VTechWorks
VTechWorks provides global access to Virginia Tech scholarship, including journal articles, books, theses, dissertations, conference papers, slide presentations, technical reports, working papers, administrative documents, videos, images, and more by faculty, students, and staff. Faculty can deposit items to VTechWorks from Elements, including journal articles covered by the University open access policy. Email vtechworks@vt.edu for help.
Open Access Policy
Virginia Tech's open access policy enables researchers to deposit the accepted version of scholarly articles with no embargo.
Theses and Dissertations
Virginia Tech was first in the world to require ETDs in 1997, and continues to add scans of older theses and dissertations.
Open Textbooks
More than 50 freely available and openly licensed textbooks are among our most downloaded items.
Communities in VTechWorks
Select a community to browse its collections.
Recent Submissions
Exploring essential aspects of chemotaxis and motility in rhizosphere-dwelling bacteria
Collett, Katelyn Idella Elizabeth (Virginia Tech, 2026-08-26)
The agricultural industry faces a constant struggle between generating profit from crop yield and the cost of production in terms of monetary loss and environmental impact. A potential solution comes from bacteria that are capable of forming a symbiotic relationship with plants to deliver nitrogen, thus alleviating the need for synthetic fertilizers that are used in excess. In addition, there are bacterial pathogens that cause disease in crops leading to reduction in yield, exacerbating the financial strain on farmers. Both groups of bacteria are capable of identifying potential hosts and biasing their movement towards them through chemotaxis and flagellar-driven motility. Key aspects for different portions of this process were examined in three agriculturally-relevant bacteria: Bradyrhizobium diazoefficiens (symbiont), Sinorhizobium meliloti (symbiont), and Agrobacterium tumefaciens (pathogen). Chapter I encompasses a literatures review of the knowledge for the chemotactic pathway and flagellar structure in the target bacteria. Additionally, it highlights the gaps in knowledge that this study set to characterize. In Chapter II we detail the investigation of the ligand binding capabilities of B. diazoefficiens chemoreceptors. A combination of bioinformatics analysis with in vitro high-throughput screening was utilized to identify chemoreceptors with periplasmic binding motifs and predict amino acid binding capabilities for the receptor Blr2932. In Chapter III we investigate the sensitivity adaptation system of S. meliloti chemoreceptors via methylation. CheR was demonstrated to crosslink to the cytosolic domain of McpX for in vitro experimental basis. We were able to demonstrate that single mutations of putative methylation sites could reduce the chemotactic ability, indicating that S. meliloti chemoreceptors likely utilizes the same residues as E. coli to adapt their signal sensitivity. Chapter IV encompasses the assembly of the A. tumefaciens strain 5A genome and characterization of its novel flagellar structure. The genome was sequenced and assembled into four complete contigs, finalizing the previously sequenced draft genome. This allowed for the characterization of individual flagellin function in motility and flagellar structure. The agricultural relevance of all three bacteria creates a growing need to characterize aspects of their functions related to plant interactions. The more knowledge we gain, the more opportunities we have to exploit them to improve formation of symbiotic relationships or to target them for reduction of disease spread.
From Margin Calls to Micro-Expressions: Leverage, Liquidity, and Physiological Signatures of Experimental Bubbles
Spoon, Ross Linwood (Virginia Tech, 2026-08-26)
This dissertation develops new experimental infrastructure to study instability in speculative asset markets, focusing on short squeezes and bubble-crash cycles alongside the behavioral and physiological states that accompany them.
The first chapter examines short squeezes: rapid and extreme price increases caused when speculative traders who borrowed an asset are forced into the market in order to acquire it again. This phenomenon is largely absent from experimental asset-pricing research despite its real-world prominence. By permitting leveraged short positions and enforcing margin requirements, the experimental design generates conditions under which squeezes can poten- tially emerge. We develop a tiered classification of squeeze events that includes voluntarily buying the asset at a loss and being forced to purchase the asset to maintain margin re- quirements; in the most extreme cases, cascading buy-ins drove prices to nearly 50 times fundamental value. We conclude that short-selling restrictions and excess liquidity amplify bubble formation, which creates an environment in which these explosive squeezes are more likely to happen.
The second chapter turns to the psychological and physiological underpinnings of bubble formation and collapse in constant fundamental value markets. Using Support Vector Re- gression alongside linear models, we find that conventional financial signals, such as lagged returns and order book imbalances, outperform biometric measures in predicting contempo- raneous market returns. At the individual level, however, electrodermal activity meaningfully predicts de-risking behavior and revealed risk aversion, with high earners exhibiting antici- patory arousal before crashes and low earners showing only reactionary spikes.
The third chapter introduces oPAL (open-source Periodic Auction Laboratory), a turnkey, extensible asset market platform built on the oTree architecture. oPAL fills a critical gap left by existing tools' near-exclusive focus on continuous double auctions, offering discrete call markets that isolate strategic bidding from execution-speed noise. Additionally, oPAL addresses practical frictions of online synchronous recruitment through an integrated pre- screening module and administrative portal. Key features enabling the two experiments presented in this dissertation include short selling, margin-based leverage, automated liquidations, flexible fundamental value trajectories, and a motion-artifact-minimizing entry interface for biometric recording.
Together, these chapters contribute new open-source infrastructure for experimental finance and new evidence on how leverage, liquidity, and emotional state jointly shape the formation and resolution of price instability in speculative markets.
