Open Access. Powered by Scholars. Published by Universities.®

Digital Commons Network™

Open Access. Powered by Scholars. Published by Universities.®

Physical Sciences and Mathematics

Institution
Keyword
Publication Year
Publication
Publication Type
File Type

Articles 721 - 750 of 58435

Full-Text Articles in Entire DC Network

Smart Kitchen: Towards Real-Time Ai Systems For Cognitive Support In Daily Activities, Ruiqi Wang Apr 2026

Smart Kitchen: Towards Real-Time Ai Systems For Cognitive Support In Daily Activities, Ruiqi Wang

McKelvey School of Engineering Graduate Student Theses & Dissertations

The rapid growth of the aging population and the rising prevalence of Subjective Cognitive Decline (SCD) highlight the need for continuous, unobtrusive assessment of functional cognition during everyday activities. Vision-based smart home systems offer a promising pathway for monitoring behavior and supporting independent living. However, enabling real-time cognitive assistance remains challenging. It requires not only accurately interpreting complex human behaviors to detect cognitive errors, but also supporting real-time deployment on resource-constrained edge devices and under dynamic wireless network conditions. This dissertation presents Smart Kitchen, an AI-driven system for real-time cognitive error detection through the monitoring of daily cooking activities. Under …

Go to article

Revisiting Visualization Literacy: What Standardized Assessments Reveal And Conceal Across Instruments, Cultures, And Ai Systems, Saugat Pandey Apr 2026

Revisiting Visualization Literacy: What Standardized Assessments Reveal And Conceal Across Instruments, Cultures, And Ai Systems, Saugat Pandey

McKelvey School of Engineering Graduate Student Theses & Dissertations

Data visualizations are now a primary medium for public communication, scientific reasoning, and decision-making. By transforming complex data into accessible graphical forms, visualizations are widely assumed to make information comprehensible to broad audiences. Yet the ability to accurately read, interpret, and critically evaluate a visualization, what researchers call visualization literacy, is neither uniform nor universal. It is a learned, multidimensional skill shaped by prior exposure, educational opportunity, and context. While the field has developed standardized instruments to measure visualization literacy, these tools were built under narrow assumptions: participants are typically paid, English-speaking, and recruited from Western online panels. When any …

Go to article

Benefits Of Traffic Reprofiling For Delay Sensitive Networking, Jiaming Qiu Apr 2026

Benefits Of Traffic Reprofiling For Delay Sensitive Networking, Jiaming Qiu

McKelvey School of Engineering Graduate Student Theses & Dissertations

Deterministic networking systems, such as Time-Sensitive Networking (TSN) and Deterministic Networking (DetNet), require strict end-to-end delay guarantees while efficiently utilizing limited network resources. Conventional approaches typically focus on fixed traffic profiles, which can lead to suboptimal resource utilization in scheduling or admission control problems. This dissertation investigates traffic reprofiling—the proactive reshaping of traffic arrival patterns—as a complementary mechanism for improving both resource efficiency and delay performance under strict service guarantees. The dissertation consists of three parts. The first part studies bandwidth minimization under hard delay constraints for Service Curve Earliest Deadline First (SCED) schedulers. We show that traffic reprofiling can …

Go to article

Cognition Is Not Content: A Structural Account Of Processing Conditions, Griselda Poe Apr 2026

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 …

Go to article

Finiteness Of Torsion Loci And Normal Functions Arising From Cycles On Hadamard Products, Devin Akman Apr 2026

Finiteness Of Torsion Loci And Normal Functions Arising From Cycles On Hadamard Products, Devin Akman

Arts & Sciences Graduate Student Theses and Dissertations

The unifying theme of this dissertation is normal functions. In the first chapter, we study an invariant of knot exteriors and similar manifolds called the A-polynomial through the lens of vanishing loci of normal functions. Using a special case of the Zilber-Pink conjecture, proven in the second chapter, we show that only finitely many irreducible Laurent polynomials of bounded overgenus appear as factors of A-polynomials. In the third chapter, we construct normal functions from hypergeometric variations of Hodge structure and compute their regulators. Finally, we determine under which conditions incomplete motivic cohomology cycles on families complete to cycles on their …

Go to article

Probing Composition Effects On The Intrinsic Energetics Of Proton Transfer In Binary (Hcl)M(H2o)N Clusters, Max R. Tucker, Sarah N. Arradondo, Olivia M. Longsworth, George C. Shields, Gregory S. Tschumper Apr 2026

