Digital Commons Network^™

Letter To Laure, Lloyd C. Clark

Lloyd Clark Papers

Letter to Laure Anderson, Development Director at The University of Texas at Arlington, from Lloyd C. Clark, with an attached publication, covering Allyn Keiser's scholarship.

Letter To Nez, Lloyd C. Clark

Lloyd Clark Papers

Letter to Nez from Lloyd C. Clark, welcoming him home and discussing his recent life updates.

Letters To Anne And Bob, Lloyd C. Clark

Lloyd Clark Papers

Letters to Anne and Bob Witt from Jean and Lloyd Clark, one thanking them for the bouquet they sent, and the other expressing their desire to see them.

Uta Alumni Check, Lloyd C. Clark

Lloyd Clark Papers

Check to the UTA Alumni Association for $150, from Lloyd C. Clark, for Dawn Prejean's gift.

Letter To Lloyd And Jean, Laure Anderson

Lloyd Clark Papers

Letter to Lloyd and Jean Clark from Laure Anderson, expressing thanks.

Letter To Bob, Lloyd C. Clark

Lloyd Clark Papers

Letter to Robert E. Witt, President of The University of Texas at Arlington, from Lloyd C. Clark, with a photo of the bouquet gifted to him.

Letter To Laure, Lloyd C. Clark

Lloyd Clark Papers

Letter to Laure Anderson, Development Director at The University of Texas at Arlington, from Lloyd C. Clark, regarding stocks.

Stock Certificate

Lloyd Clark Papers

Central Newspapers stock certificate and related stock power documents.

Supporting Student Use Of Artificial Intelligence Technologies, Deborah A. Mcallister, Jennifer R. Lynberg, Jason R. Gordon, Mary L. Buehner, Anna Grace L. Sorrells, Taylor L. Behrman

UTC Spring Research and Arts Conference

This project has a focus of preparing presentations, regarding various artificial intelligence (AI) topics, for use by middle and high school students participating in a science enrichment program at a local school. The project is funded through a supplemental grant to the National Science Foundation Robert Noyce Teacher Scholarship Program Noyce Track 2 grant. In a separate project, the middle and high school students submitted a project to the Presidential AI Challenge (AI.gov, n.d.). The presentations created through this funding supplement will support student learning as the students continue to improve and expand skills with AI technologies. Topics for the …

Continuing To Assess The Programming Self-Efficacy Of Teachers Through Professional Development With Drones, Deborah A. Mcallister, Jennifer R. Lynberg, Anna Grace L. Sorrells, Taylor L. Behrman

UTC Spring Research and Arts Conference

This project focused on work with the Robolink (n.d.) CoDrone EDU, which is programmable through block-coding or Python, or can be flown with a controller. Each teacher self-evaluated knowledge and skills, before and after a 2-day workshop. This project served as a replication of a previous study (McAllister et al., 2023) with a different educational drone. CoDrone EDU was selected to familiarize teacher participants with a drone used for student competitions. Menkhoff et al. (2022) discussed a college course that taught students how to use drones so that they could be used effectively in an educational setting. Walach and Harrell …

Assessing The Programming Self-Efficacy Of Teachers Through Professional Development Using The Ozobot Evo Robot, Deborah A. Mcallister, Jennifer R. Lynberg, Anna Grace L. Sorrells, Taylor L. Behrman

UTC Spring Research and Arts Conference

This project focused on work with the Ozobot (Ozo EDU, Inc., 2025) Evo robot, which is programmable through block-coding and color-coding. Ozobots have been consistently proven to enhance student learning. Improvements in student motivation, engagement, collaboration skills, and computational thinking have all resulted from the implementation of Ozobots (Eure, 2023). A study by Powers et al. (2025) supports a co-design approach through the use of interdisciplinary teams to advance STEAM integration into the classroom. However, the students’ knowledge on robot concepts (coding, programming, and building) through the use of Ozobots was mixed, with a wider range of reported perceived knowledge. …

Remote Dispatch And Initiation Of Client-Side Agentic Tasks Via Cross-Device Control Protocol, Yijo Liu, Shadia Walsh, Lily Duraiswami, Yeun Su Shin, Tiffany Deonne Morris, Carl Magnus Borg, Golden Gopal Krishna, Clifford Tse Yan Chan

