Artificial Intelligence and Robotics Commons^™

Probing Representational Emergence In Large Language Models, Shawn Ismail

Master's Theses

This thesis investigates whether abrupt behavioral gains in large language models under scaling are accompanied by systematic changes in internal representations. It combines a behavioral screen of 65 tasks per family with targeted layerwise probing across eight decoder-only, open-weight model families. Behavioral emergence is defined for each family-task trajectory using an empirical jump detector, with segmented regression retained only as a diagnostic. The representational follow-up analyzes 27 selected MMLU subtasks shared across all families, spanning 37 checkpoints and 216 family-task units.

For each follow-up checkpoint, frozen linear probes are trained on every layer's hidden states to measure how much task-relevant …

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 …

Efficient Deep Neural Networks For Autonomous Perception, Reeshad Khan

Graduate Theses and Dissertations

Autonomous perception systems must operate reliably under uncertainty arising from noisy observations, incomplete supervision, and hardware constraints. This dissertation investigates the design of efficient deep neural networks for autonomous perception through a unified perspective that treats uncertainty, efficiency, and sensing as interconnected challenges. The first contribution develops adaptive extensions of unbiased risk estimators, including eSURE and ePURE, enabling unsupervised training of deep neural networks for magnetic resonance image denoising under Gaussian and Poisson noise. However, these methods rely on known noise assumptions, which motivates the second contribution: a unified diffusion and Bayesian risk framework that estimates and adapts to unknown …

Toward Causal Generative Modeling: From Representation Learning To Controllable Generation, Aneesh Komanduri

Graduate Theses and Dissertations

The hallmark of human intelligence is causal reasoning, the ability to infer relationships between causes and effects through observation and intervention. While modern deep learning has excelled at identifying statistical patterns, current generative models often struggle to capture the underlying structural causal mechanisms of the data-generating process, leaving them vulnerable to shortcut learning and spurious associations. To achieve true generalizability and interpretability, artificial intelligence must transition from simple association to higher-level causal reasoning to be capable of scheduling and planning in the real world. This dissertation develops fundamental methodologies for causal generative modeling by integrating Pearl’s Structural Causal Model (SCM) …

Mease: Multi-Agent Episodic Action Sequence Explanation, Phyo Wai Khaing, Minghong Geng, Shubham Pateria, Budhitama Subagdja, Ah-Hwee Tan

Research Collection School Of Computing and Information Systems

Multi-agent reinforcement learning (MARL) achieves remarkable performance in complex coordination tasks, yet interpreting the emergent behaviors of trained agents remains a fundamental challenge. Most current explainability methods focus on individual agent decisions, overlooking the critical interplay of joint strategiesand temporal coordination patterns that define successful multi-agent policies. We present MEASE (Multi-agent Episodic Action Sequence Explanation), a novel explainable MARL (XMARL) framework that explains trained MARL policies as human-interpretable emergent cooperative joint behaviors. MEASE employs a cognition-inspired episodic memory model to learn spatio-temporal multi-agent interaction patterns, coupled with abstraction algorithms that identify significant cooperative agent behaviors. We evaluate MEASE on diverse …

Synthesis And Evaluation Of Long-Term History-Aware Medical Dialogue, Hebin Hu, Renke Dai, Ah-Hwee Tan, Yilin Kang

Research Collection School Of Computing and Information Systems

An effective healthcare agent must be able to recall and reason over a patient’s longitudinal medical history. However, the absence of datasets with realistic long-term dialogue timelines limits systematic evaluation. Real clinical text is constrained by privacy and ethics, while existing benchmarks focus on isolated interactions, failing to capture cross-session reasoning. We introduce a framework for synthesizing high-quality, long-term medical dialogues with LLMs. Our approach entails a knowledge-guided decomposition into three stages: constructing synthetic patient profiles with diverse disease and complication trajectories, generating multiturn dialogues per encounter, and integrating them into a coherent longitudinal history dataset, MediLongChat. We establish three …

Market Reactions To Deceptive Language In Fake News: Implications From Language Expectancy Theory And Transfer Learning, Ka Chung Ng, Ping Fan Ke, Ping Fan, Mike So, Tam, Kar Yan

Research Collection School Of Computing and Information Systems

The advent of generative artificial intelligence (AI) has heightened the proliferation of fake news. A key challenge is the limited real-world data to investigate the societal impact of fake news produced by generative AI. In this paper, we examine stock market reactions to financial news articles that exhibit stylometric similarity to human-crafted and AI-crafted fake financial news. Grounded in language expectancy theory, we employ a style-based transfer learning model, pre-trained to recognizing deceptive language employed in various types of fake news intricacies. We then apply this model to a comprehensive dataset of financial news, assigning a “veracity style score” to …