Application-Aware I/O Prediction and Optimization for HPC and AI Workloads
Yazdani, Ahmad Hossein (Virginia Tech, 2026-08-26)
High-performance computing (HPC) systems are essential for executing large-scale scien- tific and artificial intelligence workloads that demand extensive parallelism across compute, storage, and memory resources. Modern systems, such as Aurora at Argonne National Lab- oratory and the exascale Frontier system at Oak Ridge National Laboratory, have significantly expanded computational and I/O capabilities. However, storage and memory subsystems often struggle to keep pace with the scale, diversity, and dynamic behavior of modern applications. This gap creates performance bottlenecks that are difficult to diag- nose and mitigate using rigid, application-agnostic I/O management strategies. Addressing these challenges requires a deeper understanding of how application semantics, execution configurations, and workload behavior interact with storage and memory systems. This thesis addresses these challenges across three complementary directions. First, we model application I/O behavior using historical execution patterns and user-specified parameters.
Through analysis of E3SM and LAMMPS workloads, we show that prior application con- figurations are strongly correlated with future I/O behavior, enabling accurate prediction over 1–10-day windows with uncertainty below 0.25 and 80–90% prediction accuracy. Sec- ond, we introduce IO-Sense, an agentic I/O reasoning pipeline that combines Darshan logs with application-level configuration semantics to summarize large-scale traces and explain how configuration knobs influence I/O patterns. On AMReX-based HPC applications, IO- Sense achieves over 90% log similarity with the ground-truth log derived from the knob change for most evaluated models and demonstrates that finer-grained I/O counters can improve reasoning accuracy, though at the cost of higher inference overhead. Third, we study I/O and memory behavior in deep learning training workloads through SHADE and SODA-LLM. SHADE introduces a rank-based importance sampling policy that improves data locality and cache efficiency for large-scale computer vision training workloads. Build- ing on this direction, SODA-LLM targets transformer-based language model training/fine- tuning by exploiting the spatial and temporal stability of activation importance to guide token-level activation retention, offloading, and prefetching during the backward pass. The proposed system reduces end-to-end training latency by up to 20% while lowering memory usage by 5%. Together, these projects show that application-aware reasoning can improve prediction, diagnosis, and optimization across HPC and AI workloads. By connecting appli- cation configurations and execution semantics to observed I/O and memory behavior, this thesis provides a foundation for more adaptive storage and memory management strategies in next-generation computing systems.
Understanding and Enhancing Sequential Decision-Making and Alignment in Foundation Models
Sel, Bilgehan (Virginia Tech, 2026-08-26)
Foundation models are used as sequential decision-makers, in tasks whose outputs unfold as dependent steps: reasoning, planning, decisions affecting several parties, and generation under safety requirements. This dissertation asks which factors govern the performance of foundation models in such settings and how that performance can be improved. The answer locates those factors in training data and training procedure. Pretraining text preserves finished solutions far more often than the failed attempts behind them, so a model trained on it defaults to confident, linearly coherent continuation; exploring alternative paths and retracting mistaken steps are underrepresented behaviors, not absent capabilities. The contributions elicit these behaviors in context and then teach them by supervised and reinforcement learning. One observation recurs: a model given room to explore diverse solution paths and to back out of its own mistakes makes better decisions, and in safety-critical settings the same capacity lets it recover from unsafe trajectories.
The body has three parts. Part I treats foundation models in reinforcement learning: a meta-learning method for sequences of derivative-free optimization tasks, with task-averaged regret guarantees; policy optimization under several reward objectives and hard safety constraints, with a rectification step that restores feasibility after a detected violation; and an analysis tracing in-context reinforcement learning to the diversity of the pretraining task distribution. Part II treats large language models in sequential decision-making: a tool-use framework that translates natural-language energy-management requests into solver-ready optimization programs; the Algorithm of Thoughts, a prompting strategy whose exemplars record a search process so that the model explores, prunes, and backtracks within a single generation; an extension to autonomous long-horizon planning; and a training pipeline that makes the behavior a concise default. Part III turns the same capacity to alignment and safety: a prompting framework that surveys a decision's consequences for every affected stakeholder before answering, and a reinforcement-learning method that trains the backtracking step as a safety signal, so that the model retracts an emerging violation and continues from the safe prefix. Recovery complements avoidance rather than replacing it; an integrated red-team study of a static classifier defense motivates judging safety on the generated trajectory.
Epidemic Modeling with Generative Agents: Methodology, Prompt Sensitivity, and LLM Sensitivity
Williams, Ross Fantus (Virginia Tech, 2026-08-26)
Generative agent-based modeling (GABM) replaces the rule-based decision-making of classical agent-based models with agent decisions powered by large language models (LLMs), enabling researchers to simulate human behavior that resists clean mathematical specification. This dissertation establishes the methodology, characterizes its sensitivity to prompt design, and characterizes its sensitivity to the underlying LLM. Three studies — conducted within a shared epidemic GABM framework in which generative agents decide each day whether to stay home or go out — support the central claim: GABM is a viable simulation method, but the prompt and the LLM are both parameters of any GABM finding.
Agents endogenously self-isolate as community case counts rise and quarantine when they feel ill. Collectively, this produces multi-wave dynamics followed by an endemic period. Heterogeneous behavior emerges from Big Five personality traits, age, and gender.
The prompt is a parameter of GABM, but only along certain dimensions. Synonymous rewording and persona names leave epidemic trajectories statistically unchanged; minor wording variations and contextual shifts both alter outcomes.
LLM choice is also a parameter. Across 21 LLM configurations from OpenAI, Anthropic, and Google, most LLMs agree on the direction of persona effects but disagree several-fold on magnitude. Standard model attributes — size, version, knowledge cutoff, release date — do not predict where an LLM's collective agent behavior lands. Mentioning a trait in the response text moves that trait's effect on the decision beyond the trait alone.
Together the three studies argue that GABM is most defensible for qualitative claims about agent behavior and weakest for the quantitative claims modelers most often want to make. LLM sensitivity testing should become standard practice in GABM publications, alongside prompt sensitivity. The methodology is real; its parameters include the prompt and the LLM.