Probing Composition Effects On The Intrinsic Energetics Of Proton Transfer In Binary (Hcl)M(H2o)N Clusters, Max R. Tucker, Sarah N. Arradondo, Olivia M. Longsworth, George C. Shields, Gregory S. Tschumper

Chemistry Faculty Research & Creative Works

This work examines the effects on intrinsic energetics as the composition of finite (HCl)m(H2O)n cluster families containing at least one water molecule is changed at the molecular level, where m ≥ 2 and m + n = 4 − 6, and the associated (Cl−)x(H3O+)x(HCl)m−x(H2O)n−x clusters resulting from the dissociation of x HCl fragments via proton transfer (PT). They are collectively labeled m:n, and xPT indicates the degree of dissociation, from 0PT (no dissociation) up to 3PT. More than 1000 unique minima …

Go to article

Complexity Of The Zero Set Of A Matrix Schubert Ideal, Cesar Julian Meza Apr 2026

Complexity Of The Zero Set Of A Matrix Schubert Ideal, Cesar Julian Meza

Arts & Sciences Graduate Student Theses and Dissertations

T-varieties are normal varieties equipped with an action of an algebraic torus T. When the action is effective, the complexity of a T-variety X is dim(X)−dim(T). Matrix Schubert varieties, introduced by Fulton in 1992, are T-varieties consisting of n×n matrices satisfying certain constraints on the ranks of their submatrices. In this dissertation, we focus on the complexity of certain torus-fixed affine subvarieties of matrix Schubert varieties. Concretely, given a matrix Schubert variety X_w where w∈S_n, we study the complexity of Y_w obtained by the decomposition X_w = Y_w ×C^k with k as large as possible. Building on results by Escobar–Mészáros …

Go to article

Exactly Solving One-Dimensional Models For The Fractional Quantum Hall Effect, Joseph R. Cruise Apr 2026

Exactly Solving One-Dimensional Models For The Fractional Quantum Hall Effect, Joseph R. Cruise

Arts & Sciences Graduate Student Theses and Dissertations

Nearly fifty years after its experimental debut, the quantum hall effect has produced perhaps more unique descriptions than any other subject of condensed matter physics. From topological band theory to non-commutative geometry, topological and conformal field theories, random matrix models and more, each approach captures a subset of the essential features, and compliments the others to produce a beautiful mosaic description. In this thesis, we revisit the fractional quantum hall (FQH) effect from its oldest vantage point, that of the trial wavefunctions and the parent Hamiltonians which stabilize them. Laughlin's observation of the importance of the theory of holomorphic functions …

Go to article

Isolation, Identification And Antibiotic-Resistance Profiling Of Bacteria Isolated From Mobile Phone Surfaces In Karbala., Kawkab Abdullah Alsaadi, Zahraa Raheem Murshidy, Dhuha Ali Hussein, Sura Abd Ali Kadhim, Kawakib Aboudi Hanoon Apr 2026

Isolation, Identification And Antibiotic-Resistance Profiling Of Bacteria Isolated From Mobile Phone Surfaces In Karbala., Kawkab Abdullah Alsaadi, Zahraa Raheem Murshidy, Dhuha Ali Hussein, Sura Abd Ali Kadhim, Kawakib Aboudi Hanoon

Karbala International Journal of Modern Science

Due to the diverse environments that mobile phones are exposed to via human handling they become reservoirs of microorganisms. The objective of the current study was to determine the level and character of bacterial contamination on mobile phones by identifying bacterial strains, and the degree of antimicrobial susceptibility. A cross-sectional study was conducted from April to July 2025 where phones belonging to 115 individuals were randomly chosen from various departments and units at Kerbala University and swabbed. The VITEK 2 automated system was used to identify bacterial taxa and test for antimicrobial susceptibility, and 50 of the swabs exhibited bacterial …

Go to article

Fast Hyperspectral And Super-Resolved Mapping Of Lipid Membrane Polarity With Single-Molecule Sensitivity, Elric Dion Pott, Meek Yang, James Ethan Batey, Jole Embree, Bin Dong Apr 2026

Fast Hyperspectral And Super-Resolved Mapping Of Lipid Membrane Polarity With Single-Molecule Sensitivity, Elric Dion Pott, Meek Yang, James Ethan Batey, Jole Embree, Bin Dong