Fighting Against Recruitment Scams: Theory-Driven Supervised Learning And Empirical Analysis For Digital Fraudulent Recruitment Posting Behavior, Tom (Tianteng) Wang, David (Jingjun) Xu, Keng Siau, Zhongju (John) Zhang

Research Collection School Of Computing and Information Systems

The number of recruitment postings on digital recruitment hiring platforms has increased since the COVID-19 pandemic. However, the weak surveillance and operations of these platforms, combined with the fact that most job seekers have relatively low vigilance and a strong desire for recruitment offers, enable scammers to easily deceive job seekers for their money and confidential information. In this work, we combine prevailing text mining techniques (i.e., ChatGPT with prompting engineering and supervised machine learning) with interpersonal deception theory (IDT) from social science to design an interpretable IT system to predict fraudulent recruitment postings on digital recruitment-hiring platforms. We compare …

Enhancing Action And Ingredient Modeling For Semantically Grounded Recipe Generation, Guoshan Liu, Bin Zhu, Yian Li, Jingjing Chen, Chong-Wah Ngo, Yu-Gang Jiang

Research Collection School Of Computing and Information Systems

Recent advances in Multimodal Large Language Models (MLMMs) have enabled recipe generation from food images, yet outputs often contain semantically incorrect actions or ingredients despite high lexical scores (e.g., BLEU, ROUGE). To address this gap, we propose a semantically grounded framework that predicts and validates actions and ingredients as internal context for instruction generation. Our two-stage pipeline combines supervised fine-tuning (SFT) with reinforcement fine-tuning (RFT): SFT builds foundational accuracy using an Action-Reasoning dataset and ingredient corpus, while RFT employs frequency-aware rewards to improve long-tail action prediction and ingredient generalization. A Semantic Confidence Scoring and Rectification (SCSR) module further filters and …

Selective Concolic Testing, Guofeng Zhang, Zhenbang Chen, Ziqi Shuai, Jun Sun, Weijiang Hong, Yufeng Zhang, Ji Wang, Yang Liu

Research Collection School Of Computing and Information Systems

The principled combination of symbolic execution and random testing lacks a formal foundation, especially in deciding which inputs to symbolize. We propose selective concolic testing, a cost-aware framework that formulates this choice as an optimized policy problem of a MDP (Markov Decision Process). We model program exploration over a finite control-flow graph, where MDP states represent covered statements, actions partition path constraints into symbolic and random fragments, rewards reflect coverage gain, and costs account for SMT solving effort and sampling inefficiency. Our framework yields the first formal characterization of selective symbolization as policy synthesis in a probabilistic system. We prove …

Detecting Doubt In Reflective Learning: A Learning Analytics Study With Large And Small Language Models, Eng Lieh Ouh, Kar Way Tan, Siaw Ling Lo, Yuhao Zhang

Research Collection School Of Computing and Information Systems

Reflective learning enhances understanding, especially when instructors promptly address difficulties raised in student reflections. Automated doubt detection can reduce time for instructors, yet existing classification approaches take substantial time for manual annotation and model training. This paper investigates whether large and small language models (LLMs, SLMs) can automate doubt detection without time-consuming training. Using a dataset of anonymized student reflections, we evaluate zeroshot, few-shot prompting, and multi-step reasoning against prior supervised classification baselines. We show that LLMs (GPT-4o, Claude-4, Gemini-2.5) surpass earlier F1 scores without prompting, while prompting further improves their performance. However, using proprietary LLMs can raise cost and …

Gencode: A Generic Data Augmentation Framework For Boosting Deep Learning-Based Code Understanding, Zeming Dong, Qiang Hu, Xiaofei Xie, Maxime Cordy, Mike Papadakis, Yves Le Traon, Jianjun Zhao

Research Collection School Of Computing and Information Systems

Pre-trained code models lead the era of code intelligence, with multiple models designed with impressive performance. However, one important problem, data augmentation for code data that automatically helps developers prepare training data lacks study in this field. In this paper, we introduce a generic data augmentation framework, GenCode, to enhance the training of code understanding models. Simply speaking, GenCode follows a generation-and-selection paradigm to prepare useful training code data. Specifically, it employs code augmentation techniques to generate new code candidates first and then identifies important ones as the training data by influence scores. To evaluate the effectiveness of GenCode, we …