Defensive Publications Series

Architectures for agentic operations often involve a coupling of the input interface and the execution environment on a single piece of hardware. Such configurations can result in resource strain on mobile hardware, affecting battery longevity and data usage, and may encounter compatibility issues with desktop-centric web workflows. A dispatch protocol is disclosed that decouples the input interface from the execution environment. A dispatch payload is generated on a source device (e.g., a smartphone) or relayed via a third-party messaging channel (e.g., enterprise communication platforms or instant messaging applications). This payload includes a task prompt, browser configurations, and user context. The …

Proactive Presentation Of Automated Assistant Prompts Based On Resource Locator Analysis, Yijo Liu, Carl Magnus Borg, Tiffany Deonne Morris, Yeun Su Shin, Shadia Walsh, Clifford Tse Yan Chan, Lily Duraiswami, Golden Gopal Krishna

Defensive Publications Series

A discovery gap exists where users often fail to recognize that an automated assistant is capable of assisting with tasks during a browsing session. This frequently leads to manual navigation through complex web workflows. To address this, a method is disclosed for the proactive presentation of assistant prompts based on resource locator analysis. Upon navigating to a resource locator or receiving a resource locator via a notification, the target page is analyzed to categorize functional capabilities. If a workflow compatible with an automated assistant is identified, such as a multi-step booking or purchasing process, a context-specific prompt is presented. This …

Surface Classification And Predictive Path Analysis For Hazard Avoidance In Wearable Display Devices, Michael Christopher Yu

Defensive Publications Series

Wearable electronic devices face challenges in identifying surface hazards, such as ice, potholes, or cliffs, while providing relevant safety information is prioritized for the wearer. Traditional methods often struggle with detecting translucent or low-contrast hazards and can result in information overload. This disclosure describes a method for hazard identification and avoidance that utilizes signal artifacts from time-of-flight sensors combined with predictive path analysis. Surface composition is classified by analyzing the noisiness and reflection patterns of sensor signals, which allows for the differentiation of materials like snow, ice, or water from diffuse surfaces like gravel. Concurrently, a projected path is generated …

Automated Booking Using Ai-Driven Browser Automation System And Calendar Availability, Benjamin Albert Azose, Jonathan Jerome Azose

Defensive Publications Series

An AI-driven browser automation system is disclosed to autonomously manage reservation and booking tasks by synchronizing real-time web availability with a user’s calendar data. The system navigates booking interfaces, via automated browser interactions or application programming interfaces (APIs), to extract available slots and performs duration-aware compatibility analysis against the user's existing commitments. Supporting both "availability-first" and "calendar-first" processing models, the system identifies non-conflicting intervals to execute bookings in fully automated, semi-automated, or recommendation-based modes. By automating form completion and monitoring for new openings, the framework reduces manual interaction overhead, eliminates scheduling errors, and improves the success rate of securing high-demand …

Adding A Task To A List After Unsuccessful Completion By Ai Agent, Benjamin Albert Azose, Jonathan Jerome Azose

Defensive Publications Series

A technique is described for capturing the intent of unsuccessful automated browser tasks by integrating them into a task management interface. When an agentic flow is initiated within a web browser, the state of the task is monitored. Upon detection of a failure or a terminal error state, a task management interface associated with the user is identified. A human-readable summary of the unsuccessful task is generated and added to the identified interface as a new entry. Metadata, such as execution timestamps and source domains, may be included. This ensures that the context of incomplete automated operations is preserved for …

Suggesting An Activity While Waiting For An Ai Agent, Benjamin Albert Azose, Jonathan Jerome Azose

Defensive Publications Series

A computing system proactively suggests activities to a user while an AI agent performs a task. An activity list is generated based on user-specific behavioral data and historical metrics. An estimate is made of the expected duration of the task, and the activity list is filtered to identify a subset of options that match the estimated duration. A contextual filter can be applied based on the subject matter of the task, prioritizing topically related activities or selecting disparate activities to provide a mental break. The selected activity is presented to the user, filling the latency period and enhancing user engagement.