Chemistry & Biochemistry Faculty Publications and Presentations

Cell membranes display nanoscale heterogeneity in lipid composition and organization that regulates vital biological processes yet remain challenging to resolve with conventional imaging. We introduce spectral phasor single-molecule localization microscopy (SP-SMLM), a hyperspectral and super-resolution method that combines wavefront-like optical filtering with single-molecule imaging for simultaneous spatial and spectral analysis. A lab-built three-channel imager with sine/cosine filters encodes emission spectra of single molecules into the phasor space, enabling high-throughput, high-SNR mapping of membrane polarity at sub-50 nm spatial and 15 s temporal resolutions. Through simulation, we validate that the phasor angle correlates with the spectral mean for single dye molecules. …

Go to article

Static And Dynamic Properties Of Dense Matter, Liam Brodie Apr 2026

Static And Dynamic Properties Of Dense Matter, Liam Brodie

Arts & Sciences Graduate Student Theses and Dissertations

Neutron stars and their mergers create environments of ultra-dense matter where the strong interaction governs the behavior, but where quantum chromodynamics, the theory of the strong interaction, cannot be solved exactly or perturbatively using current methods. Neutron stars are objects with more mass than our Sun, but with a diameter of about 25 km. Observations of neutron star macroscopic properties like their masses and radii, their surface temperatures, and the gravitational-wave and electromagnetic signals from their mergers allow us to use theoretical tools to infer the behavior of matter at a microscopic scale. In this thesis, we explore the static …

Go to article

Development Of New Technical Approaches To Speed The Search For Axion Dark Matter With Haloscopes, Jonah Harper Hoffman Apr 2026

Development Of New Technical Approaches To Speed The Search For Axion Dark Matter With Haloscopes, Jonah Harper Hoffman

Arts & Sciences Graduate Student Theses and Dissertations

The axion is a hypothetical particle that may solve two major problems in physics: the strong CP problem in quantum chromodynamics (QCD) and the unknown particle nature of the dark matter. Axion haloscopes like the Axion Dark Matter eXperiment (ADMX) are capable of probing the region of parameter space where axions could constitute the dark matter. These experiments use a microwave resonant cavity in a strong magnetic field to convert axions into detectable photons when their Compton wavelength matches the resonant mode of the cavity. By moving a metal rod within the cavity volume, the resonant frequency can be tuned, …

Go to article

The Role Of Interactions In Moiré Quantum Materials, Liangtao Peng Apr 2026

The Role Of Interactions In Moiré Quantum Materials, Liangtao Peng

Arts & Sciences Graduate Student Theses and Dissertations

Moiré quantum materials have emerged as a highly tunable platform for studying strongly interacting electrons in low dimensions. By twisting or lattice-mismatching atomically thin crystals, one can generate long-wavelength moiré superlattices that reconstruct the electronic bands, produce narrow bandwidths, and greatly enhance the importance of Coulomb interactions. As a result, these systems host a wide range of correlated phenomena, including symmetry breaking, magnetism, and superconductivity. This dissertation develops and applies Green's-function-based many-body methods to understand the role of interactions in moiré quantum materials. This dissertation consists of four chapters. The first chapter introduces moiré quantum materials and reviews the basic …

Go to article

3Ω Thermal Transport In Atomically Thin Materials, Yiwei Le Apr 2026

3Ω Thermal Transport In Atomically Thin Materials, Yiwei Le

Arts & Sciences Graduate Student Theses and Dissertations

Since the discovery of graphene, research on two-dimensional (2D) materials has become a central focus of condensed matter physics. While an enormous number and variety of investigations of their electrical, optical, and mechanical properties have been undertaken, their thermal properties remain comparatively underexplored, despite the importance for understanding low-energy excitations and phase transitions. Here, we introduce an extension of the 3ω method thermal measurement technique to the 2D limit of materials. We develop analytical models of thermal transport in the 3ω geometry with boundary conditions tailored for suspended 2D flakes. We then experimentally confirm the feasibility of this approach in …

Go to article

Advancing High-Energy Astrophysics Through X-Ray Spectropolarimetry And Balloon-Borne Instrumentation, Sohee Chun Apr 2026

Advancing High-Energy Astrophysics Through X-Ray Spectropolarimetry And Balloon-Borne Instrumentation, Sohee Chun

Arts & Sciences Graduate Student Theses and Dissertations

The strong gravitational field around black holes provides a unique laboratory for investigating the behavior of matter and radiation under extreme conditions. As matter accretes, the interplay between the accretion disk, hot corona, and surrounding stellar environment produces complex spectral and polarimetric X-ray signatures. Thus, X-ray spectropolarimetry has become a powerful tool for constraining the accretion geometry. Building on these advances, this dissertation integrates theoretical modeling, observational data analysis, and instrumentation development to address open questions in high-energy astrophysics. On the theoretical side, relativistic ray-tracing simulations test the truncated disk-hot inner flow accretion geometry against observations of Cygnus X-1, identifying …

