visual-only detection approaches often struggle with information insufficiency and high false-positive rates in complex environments. To address these limitations, we present a novel weakly supervised framework that leverages audio-visual collaboration for robust video anomaly detection. Capitalizing on the exceptional cross-modal representation learning capabilities of Contrastive Language-Image Pretraining (CLIP) across visual, audio, and textual domains, our framework introduces two major innovations: an efficient audio-visual fusion that enables adaptive cross-modal integration through lightweight parametric adaptation while maintaining the frozen CLIP backbone, and a novel audio-visual prompt that dynamically enhances …

Purifai: Detecting And Fixing Search-Induced Distortions In Web-Augmented Llms, Guoqing Wang, Zhao Zhang, Zeyu Sun, Xiaofei Xie, Yizhou Chen, Yanchao Tan, Dan Hao

Research Collection School Of Computing and Information Systems

As Large Language Models (LLMs) increasingly serve as interfaces for proprietary data (e.g., enterprise knowledge bases, legal statutes), ensuring their fidelity to trusted internal information is paramount. While integrating real-time web search can enhance model utility, it introduces a critical vulnerability: the ingestion of conflicting, misleading, or hallucinated content from the open web can override the model's adherence to its verified internal knowledge. We define this failure mode as search-induced distortion, a significant risk in high-stakes domains where the internal knowledge base serves as the absolute ground truth.To address this challenge, we present PurifAI, a proactive, model-agnostic, cache-level purification system …

A Robust Hybrid Security Framework: Integrating Multi-Layered Text Encryption With Barcode-Based Steganography, Mohamed Sayed, Talaat M. Wahbi, Farooq Abdalwahab Haboub

BAU Journal - Science and Technology

The widespread use of the Internet is causing increasing security concerns regarding online communications. One method for achieving secure communication between authorized parties is steganography. We herein employ multilevel technologies, including compression, encryption, barcoding, and steganography to secure a secret text message. Type I multilevel steganography is used with a two-level setup. The first level uses enhanced least significant bit (secure LSB-L1) image steganography; the output is a stego-image file, the cover is an image file, and the secret data in this level is English text. The output from the first level is encrypted using the RSA algorithm, and the …

Operationalizing Supply-Chain Hygiene In Graduate Is Education: A Hands-On Module For Secure Software And Ai/Ml Pipelines, Dominic A. Wilson

Journal of Cybersecurity Education, Research and Practice

Supply-chain attacks (including typosquatting, dependency confusion, compromised builds, dataset poisoning, and backdoored models) pose growing threats to analytics platforms central to Information Systems (IS). While frameworks like the Secure Software Development Framework (SSDF) and Supply-chain Levels for Software Artifacts (SLSA) offer guidance, IS curricula often lack accessible, infrastructure-light modules that build practical skills for mitigating these risks. This experience report presents a two-week module embedded in a graduate Secure Coding course required for a Master’s in Applied Security and Analytics degree. The module operationalizes secure development habits across both traditional software and machine learning (ML) pipelines. The module addresses a …

The Nist Artificial Intelligence Risk Management Framework: Adoption Challenges And Opportunities, Gillian Kennedy, Devin Patel, Humza Sheikh, Paul Wagner, Robert J. Honomichl

Journal of Cybersecurity Education, Research and Practice

Artificial intelligence (AI) is being adopted at an exponential rate to improve efficiency, decision-making, and cybersecurity, but its rapid integration introduces new and often poorly understood risks, including system errors, algorithmic bias, data privacy concerns, security vulnerabilities, and ethical dilemmas. This paper examines how organizations are implementing AI and evaluates the National Institute of Standards and Technology's AI Risk Management Framework (NIST AI RMF) as a tool for managing these risks. It reviews the benefits of AI adoption alongside the risks emerging from its use in business and broader society and examines the legal and ethical challenges organizations face when …

To What Extent Could Quantum Computing Pose A Threat To Global Modern Data Security?, Aniket Maheshwari

Journal of Cybersecurity Education, Research and Practice

Quantum computing has emerged as a transformative technology with the potential to fundamentally disrupt modern cryptographic systems that underpin global data security. This paper examines the extent to which quantum computing could pose a threat to modern global data security by synthesising existing technical, institutional, and policy-oriented literature. Drawing on a narrative review of scholarly research, industry reports, and government frameworks, the analysis focuses on the implications of quantum algorithms such as Shor’s and Grover’s, which challenge the mathematical foundations of widely used cryptographic schemes. The findings suggest that while quantum computing presents a credible long-term threat to asymmetric encryption …

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 …

The Intersection Between Mindfulness And Cybersecurity: A Tool To Reduce Burnout And Improve Operational Effectiveness, Ivo Ricardo Dias Rosa