Automated Generation And Dynamic Insertion Of Image Alternative Text Via Generative Models, Benjamin Albert Azose

Defensive Publications Series

Missing or low-quality alternative text, or alt text, frequently impedes web and document accessibility. Alt text includes descriptive information attached to an image that screen readers use to describe visual content to visually impaired users. A deficiency in alt text creates a barrier for users who depend on screen readers to interpret visual content on webpages. The disclosure describes a method whereby a web browser or document viewer utilizes a generative model to analyze the content of an image to be displayed. If the image is not associated with alternative text that describes the image adequately, the web browser or …

Linguistic Transformation And Multimodal Content Simplification Logic, Robert Gerhard Rudolf Brauer, Isabelle Evelyn Joswig

Defensive Publications Series

A system for linguistic transformation performs modification of an unstructured data payload. The system comprises a content ingestion module, a linguistic transformation module, a multimodal interpreter, and a content output module. The content ingestion module identifies textual elements and non-textual elements within the unstructured data payload. The linguistic transformation module applies standardized parameters to the textual elements, modifying syntax and vocabulary. The multimodal interpreter generates descriptive text strings from the non-textual elements. The linguistic transformation module applies the standardized parameters to the descriptive text strings. The content output module aggregates the modified textual elements and the modified descriptive text strings …

Holographic Data Sharding For Cross-Jurisdictional Storage And Ephemeral Edge Reconstruction, Mohsin Alam

Defensive Publications Series

A waveform security model may use a holographic sharding process to transform a data object into a set of mathematically incomplete and individually unintelligible interference patterns. An orchestration engine can distribute these shards across multiple storage providers and/or jurisdictions, configured to prevent a single entity from possessing sufficient information for reconstruction. The complete data object can be reconstituted, for example, ephemerally within a secure processing environment on an authorized client device (e.g., a smartphone, a laptop computer) at a time of use. This process can render data non-local, disaggregated, and mathematically incomplete at rest, potentially improving data sovereignty and secrecy …

Autonomous Reconciliation Of Cloud Configuration Drift Into Infrastructure As Code, Paras Mamgain

Defensive Publications Series

Cloud environments can experience configuration drift, where manual interventions may cause the live state of resources to diverge from their infrastructure as code (IaC) definitions, creating a risk that automated deployments could overwrite certain fixes. To address this, an autonomous reconciliation agent can programmatically discover live cloud resources, analyze their configurations and dependencies, and synthesize relational IaC that represents the current environment. This generated code may then be integrated into a version control system, for example, by creating a pull request for human review. The process can provide an automated pathway to codify out-of-band changes, helping to reduce drift and …

Hardware-Decoupled Composition Of Live-Data Overlays On A Dedicated Graphics Plane For A Television Os, Richa Gupta

Defensive Publications Series

A system is described for displaying persistent, live-data overlays on a computing device, such as a smart television or media streaming device, which can reduce interruption to a primary, full-screen viewing experience. The system may use hardware-decoupled composition, where a method can include rendering a primary video stream onto a video plane and a data widget onto a separate graphics plane. These hardware planes can then be combined by a hardware composer. This architecture can utilize an asynchronous, low-frequency refresh mechanism for the widget and a javascript object notation (to vector) data pipeline. This approach may allow the overlay to …

Unified Probabilistic Model For Robot Calibration Fusing Kinematic, Sensor, And Manufacturing Priors, Taemin Kim

Defensive Publications Series

Robot calibration methods that rely on external sensor measurements may face challenges with accuracy and robustness, for example, in the presence of sensor noise, occlusions, or geometric singularities. A disclosed technology can utilize a unified probabilistic model, which may be implemented as a factor graph, to fuse heterogeneous data sources. This model can integrate data sources, such as high-frequency robot control inputs like encoder trajectories and manufacturing specifications like computer-aided design tolerances, as probabilistic priors with sensor measurements. A reliability gating mechanism may help manage geometric degeneracies, for instance, by dynamically weighting data sources based on observability. This approach may …

Asynchronous Remote Interaction And Conversational Interrupt Handling For Long-Running Agentic Tasks, Yijo Liu, Yeun Su Shin, Clifford Tse Yan Chan, Shadia Walsh, Lily Duraiswami, Carl Magnus Borg, Tiffany Deonne Morris, Golden Gopal Krishna