Learning Global Context For Sparse Activity Recognition In Lengthy Recordings With Limited Dataset Size, Zeyu Tang

All Dissertations

This dissertation describes methods to analyze lengthy recordings of data in order to detect sparsely occurring activities. The narrative below describes the progression of research that led to the development of these methods and their generalization into a unified framework. My research started with designing models for dietary monitoring, including detecting meals from day-long recordings and detecting intake gestures from meal-length recordings. Both tasks share some common characteristics: (a) the target event takes only a small portion of data recordings, and (b) there is global context within full-length data recordings that can help a model make better decisions. After finishing …

Conditioning Hierarchical Diffusion Transformers On Rf Circuit Parameters, Ethan Morton

McKelvey School of Engineering Graduate Student Theses & Dissertations

As modern ML techniques have become increasingly advanced, they have begun to be integrated into wireless RF systems for classification, identification, and spectrum management. Deep Neural Networks (DNNs) enable RF system operators and designers to design more flexible systems with greater robustness to errors and attacks. However, neural networks require significant amounts of properly annotated data to train. Current data labeling methods lack the ability to obtain reliable true labels for circuit properties such as carrier frequency offset (CFO), power amplifier (PA) non-linearity, and in-phase/quadrature (IQ) imbalance. This thesis investigates the efficacy of a novel architecture, RF-Diffusion, for generating high-quality …

A Generative Ai Method For Minority Class Handling In Anomaly Detection With Drift And Explainability Analysis, Kelvin J. Mwiga, Mussa A. Dida, Ahmad Mohsin, Iqbal H. Sarker

Research outputs 2022 to 2026

Artificial Intelligence, particularly machine learning (ML) algorithms, plays a crucial role in detecting cyberattacks, including anomalies and intrusions. However, machine learning models trained on imbalanced cybersecurity datasets often struggle to accurately detect minority data instances and potential threats, thereby weakening overall system security. Despite extensive research, a persistent challenge is the inadequate explanation for model predictions concerning minority data classes. This study aims to address these limitations by developing a generative AI-based approach to manage minority classes in anomaly detection, incorporating concept drift handling and explainability analysis. We introduce an over-sampling technique, CGGReaT, designed to enhance the presence of minority …

Privacy Protection In Machine Learning: Methods For Structured And Unstructured Data, Karuna Bhaila

Graduate Theses and Dissertations

As machine learning models become increasingly integrated into data-driven decision-making, the protection of sensitive information throughout the model lifecycle is a paramount concern. As these models process and memorize sensitive, proprietary, or personal data, they risk leaking information through their outputs or internal states, especially in domains such as healthcare and finance. The protection of data in machine learning has thus been a crucial field of study. Within this paradigm, researchers have studied theoretical and application-oriented mechanisms for realizing privacy protections for various data formats. Nonetheless, privacy in machine learning still has many open problems, especially with the emergence of …

From Data Digitization To Personalized Care: Deep Learning And Decision Analytics In Healthcare Systems, Ahla Ko

Graduate Theses and Dissertations

This dissertation traces a progression from foundational data infrastructure to individualized clinical decision-making, developing and evaluating three computational frameworks that advance healthcare efficiency and personalization through deep learning and decision analytics. The first study presents an end-to-end pipeline for automated recognition of handwritten medical forms, addressing persistent challenges in health data digitization. Integrating a YOLO-based field detection model, a Convolutional Recurrent Neural Network (CRNN) for text transcription, and a confidence-based human-in-the-loop quality assurance framework, the system achieved 99.75\% Exact Match Accuracy and a 0.13\% Character Error Rate on medical forms collected from a tuberculosis research project in Moldova. A GPU-accelerated …

Aircraft Fault Detection Via Weight And Bias Analysis Of A Custom First Neural Network Layer, George Harrison Chen

Theses and Dissertations

Fault detection in aircraft is traditionally handled through redundant hardware and comparison algorithms to detect failures. Alternatives like model-based residual generation and data-driven approaches such as supervised fault classification and unsupervised anomaly detection have been explored, but they suffer from practical limitations; model-based methods require accurate system models, and data-driven methods have large constraints on the data limiting scalability and adaptability. This work presents a purely data-driven neural network architecture featuring a custom first layer designed for real-time fault detection where the weights and biases of this layer are used to detect faults. The network requires zero supervision and complements …

Adaptive Artificial Potential Field Guidance And Control For Autonomous Docking With Uncooperative And Unknown Spacecraft, Steven Holmberg

Theses and Dissertations

The increasing demand for on-orbit servicing (OOS), active debris removal (ADR), and space domain awareness (SDA) missions has increased the need for autonomous spacecraft rendezvous and proximity operations (RPO) with uncooperative and unknown targets. Traditional guidance and control methods are typically designed for cooperative systems with known geometry and state information. This work builds on previous research to develop and evaluate an artificial potential field (APF)-based control framework capable of autonomous operation with minimal prior target knowledge and applicability to both relatively static and tumbling spacecraft.

