Software Engineering Commons^™

Large Language Model Communication And Data Transfer Across A Simulated Telephone Line, Jared Reyes

Dissertations and Theses

This thesis presents a proof-of-concept communication system in which two local large language model endpoints communicate across a simulated analog telephone line using legacy USB modems. The project combines modem voice mode, modem data mode, local speech processing, and structured machine messaging into a single staged session. During the voice phase, one endpoint places a call, the other answers, speech generated by a locally hosted LLaMA-family model is synthesized with Piper, transmitted through the modem voice path, captured on the remote side, and transcribed with Faster-Whisper to drive the next response. After the voice exchange, the system transitions to a …

Algorithmic Trading In Idiosyncratic-Payoff Markets: A Multi-Agent System For On-Chain Prediction Contracts, Saif Aldeen A.K. Agha

CMC Senior Theses

This thesis documents the design, deployment, and forward-test evaluation of an evolutionary multi-agent algorithmic trading system on Polymarket, the largest decentralized prediction market. The system pairs a locally-hosted 72-billion-parameter language model with a gradient-boosted statistical filter and an evolutionary selection mechanism that maintains a population of approximately 500 autonomous trading agents. Each agent generates a probability estimate for an event, compares it to the prevailing market price, and trades the resulting disagreement.

The central empirical exercise estimates a panel regression of trade-level profit on the absolute disagreement between the agent's probability estimate and the market price, controlling for agent identity, …

Do Comments And Expertise Still Matter? An Experiment On Programmers’ Adoption Of Ai-Generated Javascript Code, Changwen Li, Christoph Treude, Ofir Turel

Research Collection School Of Computing and Information Systems

This paper investigates the factors influencing programmers’ adoption of AI-generated JavaScript code recommendations within the context of lightweight, function-level programming tasks. It extends prior research by (1) utilizing objective (as opposed to the typically self-reported) measurements for programmers’ adoption of AI-generated code and (2) examining whether AI-generated comments added to code recommendations and development expertise drive AI-generated code adoption. We tested these potential drivers in an online experiment with 173 programmers. Participants were asked to answer some questions to demonstrate their level of development expertise. Then, they were asked to solve a LeetCode problem without AI support. After attempting to …

Do Developers Read Type Information? An Eye-Tracking Study On Typescript, Samuel W. Flint, Robert Dyer, Bonita Sharif

Research & Publications

Statically-annotated types have been shown to aid developers in a number of programming tasks, and this benefit holds true even when static type checking is not used. It is hypothesized that this is because developers use type annotations as in-code documentation. In this study, we aim to provide evidence that developers use type annotations as in-code documentation. Understanding this hypothesized use will help to understand how, and in what contexts, developers use type information; additionally, it may help to design better development tools and inform educational decisions. To provide this evidence, we conduct an eye tracking study with 26 undergraduate …

Distilling The Complexity Of Agent-Based Simulations Into Textual Explanations Via Large Language Models, Noé Y. Flandre, Philippe J. Giabbanelli

VMASC Publications

Communicating the design and results of agent-based models (ABMs) to subject matter experts is challenging, which hinders participation and limits trust in simulation-based decision support. Large language models (LLMs) can communicate ABMs as textual summaries, thus complementing traditional disclosure through statistical and visualization techniques. While prior work translated the structure of conceptual models into narratives via LLMs, our extension covers the dynamics of simulation models via an automated simulation-to-text method that extracts contextual information from NetLogo ABMs, performs repeated simulations, and generates narrative descriptions (including the model’s purpose, parameters, and simulation dynamics) using mutimodal LLMs. Furthermore, four summarization algorithms spanning …

Deep Learning Approaches For Voltammetric Analysis Of Coffee, Ryan Koes

Honors Theses

This thesis investigates deep learning approaches for voltammetric analysis of brewed coffee using a low-cost electrochemical system and screen-printed electrodes (SPEs). Traditional analytical methods, such as high-performance liquid chromatography (HPLC) and gas chromatography-mass spectrometry (GC-MS), provide precise quantification of key compounds but require expensive instrumentation and specialized expertise, limiting accessibility. While SPEs offer a more accessible alternative, they yielded poor results with traditional processing; however, when combined with a neural network, the system proved more effective. In experiments with 132 coffee samples, mean errors for caffeine, CGA, and TDS predictions were 52.98 ppm, 70.48 ppm, and 0.08%, respectively. These findings …

A Web-Based Wizard-Of-Oz Platform For Collaborative And Reproducible Human-Robot Interaction Research, Sean O'Connor