Go to article

Optical Mechanisms Behind Vitrinite Reflectance Suppression Caused By Strong Anisotropy: A Case Study Of Contact Metamorphic Coals, An Yanfei, He Shuyang, Chen Kaixin, Gao Junhan Apr 2026

Optical Mechanisms Behind Vitrinite Reflectance Suppression Caused By Strong Anisotropy: A Case Study Of Contact Metamorphic Coals, An Yanfei, He Shuyang, Chen Kaixin, Gao Junhan

Coal Geology & Exploration

Objective For vitrinite with weak anisotropy, which proves uniaxial and optically negative, replacing random vitrinite reflectance (Rran) with the maximum vitrinite reflectance (Rmax) allows for the accurate correction of vitrinite reflectance suppression induced by anisotropy. However, as its anisotropy intensifies, vitrinite exhibits increasingly complex optical properties, rendering it challenging to reflect the thermal metamorphic temperature of vitrinite using Rmax. Methods This study investigated the strongly anisotropic vitrinite in coals and natural coke subjected to magma intrusion in the Shitai coal mine, Huaibei City, Anhui Province. Using a reflected-polarized light microscope (R-PLM), an …

Go to article

Tiny Engineers: How Polycheateworms Can Extend Sedimentary Oxic Layers And Create Better Seagrass Habitats, Natalie Spohn Apr 2026

Tiny Engineers: How Polycheateworms Can Extend Sedimentary Oxic Layers And Create Better Seagrass Habitats, Natalie Spohn

Ocean Engineering and Marine Sciences Student Publications

The Indian River Lagoon (IRL) is a shallow, well mixed estuary on the East Coast of Florida. Presently, the IRL is suffering eutrophication due to a cycle of harmful algal blooms (HABs) that deplete the water column of oxygen (2,5). Deposition of organic matter from HABs in the sediment leads to reduced sedimentary oxygen, continuing to support the cycle of eutrophication through benthic-pelagic coupling (1,5). Anaerobic sedimentary conditions can increase concentrations of ammonium and sulfide, leading to seagrass loss (3,4,6,7,10). Bioturbation by infauna can extend the oxic layer of sediments, promoting aerobic processes and helping to remove nitrogen and sulfur …

Go to article

Mcgehee Blow Up And Collision Manifold Of Planar (2+2)-Body Problem, Nathaniel Scott Sill Apr 2026

Mcgehee Blow Up And Collision Manifold Of Planar (2+2)-Body Problem, Nathaniel Scott Sill

Theses and Dissertations

This thesis analyzes the dynamics of the Planar (2+2)-Body Problem which consists of two asteroids moving under the gravitational force of each other and of two larger primaries. We use the McGehee blow up technique to remove the singularities in the dynamics associated with a triple collision with one of the primaries by introducing a new set of variables. Additional variables are introduced to reduce the dimension of the problem. We then derive the dynamics for these new variables. We then use the energy relation that comes from the original Hamiltonian to describe the collision manifold which is pasted in …

Go to article

Differentiable Objectives For 3d Scene Relighting Via Gradient Descent On Olat Basis Coefficients, Anson Savage Apr 2026

Differentiable Objectives For 3d Scene Relighting Via Gradient Descent On Olat Basis Coefficients, Anson Savage

Theses and Dissertations

Designing effective lighting is an iterative and often time-consuming process. This work contributes to automatic lighting design research by presenting a render-engine agnostic optimization routine: gradient descent on RGB multipliers of one-light-at-a-time (OLAT) basis images. We compare several objective functions to accomplish lighting tasks and show that our method is capable of quickly and effectively exploring different lighting styles using either text prompts or reference images. We also present several datasets specific to lighting tasks and show that fine-tuning on these datasets can improve performance.