The proposed APF formulation incorporates established safety constructs from cooperative docking systems, including an …

Dementia Detection In Low-Resource Languages: Evaluating Translation-Assisted Transfer Learning For Multilingual Clinical Assessment, Kylar A. Deloach

Honors Theses

Alzheimer's disease (AD) is a growing global health concern, with millions of people affected worldwide and cases expected to rise significantly in the coming decades. Early detection is critical for patient treatment and care, and recent advances in natural language processing (NLP) have shown promise in identifying linguistic markers associated with AD. However, most existing work has focused on English, leaving speakers of other languages with limited access to such tools. This study investigates how effective AD detection models trained on English data are at transferring to Greek, a low-resource language with limited dementia-related speech data available. We propose a …

Ai-Driven Multispectral Drone Monitoring For Afforestation In Arid Environments, Hesham Morgan, Ali Elgendy, Brandon Tran, Tamer Ismail, Mohamed M. Moursy, Yehya Kh. Shehadeh, Ahmed Elgharib, Ahmed Abdullah Al-Dughairi, Ali El Mubarak, Khaled Allam Harhash, Hesham El-Askary

Mathematics, Physics, and Computer Science Faculty Articles and Research

Monitoring large-scale afforestation projects in arid and semi-arid environments requires accurate, high-resolution, and repeatable methods to assess tree survival and growth. In this study, we integrated unmanned aerial vehicle (UAV) multispectral imaging with an advanced object detection framework to evaluate vegetation establishment in the Shuayb Al-Budai afforestation site, part of the Imam Turki bin Abdullah Royal Natural Reserve, Kingdom of Saudi Arabia (KSA). Multispectral datasets were acquired using a MicaSense Altum-PT sensor and processed through a masked Region-based Convolutional Neural Network (RCNN) with two backbone architectures: ResNet-101 and VGG19-BN. The Mask R-CNN–ResNet-101 model achieved superior performance, with an overall accuracy …

Ai Literacy: An Annotated Oer Bibliography, Houy Yvonne

UNLV Best Teaching Practices Expo

"Scalable, discipline-agnostic AI literacy instruction can be implemented incrementally without requiring full course redesign in higher education, supporting both technical understanding and critical engagement with the social and ethical dimensions of AI: The curated list of open educational resources (OER) on this poster enable a flexible, modular approach to teaching foundational AI literacy. The annotated list includes self-paced, hands-on projects with complementary educator-guided activities and discussion to support conceptual understanding of machine learning, training data, and algorithmic bias, drawing on OER such as Code.org’s AI curriculum, MIT RAISE’s Day of AI, and the multi-lingual Elements of AI course. Many resources …

A Deep Learning-Based Approach For Bot Detection In Trending Hashtags On X, Mehboob Hussain, Muhammad Rizwan Rashid Rana, Muhammad Imran, Muhammad Shoaib, Muhammad Hasaan Mujtaba

Makara Journal of Technology

The widespread presence of bots on social media platforms, such as X (formerly Twitter), poses a significant threat to the integrity of online information by facilitating the dissemination of misinformation and manipulating public discourse. This study proposes a robust deep learning-based framework, DeepBot, to detect bot participation in trending hashtags and discussions on X. The approach uses a dataset sourced from Kaggle, comprising user profile metadata, including follower count, tweet frequency, account verification status, and engagement metrics. The data were subjected to comprehensive preprocessing, including noise removal, part-of-speech (POS) tagging, and word embedding using the pre-trained GloVe model. RoBERTa is …

Cognition Is Not Content: A Structural Account Of Processing Conditions, Griselda Poe

Publications and Research

Human cognition has been described in terms of content. This description becomes insufficient once artificial systems make output observable apart from subject attribution, intention, and relational context. Under this contrast condition, what becomes visible is a layered structure in which output, reconstruction, evaluation, and termination do not necessarily coincide. The same input may register as complete under one processing condition while remaining unresolved under another. This separability means that content-based description does not merely omit an additional variable: it can mislocate a processing difference as a difference in meaning, personality, intention, ability, or attitude. This is not a proposal for …

Dynamic-Query Robustness Of Ann Indexes Under Time-Indexed Drift, Stellamaris Nakacwa, Majid Shaalan