Honors Theses

The Wizard-of-Oz (WoZ) technique is widely used in Human-Robot Interaction (HRI) research, but two persistent problems limit its effectiveness: existing tools impose technical barriers that exclude non-engineering domain experts (the Accessibility Problem), and the fragmented landscape of robot-specific implementations makes interaction scripts difficult to port across platforms (the Reproducibility Problem- concerning execution consistency and portability, not third-party replication). Through a literature review, I identified three design principles to address both: a hierarchical specification model, an event-driven execution model, and a plugin architecture that decouples experiment logic from robot-specific implementations. I realized these principles in HRIStudio, an open-source, web-based platform providing …

An Empirical Framework For Evaluating Semantic Preservation Using Hugging Face, Nan Jia, Anita Raja, Raffi Khatchadourian

Publications and Research

As machine learning (ML) becomes an integral part of high-autonomy systems, it is critical to ensure the trustworthiness of learning-enabled software systems (LESS). Yet, the nondeterministic and run-time-defined semantics of ML complicate traditional software refactoring. We define semantic preservation in LESS as the property that optimizations of intelligent components do not alter the system's overall functional behavior. This paper introduces an empirical framework to evaluate semantic preservation in LESS by mining model evolution data from HuggingFace. We extract commit histories, $\textit{Model Cards}$, and performance metrics from a large number of models. To establish baselines, we conducted case studies in three …

Printer For Music Box, Chad Lewis, Caleb Murawski, Zion Smith, Bryan Tibbs

Williams Honors College, Honors Research Projects

The traditional method of creating music box sheet music involves manually punching holes into a paper strip using a hand-operated hole punch. This process involves precise knowledge of each note’s location and the ability to achieve perfect accuracy for hours.

The goal of this system is to automate this process, significantly reducing the time required while greatly improving the accuracy of the resulting music box playback. The user simply uploads a MIDI file of their choice into a user-friendly application. Here, the file is modified based on the user’s needs and sent to an automated hole-punching system to punch the …

A Decade Of Programming Languages: Trends In Popularity And Influence, Jonathan C. Erb

Williams Honors College, Honors Research Projects

Programming languages play a central role in open-source software ecosystems, yet their adoption, visibility, and influence shift over time as technologies, developer communities, and industry practices evolve. The study aims to investigate long-term trends in programming-language usage on GitHub from 2014 through 2024, focusing on ten major languages that represent diverse domains and ecosystems. Using repository metadata, engagement metrics such as stars and forks, and language-level code statistics measured with cloc, the analysis will examine changes in repository creation, code contribution volume, and popularity. Since popularity remains an unsettled and multidimensional concept, part of this research involves determining how it …

Automatic Generation Of Introductory Programming Exercises With Large Language Models, Nguyen Binh Duong Ta, Hua Gia Phuc Nguyen, Gottipati Swapna

Research Collection School Of Computing and Information Systems

Despite recent advances in code generation made possible by large language models (LLMs), programming is still an essential skill that computing students need to master now and in the foreseeable future. In learning programming, frequent practices with exercises set at an appropriate difficulty and knowledge level is of crucial importance for students. However, it’s not a trivial task for instructors to create many good quality exercises customized for each student. Programming problems found on Internet sources such as LeetCode are mostly too challenging for novice programmers with no prior coding knowledge. Recent work in AI-enabled education has been leveraging LLMs …

A Case Study Of Gender And Online Team Communication In Software Engineering Education, Rita Garcia, Christoph Treude

Research Collection School Of Computing and Information Systems

Collaboration is crucial in Software Engineering (SE), yet factors like gender bias can shape team dynamics and behaviours. This descriptive case study examines an eight-week project involving 39 SE students across eight teams contributing to GitHub projects. Focusing on gender, we used a mixed-methods approach to analyse Slack communications, identifying gender differences in how students respond to initiated communications and comparing how students’ communications influenced other aspects of students’ performance, including learning gains. We found higher help-seeking and leadership behaviours in the all-woman team involved in this case study, while men responded more slowly. Although communication did not directly affect …

Gui Test Migration Via Abstraction And Concretization, Yakun Zhang, Chen Liu, Xiaofei Xie, Yun Lin, Jin Song Dong, Dan Hao, Lu Zhang

Research Collection School Of Computing and Information Systems

GUI test migration aims to produce test cases with events and assertions to test specific functionalities of a target app. Existing migration approaches typically focus on the widget-mapping paradigm that maps widgets from source apps to target apps. However, since different apps may implement the same functionality in different ways, direct mapping may result in incomplete or buggy test cases, thus significantly impacting the effectiveness of testing the target functionality and the practical applicability of migration approaches.In this article, we propose a new migration paradigm (i.e., the abstraction-concretization paradigm) that first abstracts the test logic for the target functionality and …

Dual-Lora And Quality-Enhanced Pseudo Replay For Multimodal Continual Food Learning, Xinlan Wu, Bin Zhu, Feng Han, Pengkun Jiao, Jingling Chen