Go to article

Mechanism Of Improved Miscibility Of Co2–Oil Systems By Multiester-Headgroup Surfactants, Ning Xu, Yanling Wang, Baojun Bai, Yu Zhang, Wenjing Shi, Zhaonian Zhang, Wenhui Ding, Peixu Ma, Zan Gao Apr 2026

Mechanism Of Improved Miscibility Of Co2–Oil Systems By Multiester-Headgroup Surfactants, Ning Xu, Yanling Wang, Baojun Bai, Yu Zhang, Wenjing Shi, Zhaonian Zhang, Wenhui Ding, Peixu Ma, Zan Gao

Geosciences and Geological and Petroleum Engineering Faculty Research & Creative Works

In the pursuit of carbon neutrality, the CO2-enhanced oil recovery provides dual benefits by enabling both carbon sequestration and incremental oil production. However, its application is limited by high minimum miscibility pressure. CO2-philic and oil-affinitive surfactants have emerged as an effective, low-dosage, and cost-efficient strategy to reduce the minimum miscibility pressure. In this study, macroscopic phase behavior experiments combined with molecular dynamics simulations were employed to systematically elucidate the influence of multiester-head surfactants on CO2–oil miscibility. Using a modified pressure–volume–temperature apparatus equipped with optical power monitoring, we determined that multiester-head surfactants reduced the first-contact …

Go to article

Maximizing Uinta Basin Oil Well Yield Through Geochemical Fingerprinting Of Produced Water, Troy W. Day Apr 2026

Maximizing Uinta Basin Oil Well Yield Through Geochemical Fingerprinting Of Produced Water, Troy W. Day

Theses and Dissertations

The Uinta Basin is an unconventional oil play that has, since 2017, annually increased hydrocarbon extraction through modern drilling and fracing techniques with increased frac loading. The primary production target is the Uteland Butte Member of the Green River Formation, which consists of interbedded limestones, dolostones, and thick, organic-rich mudstones. Along with increased hydrocarbon production, volumes of produced water from the Uteland Butte Member have also risen. Water is co-produced with hydrocarbons and is an undesirable byproduct of hydrocarbon extraction. Large frac operations now consistently have highly variable percentages of recovered water volume relative to oil. This increase in water …

Go to article

Massively Parallel Kalman Filtering: Scaling State Estimation Via Cuda Kernels, Joseph Campione, Peter Chim, Hannah Depuydt, William Johnston, Jackson Phillips, Brian White Apr 2026

Massively Parallel Kalman Filtering: Scaling State Estimation Via Cuda Kernels, Joseph Campione, Peter Chim, Hannah Depuydt, William Johnston, Jackson Phillips, Brian White

Mathematics & Computer Science Student Scholarship

This project addresses the challenge of state estimation in real-world conditions by fusing data from multiple sensors using the Kalman Filter. To ensure numerical stability, we use the Joseph Form covariance update, which guarantees valid results but introduces significant computational overhead due to its complexity. To overcome this limitation, we implement a parallelized solution using custom CUDA kernels on a GPU, distributing matrix operations across thousands of threads rather than relying on sequential CPU execution. Through systematic benchmarking across matrix sizes ranging from 4×4 to 4096×4096, we identify a crossover region where GPU performance surpasses CPU efficiency. This work shows …

Go to article

Developing Physics And Computing Identities In A Computationally Integrated Physics Course, Elizabeth Hallock Apr 2026

Developing Physics And Computing Identities In A Computationally Integrated Physics Course, Elizabeth Hallock

Engineering & Physics Student Scholarship

No abstract provided.

Go to article

Creating A Random Number Generator Harnessing: Lava Lamps As A Source Of Randomness, Rachel Barter, Nicolas Guerra, Sarah O'Connor Apr 2026

Creating A Random Number Generator Harnessing: Lava Lamps As A Source Of Randomness, Rachel Barter, Nicolas Guerra, Sarah O'Connor

Mathematics & Computer Science Student Scholarship

No abstract provided.

Go to article

Investigating The Substrate Specificity Of Azo Bond Reduction In The Human Gut Microbiome, Najoude Claude, Annick Kenfack, Glorimar De Los Santos, Madeline Bormes Apr 2026

Investigating The Substrate Specificity Of Azo Bond Reduction In The Human Gut Microbiome, Najoude Claude, Annick Kenfack, Glorimar De Los Santos, Madeline Bormes

Chemistry & Biochemistry Student Scholarship

The bacteria of the human gut microbiome can metabolize drugs, often causing unintended side effects. The uniqueness and variability of each individual’s microbial composition, coupled with our limited understanding of specific enzyme functions, make the metabolic outcomes of these therapeutics difficult. Azoreductase enzymes are of particular interest due to their ability to reduce azo bonds, which are often found in drugs and food dyes. Azo dyes can also be reduced nonenzymatically by hydrogen sulfide, a bacterial metabolite produced during sulfate and cysteine metabolism. The goal of this study is to examine the specificity and relative kinetics of both enzymatic and …