Harrisburg University Other Works

Approximate nearest-neighbor search is a central retrieval primitive in dense question-answering and retrieval-augmented generation systems. Existing ANN evaluation protocols typically measure recall, latency, throughput, and search-effort sensitivity under a fixed-query assumption: a query vector is submitted to an index, approximate neighbors are retrieved, and the result is compared with exact nearest-neighbour ground truth. This assumption is appropriate for conventional vector-search benchmarking, but it is less complete for multi-step, distributed, and agent-controlled retrieval pipelines in which the retrieval-facing query may be refined, recomputed, or displaced across execution steps. This paper introduces a time-driven dynamic query evaluation framework for ANN search. The …

The Role Of Human-Centered Artificial Intelligence In Crewed Mars Missions, Jordan Neuman

Student Publications and Presentations

The suggestion of implementing Artificial Intelligence (AI) in crewed missions to the Martian surface can revolutionize mission operations, psychological support, and ethical decision‑making under challenging conditions. This study is a structured, qualitative literature review of current AI research for the future of human space exploration. It uses publicly available peer‑reviewed journal articles and NASA/ESA technical reports on AI for Mars missions. Using Perplexity, the researcher first prompted the AI tool to identify recurring topics across these sources and generate themes. The researcher independently reviewed all sources to confirm and refine those themes. Three themes—mission autonomy, local resource utilization and life‑support, …

Temperature-Induced Uncertainty In Fixed-Context Retrieval-Augmented Generation, Steven Zeng, Murat Kuzlu

Cybersecurity Undergraduate Research Showcase

This study examines how decoding temperature affects output uncertainty in a fixed-context retrieval-augmented generation (RAG) system. We define uncertainty as the semantic dispersion among repeated answers under the same fixed retrieved context, with greater dispersion interpreted as higher uncertainty. To isolate this answer-generation variability from retrieval drift, each question was paired with a fixed retrieved context, and repeated generations differed only in temperature. The experiment used nine questions drawn from a machine-learning textbook corpus, with three questions each at easy, moderate, and hard difficulty. Each question was evaluated at five temperatures (0.0, 0.25, 0.5, 0.75, and 1.0) over 30 iterations, …

Modelling Method Of Unmanned Vehicle Dynamics Based On Neural Network, Jun Wang, Min Liu, Xiaochuan Zhang, Yishan Ding, Juhui Feng, Ye Zhuang

Journal of System Simulation

Abstract: To address the challenges of high data acquisition costs of test data on dynamic characteristics between tires and soft terrain and low speed of numerical calculation for unmanned vehicles in complex terrestrial environments, a modeling method of unmanned vehicle dynamics based on a neural network was proposed. Tire-terrain contact dynamics models were built by using discrete element method (DEM) simulations for tire-terrain contact and experimental data, thereby creating a dataset of tire contact forces for various tire materials in terrestrial environments. The neural network was applied to regressively learn the dataset, and a nonlinear neural network tire model was …

Method For Testing And Evaluating Intelligence Level Of Virtual Forces Based On Operational Experiments, Dayong Liu, Zhiming Dong, Weidong Zhang, Wenjun Zhang, Jiancheng Gao

Journal of System Simulation

Abstract: The intelligence level of virtual forces is a key factor affecting the credibility and effectiveness of tactical confrontation simulations. To address the current lack of a testing and evaluation system, a method for testing and evaluating the intelligence level of virtual forces based on operational experiments is proposed. Guided by operational experiment theory, the method stimulates the intelligent behavior of virtual forces by constructing dynamic confrontation environments, and collects, calculates, analyzes, and evaluates their intelligence performance data according to a systematic process. The overall architecture, logical functional modules, and basic evaluation process of the method are designed. A "4M" …

Research On Control Of Coaxial Dual-Rotor Unmanned Aerial Vehicle Based On Improved Reaching Law, Xinhang Chen, Xiaodong Ling, Chengchang Lang, Shijun Zheng, Yiqi Tang

Journal of System Simulation

Abstract: To address the issues of significant chattering, slow convergence speed, and large overshoot in the control system of a coaxial dual-rotor unmanned aerial vehicle, a control method based on an improved double-power and hyperbolic function integral sliding mode reaching law was proposed. A novel reaching law was designed to achieve fast convergence when the system state is far from the sliding surface and smooth transition when approaching the sliding surface, thereby enhancing the overall convergence speed of the system and ensuring that the system reaches the sliding surface within a finite time. The saturation characteristic of the hyperbolic function …