Research Collection School Of Computing and Information Systems

Food analysis has become increasingly critical for health-related tasks such as personalized nutrition and chronic disease prevention. However, existing large multimodal models (LMMs) in food analysis suffer from catastrophic forgetting when learning new tasks, requiring costly retraining from scratch. To address this, we propose a novel continual learning framework for multimodal food learning, integrating a Dual-LoRA architecture with Quality-Enhanced Pseudo Replay. We introduce two complementary low-rank adapters for each task: a specialized LoRA that learns task-specific knowledge with orthogonal constraints to previous tasks’ subspaces, and a cooperative LoRA that consolidates shared knowledge across tasks via pseudo replay. To improve the …

Towards Secure Program Partitioning For Smart Contracts With Llm’S In-Context Learning, Ye Liu, Yuqing Niu, Chengyan Ma, Ruidong Han, Wei Ma, Yi Li, Debin Gao, David Lo

Research Collection School Of Computing and Information Systems

Smart contracts are highly susceptible to manipulation attacks due to the leakage of sensitive information. Addressing manipulation vulnerabilities is particularly challenging because they stem from inherent data confidentiality issues rather than straightforward implementation bugs. To tackle this by preventing sensitive information leakage, we present PARTITIONGPT, the first LLM-driven approach that combines static analysis with the in-context learning capabilities of large language models (LLMs) to partition smart contracts into critical (privileged) and normal codebases, guided by a few annotated sensitive data variables. We evaluated PARTITIONGPT on 18 annotated smart contracts containing 99 sensitive functions. The results demonstrate that PARTITIONGPT successfully generates …

When Slides Replace Thinking: How Powerpoint Culture Undermines Conceptual Learning In Engineering Education, Shahin Nayyeri Amiri, Hamid Eisazadeh

Civil & Environmental Engineering Faculty Publications

Engineering education has gradually shifted from deep conceptual teaching to a culture dominated by presentation slides. In many universities today, core courses such as Statics, Dynamics, and Mechanics of Materials are taught through PowerPoint files rather than through reasoning, derivation, and problem solving. While slide-based instruction may appear efficient, it encourages passive learning and has severely weakened students’ ability to think independently and make sound engineering judgments. Traditionally, these fundamental courses were taught by seasoned instructors who had mastered the subjects and took genuine pride in explaining complex ideas with clarity and passion. They guided students through the logic behind …

Near Real-Time Adaptive Isotropic And Anisotropic Image-To-Mesh Conversion For Cerebral Aneurysm Simulations, Kevin Garner, Chander Sadasivan, Nikos Chrisochoides

Computer Science Faculty Publications

This paper presents two performance optimization techniques for a mesh adaptation method that is designed to help streamline the discretization of complex vascular geometries within the numerical modeling process. This method is integrated into a pipeline with an image-to-mesh conversion tool to generate adaptive anisotropic meshes from segmented medical images. The pipeline is shown to satisfy quality, fidelity, smoothness, and robustness requirements while providing near real-time performance for medical image-to-mesh conversion. Tested with two brain aneurysm cases and utilizing up to 96 CPU cores within a single, multicore node on Purdue University’s Anvil supercomputer, the parallel adaptive anisotropic meshing method …

Distributed Semi-Speculative Parallel Anisotropic Mesh Adaptation, Kevin Garner, Polykarpos Thomadakis, Nikos Chrisochoides

Computer Science Faculty Publications

This paper presents a distributed memory method for anisotropic mesh adaptation that is designed to avoid the use of collective communication and global synchronization techniques. In the presented method, meshing functionality is separated from performance aspects by utilizing a separate entity for each - a multicore cc-NUMA-based (shared memory) mesh generation software and a parallel runtime system that is designed to help applications leverage the concurrency offered by emerging high-performance computing (HPC) architectures. First, an initial mesh is decomposed and its interface elements (subdomain boundaries) are adapted on a single multicore node (shared memory). Subdomains are then distributed among the …

Real-Time Immersive Wildfire Simulation: A Game Engine Approach, Jake Bova

Graduate Student Theses, Dissertations, & Professional Papers

Wildfires are growing in frequency, size, and severity, placing escalating demands on firefighting resources and putting lives, property, and ecosystems at risk. Accurate prediction of fire behavior is critical to suppression planning, evacuation decisions, and firefighter training, yet the operational models that fire managers rely upon face a persistent tradeoff. High-fidelity physics-based models capture fire–atmosphere interactions faithfully but are computationally expensive, often running far slower than real time on CPU hardware, while faster empirical models sacrifice the dynamic coupling between a fire and the wind field it generates. Furthermore, most existing tools produce abstract, two-dimensional outputs that offer limited visual …

Error-Driven Density Control For Compact Gaussian Splatting Under Sparse Supervision, Abdelrhman Elrawy

Theses and Dissertations (Comprehensive)