Defensive Publications Series

Tasks are managed through a decoupled interaction and execution framework. An execution environment on a primary device performs a task and detects an interrupt state when a decision node is reached that lacks parameters or confirmation. A natural language query object is serialized and routed via a cloud messaging service or chat to an authenticated secondary device. Remote input is captured at the secondary device and transmitted back to the primary device. The input is parsed and injected into the execution runtime, allowing the task to resume from the point of interruption. This process enables the extension of decision-making logic …

Determining When To Present Notifications, Benjamin Albert Azose, Jonathan Jerome Azose

Defensive Publications Series

An artificial intelligence (AI) agent can determine when to present notifications to a user. The AI agent can act as a virtual concierge and notify users when requested tasks, such as booking flights or summarizing research, have been completed. The AI agent can also notify the user of new features of applications. The AI agent can determine when to notify the user without interrupting the user from performing an important task. The AI agent can consider activity by the user or lack thereof, the type of activity of the user, a time of day, network connectivity, and/or how long since …

Determining Whether To Enter Incognito Mode Based On Ai Agent Request, Benjamin Albert Azose, Jonathan Jerome Azose

Defensive Publications Series

Incognito states or modes can be managed automatically during the execution of agentic browser flows to address data management concerns during sensitive tasks. When a user requests an AI agent to perform a task, a sensitivity level of the task can be determined based on stored confidentiality preferences. These confidentiality preferences can be derived, with user permission, from on-device analysis of historical behavior by the user or from global data that includes, with user permission, numerical features describing the preferences of other users. If the task, the execution plan of the task, or specific Universal Resource Locators (URLs) associated with …

The Living Memory System: A Framework For Transforming Death Care Infrastructure Into A Perpetual Relational Ecosystem For Memory Preservation, Family Continuity, And Citizen-Governed Memorial Stewardship, Joseph Jm Walker

Defensive Publications Series

The death care industry operates under structural conditions that systematically disadvantage grieving families through information asymmetry, compressed decision timelines, relationship termination at the point of transaction, fragmented archival systems, and the absence of citizen-governed transparency standards. Simultaneously, the digital era has created a second layer of fragmentation in which human lives generate vast quantities of meaningful material that remains structurally disconnected from permanent physical memorial locations and inaccessible across generations.

This disclosure presents the Living Memory System — a modular, globally applicable framework repositioning memorial infrastructure from a primarily transactional service category into a long-term continuity ecosystem serving families, communities, …

Method To Adapt The Power Limitation In A Quasi-Resonant Induction Generator, Massimo Zangoli, Alex Viroli, Giacomo Corsi, Fabio Angeli, Filippo Milanesi

Defensive Publications Series

This method aims to adapt the power limitation to different cookware. While operating for a prolonged period, induction generators get heated and may need to limit power for reliability reason. Temperature and thus power limitation is directly linked to the type of the cooking vessel so this disclosure presents an adaptive technique able to change the power limitation based on the pot characteristics, offering better cooking experience.

Method To Limit The Diode Current In A Quasi-Resonant Induction Generator, Lucia Biscontin, Alex Viroli, Massimo Zangoli, Giacomo Corsi, Keke Wang, Emanuel Urgese, Fabio Angeli, Desislava Iolova

Defensive Publications Series

This method aims to protect a crucial hardware component (reverse diode) of a quasi-resonant induction generator by estimating one specific electrical parameter (its current). This is achieved by estimating such parameter, estimating its limit and change the working point of the generator in case the limit is surpassed. This lowers breakage likelihood and lengthens lifespan of the component (and the appliance).

Spectragrid Vision Transformer, Johny Manuel-Devadoss, Deepthi Prasad

Defensive Publications Series

Vision Transformers have become the dominant paradigm for visual learning due to their ability to model long-range dependencies through self-attention. However, dense pairwise attention introduces quadratic computational complexity with respect to token count, severely limiting scalability in high-resolution and long-duration perception tasks. Existing efficiency improvements reduce attention overhead but preserve the underlying assumption that global visual understanding requires explicit token interaction.

This paper introduces the SpectraGrid Vision Transformer (SG-ViT), a unified architecture that replaces dense self-attention with structured perceptual computation, sparse routing, persistent latent world modeling, and autonomous action-driven reasoning. Instead of recomputing dense token interactions at every timestep, SG-ViT …