Go to article

Probing The Disequilibrium Chemistry Of Two Late T-Dwarfs Using Planetary Intensity Code For Atmospheric Scattering, Crystal-Lynn Jacqueline Bartier Apr 2026

Probing The Disequilibrium Chemistry Of Two Late T-Dwarfs Using Planetary Intensity Code For Atmospheric Scattering, Crystal-Lynn Jacqueline Bartier

Theses and Dissertations

We have performed a detailed analysis of the atmospheric properties of two late-T type brown dwarfs, 2MASSI J0415195-093506 and 2MASS J05591914-140448. For each brown dwarfs, we combined the spectra of each brown dwarf from the ground at 1-2.5 µm, AKARI at 2.5-5 µm, and from the Spitzer Space Telescope at 5-15 µm. For this project, we are particularly interested in probing the extent of vertical mixing. For this analysis, we used the Planetary Intensity Code for Atmospheric Scattering (PICASO), which is a code that creates a radiative transfer model for a given set of atmospheric parameters. With PICASO we find …

Go to article

The Item Response Warehouse: What It Is, How To Use It, And Targets For Potential Improvements, Savira D. Nadela, Hansol Lee, Nishka Jain, Ayaan Gupta, Xingyi Zhang, Benjamin W. Domingue Apr 2026

The Item Response Warehouse: What It Is, How To Use It, And Targets For Potential Improvements, Savira D. Nadela, Hansol Lee, Nishka Jain, Ayaan Gupta, Xingyi Zhang, Benjamin W. Domingue

Chinese/English Journal of Educational Measurement and Evaluation | 教育测量与评估双语期刊

The Item Response Warehouse (IRW) is a repository of harmonized item response datasets designed to support secondary analysis and methodological research in psychological and educational measurement. This paper serves as a practical guide for researchers interested in using the IRW. We describe the structure of IRW datasets and the quantitative and qualitative metadata available for dataset selection, and we demonstrate how researchers can navigate the IRW website to explore and compare available tables. We further show how the IRW R and Python packages can be used to filter datasets programmatically, download response-level data, and generate standardized citations for reproducible research …

Go to article

A Masked-Perturbation Model For Destabilization Security Games In Feedback Systems, Gabriel Michael Perry Apr 2026

A Masked-Perturbation Model For Destabilization Security Games In Feedback Systems, Gabriel Michael Perry

Theses and Dissertations

In this thesis, we present a mathematical formulation for increasing cyberphysical system security. This model allows a system administrator to use assumptions about exposed signals to derive defense strategies meant to mitigate the system's vulnerability to feedback perturbation attacks. We begin with a survey of game theory with the goal of building the concepts of resource and security games. We then introduce our model, which is a security game; targets in this game are exposed signals in the model map of the system, and resources are perturbed or introduced dynamical relationships between these signals. This formulation, called the masked-perturbation model, …

Go to article

Theoretical Expressivity Of Self-Attention As A Model Of Binary Sequence-To-Sequence Functions And Other Explorations, Dustin Angerhofer Apr 2026

Theoretical Expressivity Of Self-Attention As A Model Of Binary Sequence-To-Sequence Functions And Other Explorations, Dustin Angerhofer

Theses and Dissertations

The self-attention mechanism lies at the heart of modern sequential modeling with deep neural networks, especially in the context of Large Language Models (LLMs). The basic role of self-attention is to capture pairwise interactions between tokens in a sequence. This allows token embeddings to be endowed with context-dependent information before further processing. Despite its widespread use, our theoretical understanding of self-attention is still nascent. In this work, we examine self-attention as a model of binary sequence-to-sequence (seq2seq) functions, which is the simplest non-trivial case to analyze. We find, using VC dimension and the Sauer-Shelah lemma, that in the limit of …

Go to article

The Lookahead Transformer, Justin P. Olcott Apr 2026

The Lookahead Transformer, Justin P. Olcott

Theses and Dissertations

Large Language Models (LLMs) are transformer models that generate text autoregressively in two stages: prefill and decode. The prefill stage processes all input tokens in parallel and caches them into the KV cache. The decode stage uses the KV cache, processes a single token at a time, and is generally memory-bandwidth constrained. Each decode step generates a token probability distribution, samples from it, appends the token to the cache, and repeats. This left-to-right, one-at-a-time generation paradigm has proven effective for language modeling in many domains; however, it also has limitations such as token permanence, constant compute per token generation, constant …

Go to article

First Prev Next Last