This thesis studies efficiency and stability challenges in Gaussian-splatting-based reconstruction under sparse supervision. In few-shot novel view synthesis, standard 3D Gaussian Splatting (3DGS) can overfit the limited training views and grow an unnecessarily large number of primitives due to limitations in its Adaptive Density Control (ADC) mechanism. This thesis introduces an error-driven reformulation of ADC that triggers densification using opacity gradients as a lightweight proxy for rendering error, and shows that such aggressive densification must be paired with delayed and conservative pruning to prevent destructive create--destroy cycles. When combined with depth-based geometric regularization, the resulting framework produces substantially more compact …

Geovig And Purevig: Geometry-Aware Architectures For Efficient Computer Vision, Omar Ismail

Theses and Dissertations (Comprehensive)

Deploying deep learning models for medical image analysis on mobile devices requires a balance between inference latency, memory footprint, and delineating anatomical boundaries with high accuracy. While Convolutional Neural Networks (CNNs) and mobile Vision Transformers (ViTs) offer efficiency, they often struggle to model the irregular, non-local geometric structures inherent in biological tissues without incurring prohibitive computational costs. In this thesis, we introduce GeoViG (Geometric Vision Graph), an architecture that bridges the gap between efficient grid-based processing and explicit Geometric Deep Learning. GeoViG introduces a novel transition from high-resolution pixel grids to low-resolution dynamic graphs via a SpreadEdgePool operator, a geometry-aware …

Llm-Driven Weekly Newsletter To Assess Open Source Software Project Github Health, Christian Novalski, Christopher Chavez, Ghalian Fayyadh, Kostadin Damevski

UROP Posters

Open Source Software (OSS) projects increasingly depend on a diverse set of contributors, including episodic participants who contribute intermittently. Episodic contributors represent a large portion of OSS communities, yet projects often struggle to retain them, leading to decreased project health and continuity. While dashboards and real-time communication tools support continuously active contributors, they often fail to serve the unique needs of episodic participants, who may struggle to remain informed and re-engage with project activity after periods of absence. In this study, we examine the effect of a weekly, email-based newsletter intervention designed to improve awareness and engagement among episodic OSS …

Evaluating Large Language Models For Line-Level Vulnerability Localization, Jian Zhang, Chong Wang, Anran Li, Weisong Sun, Cen Zhang, Wei Ma, Yang Liu

Research Collection School Of Computing and Information Systems

Recently, Automated Vulnerability Localization (AVL) has attracted growing attention, aiming to facilitate diagnosis by pinpointing the specific lines of code responsible for vulnerabilities. Large Language Models (LLMs) have shown potential in various domains, yet their effectiveness in line-level vulnerability localization remains underexplored. In this work, we present the first comprehensive empirical evaluation of LLMs for AVL. Our study examines 19 leading LLMs suitable for code analysis, including ChatGPT and multiple open-source models, spanning encoder-only, encoder-decoder, and decoder-only architectures, with model sizes from 60M to 70B parameters. We evaluate three paradigms including few-shot prompting, discriminative fine-tuning, and generative fine-tuning with and …

Generalized Visual Relation Detection With Diffusion Models, Kaifeng Gao, Siqi Chen, Hanwang Zhang, Jun Xiao, Yueting Zhuang, Qianru Sun

Research Collection School Of Computing and Information Systems

Visual relation detection (VRD) aims to identify relationships (or interactions) between object pairs in an image. Although recent VRD models have achieved impressive performance, they are all restricted to pre-defined relation categories, while failing to consider the semantic ambiguity characteristic of visual relations. Unlike objects, the appearance of visual relations is always subtle and can be described by multiple predicate words from different perspectives, e.g., “ride” can be depicted as “race” and “sit on”, from the sports and spatial position views, respectively. To this end, we propose to model visual relations as continuous embeddings, and design diffusion models to achieve …

Benchmarking Gaslighting Negation Attacks Against Reasoning Models, Bin Zhu, Hailong Yin, Jingjing Chen, Yu Gang Jiang

Research Collection School Of Computing and Information Systems

Recent advances in reasoning-centric models promise improved robustness through mechanisms such as chain-of-thought prompting and test-time scaling. However, their ability to withstand gaslighting negation attacks—adversarial prompts that confidently deny correct answers—remains underexplored. In this paper, we conduct a systematic evaluation of three state-of-the-art reasoning models, i.e., OpenAI’s o4-mini, Claude-3.7-Sonnet and Gemini-2.5-Flash, across three multimodal benchmarks: MMMU, MathVista, and CharXiv. Our evaluation reveals significant accuracy drops (25–29% on average) following gaslighting negation attacks, indicating that even top-tier reasoning models struggle to preserve correct answers under manipulative user feedback. Built upon the insights of the evaluation and to further probe this vulnerability, …