Journal of Cybersecurity Education, Research and Practice

Abstract: This paper offers a conceptual discussion of how mindfulness, understood as present moment awareness and deliberate attention regulation, can support cybersecurity professionals. Drawing on a narrative synthesis of workplace mindfulness, burnout, and high pressure decision making literature, we map plausible self regulation mechanisms to typical cyber defense tasks. Rather than presenting new empirical data, we develop an explanatory framework linking attention, reactivity, and recovery to decision quality, team communication, and adherence to incident playbooks. We focus on two connected outcomes: reducing burnout in roles with sustained cognitive and emotional demands, and improving operational effectiveness during critical situations such as …

Predicting Cybermindfulness With The Cyber-Health Belief Model, James Robinson, Yan Tian, Thomas Skill

Journal of Cybersecurity Education, Research and Practice

This study describes the development of a Cyber-Health Belief Model (CHBM). The health belief model (HBM) is a message strategy that is widely and successfully used in public health research [1] and has been extended into phish training. Most phish training programs assume  end users are victimized because they have insufficient information to defend themselves. While near-term training effectiveness has shown to be effective, evidence for sustained behavioral change is thin [4]-[7].  This problem indicates that the traditional approaches need to be reconsidered and that new models are needed.  Recent research suggests attentional deficits, cyber-fatigue and fatalism and a sense …

Next-Generation Dns Rpz For Automated Threat Intelligence, Risk-Aware Filtering, And User-Centric Security, Jinu S, Kishore V. Krishnan, Rajarshi Middya, Annasamy Bagubali

Journal of Cybersecurity Education, Research and Practice

The Domain Name System (DNS) remains a critical attack vector exploited by adversaries for command-and-control (C2) communication, data exfiltration, and phishing campaigns. DNS Response Policy Zones (RPZ) have emerged as an effective defense mechanism by enabling the redirection or blocking of queries to malicious domains. However, current RPZ implementations encounter significant challenges related to scalability, adaptability, and user awareness, often resulting in static policies, delayed updates, and a high incidence of false positives. To address these limitations, this paper proposes an enhanced RPZ framework that integrates adaptive threat intelligence, machine learning-driven dynamic policy updates, and user-context-aware security controls. The proposed …

Cyber-Ready Libraries, Building Digital Fortresses For Tomorrow, Adeyinka B. Tella, Oluchi Precious Ogbonna Dr, Adebola Aderemi Olatoye Mrs

Journal of Cybersecurity Education, Research and Practice

Background and Purpose: Libraries are evolving into highly networked information ecosystems in this age of fast digital transformation, which increases their susceptibility to cybersecurity risks. The study "Cyber-Ready Libraries: Building Digital Fortresses for Tomorrow" looks into how prepared libraries are to face cyber threats and considers methods for creating information systems that are safe, robust, and ready for the future. To safeguard digital assets and user data, the study aims to assess existing cybersecurity practices in library settings and offer a roadmap for combining technological, human, and governance solutions.

Design/Method: The existing literature, case studies, and policy frameworks pertaining …

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 …

Sok: Understanding Zkvm: From Research To Practice, Guomin Yang, Yunbo Yang, Yuejia Cheng, Haibo Tang, Bingsheng Zhang, Kui Ren

Research Collection School Of Computing and Information Systems

Zero-knowledge virtual machine (zkVM) is a powerful infrastructure for proving the correctness of a program execution with a succinct proof, attracting significant interest from researchers, developers, and users. It has been widely used in applications such as blockchain rollups, privacy-preserving machine learning, and off-chain computation. As the field grows, a wide range of zkVMs have been proposed. However, they adopt different choices in instruction formats, trace layouts, and proving backends, which results in a highly heterogeneous design landscape and makes it difficult to understand the relations among these systems.To bridge this gap, we provide a comprehensive study of zkVMs that …

2026 Cyber-Resilient Health Care Workshop Report, Malcolm Schongalla, Sergey Bratus

Computer Science Technical Reports

The ISTS and the Dartmouth College Cybersecurity Cluster hosted the successful, inaugural Cyber-Resilient Health Care (CRHC) Workshop, March 5th & 6th, 2026. The event theme was "Innovation and Implementation," in response to the need to shift from reactive to proactive resiliency measures in the healthcare sector. Approximately 30 experts in clinical health care, cybersecurity, medical technology, policy, and innovation met to discuss solution-focused innovations addressing hard, cyber-related problems in health care. The agenda featured keynotes, an expert panel, innovation pitches, small group discussions, and a tabletop infrastructure disaster exercise. Participants gained insights into the obstacles and solutions involved in supporting …

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 …

Adversarial Robustness Of Perceptual Hashing Systems: A Unified Security Evaluation Framework, Avijit Roy

Student Theses

Social network platforms, child safety organizations, and image provenance systems use perceptual hashing to identify known child sexual abuse material (CSAM), support content moderation and reverse image search, and verify image integrity. Perceptual hashing works by producing similar fingerprints for visually similar images, even after common transformations such as compression, resizing, or minor brightness changes. This useful similarity-preserving property also creates an adversarial attack surface, as attackers can use AI-assisted or conventional image manipulation techniques to move a hash across a matching threshold while maintaining visual similarity, often without access to specialized hardware.

The security failures produced by adversarial attacks …

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 …