large language models

Network Flow-Based IoT Botnet Behavior Detection and Explanation Using Multi-Agent LLM Reasoning

IoT botnets generate diverse network behaviors, including scanning, command-and-control communication, and distributed denial-of-service activity. Traditional machine learn-ing models can achieve high detection accuracy using engineered network-flow features, but they often provide limited behav-ioral explanation. This...

Artificial Intelligence for Engineering-Grade Parametric CAD: A Critical Review of Representations, Large Language Models, and Agentic Design Systems

Artificial intelligence (AI) is rapidly expanding computer-aided design (CAD) from shape generation toward generative, conversational, and agentic engineering. However, most AI–CAD evaluations remain dominated by geometric similarity and execution success, while industrial CAD depends on editable parameters, design int...

The Missing Link: Why Legal Large Language Models Need Knowledge Graphs

The Impact of Patient Sociodemographic Characteristics on Primary Care Clinician Empathic Communication over Patient Portal Messages

Empathic communication in the clinical setting has been associated with lower blood sugar and cholesterol levels among diabetic patients, decreased anxiety, and increased satisfaction, treatment adherence, and trust. Despite recent growth in patient portal use, the expression of empathic communication through patient p...

Differential Privacy Tradeoffs in Large-Scale ML Model Training

Differential privacy (DP) provides the gold standard for formal privacy guarantees in machine learning, offering mathematically rigorous bounds on the information an adversary can infer about any individual training record from an observed model.However, the practical application of differential privacy to large-scale...

Advancing Machine Learning for Resource-Constrained Environments

This dissertation addresses the fundamental challenge of enabling machine learning intelligence on resource-constrained systems. These systems include embedded edge devices, mobile platforms, and robots operating under both on-device and cloud-based paradigms. The resource-constrained systems must perform low-latency,...

STOCHASTIC TRAJECTORY PROCESSES AND GRAPH NEURAL NETWORK LEARNING FOR ADAPTIVE COVERAGE IN UAV NETWORKS

This dissertation develops a framework for UAV trajectory design that bridges theoretical coverage guarantees and practical implementation constraints. The research introduces stochastic trajectory processes as a unifying paradigm, departing from deterministic optimization approaches that require precise user locations...

Towards Realistic Simulated Students: Aligning Language Models with Student Knowledge and Behavior

As large language models (LLMs) are increasingly used in education, there is a growing need to quickly verify and steer LLM-generated content for the benefit of students. In particular, it is important to ensure that educational AI systems improve learning outcomes and engagement while minimizing potential harms. While...

An Oriented Finite-Spin Vertex with Retained Tetrahedral Geometry: construction, canonical measure, and the gluing obstruction

We construct a finite-spin, orientation-resolved four-simplex vertex amplitude that retains the full tetrahedral intertwiner algebra. Orientation is selected by restricting the history integration to the domain on which a native Spin(4)-invariant pseudoscalar has fixed sign. Because the restriction acts on the integrat...

Advancing Responsible and Trustworthy AI Using Human-Centered Methods

Data-driven systems that use machine learning are now embedded in high-stakes domains including healthcare, hiring, and the criminal justice system, yet they routinely exhibit biases that cause real harm, generate stigmatizing language about marginalized groups, and encode discrimination into consequential decisions. A...

Natural Language Processing for Scalable Educational AI Systems

Recent advances in natural language processing (NLP), and in generative large language models (LLMs) in particular, have created new opportunities to develop scalable educational AI systems that support assessment, feedback, and personalized learning. Realizing this promise, however, requires more than applying general...

INTERPRETING RANKING MODELS: FROM FEATURE ATTRIBUTIONS TO MECHANISTIC COMPREHENSION

The rapid adoption of Neural Ranking Models (NRMs) in Information Retrieval (IR) underscores the urgent need for robust interpretability methods to ensure their reliability and transparency. Despite their widespread use, how NRMs model query-document relevance remains poorly understood, limiting their application in hi...