Digital Commons Network^™

A Biodegradable Particulate-Polymer Composite, Natalie M. Durdle Miss

Defensive Publications Series

This disclosure describes a low-cost, room-temperature curing, biodegradable particulate-polymer composite material consisting of a crystalline carbohydrate filler (sucrose), an organic polysaccharide binder (gum tragacanth or equivalent water-soluble vegetable gums), and a liquid vehicle (water). This disclosure details the mechanical formulation, processing methods, and structural applications of this material for non-culinary industrial uses, including sacrificial tooling, casting mandrels, rapid prototyping, and biodegradable architectural modeling. The physical mechanics and chemical formulations of this material are inherently enabled by historical cultural documentation, specifically the 1815 publications of Marie-Antoine Carême regarding pastillage matrices.

Structural Hydrocolloid Scaffolding In 19th Century Pastry, Natalie M. Durdle Miss

Defensive Publications Series

This disclosure details the technical methods for creating shelf-stable, structurally sound, and aerated dairy compositions without the use of modern synthetic stabilizers. It documents the "Diplomat" and "Mousseline" systems as prior art for structural hydrocolloid scaffolding and fat-protein emulsification, specifically addressing the prevention of syneresis (weeping) and the deactivation of starch-degrading enzymes.

The Art Of The Kitchen Insult, Natalie M. Durdle Miss

Defensive Publications Series

This disclosure identifies a systematic method of political subversion embedded in the "High Art" of 19th-century French pastry, specifically the works of Marie-Antoine Carême. It identifies the "Crème Diplomate" and "Crème Bavaroise" as deliberate semiotic parodies of the diplomatic class and the Bavarian state. By utilizing gelatin-stabilized aerated structures, the "Chef-Architect" performed a public satire of "spineless" and "insubstantial" political entities, effectively "manufacturing" an illegitimate social class of chef-experts (the "Tiddy-Doll" effect).

Generating Dynamic Hazard Alerts Via Multidimensional Crowdsourced Data Collection, Chaim Zucker

Defensive Publications Series

Real-time navigation systems may issue generic hazard alerts, such as "car on shoulder," which may not differentiate between varying risk situations. A system for collecting multidimensional data about a reported event can be used to enhance these alerts. After an initial hazard report, the system can distribute simple prompts, for example, binary-choice questions such as "Are people outside the car? To subsequent passing drivers to gather contextual details. These crowdsourced responses can be aggregated, and an algorithm on a server may assess the data against a set of rules to determine a severity level. Based on this assessment, the system …

Context-Driven Grid Deformation And Deep State Injection For 3d Perception, Idan Kligvasser, Ehud Rivlin, Regev Cohen, George Leifman, Yotam Intrator

Defensive Publications Series

Existing 3D perception methods can be computationally inefficient due to static grids that may process irrelevant spatial regions and may struggle to integrate driving context. A framework can use a global state vector, derived from data such as vehicle ego-state and environmental information, to manage a perception pipeline. The state vector may concurrently drive two mechanisms: a context-driven grid deformation that can create a dynamic, foveated voxel grid to concentrate resolution on pertinent areas, and a deep state injection method using adaptive instance normalization to condition a network’s feature extraction at multiple layers. This integrated system may improve computational efficiency …

Cooperative Ue-Network Delivery Of Map Data For A Predicted Area Of Interest, Arunoday Bhattacharya, Andy (Jibing) Wang

Defensive Publications Series

A method is described for delivering map data to resource-constrained devices, such as smartwatches, operating on bandwidth-limited networks, for example, cellular or non-terrestrial networks. A user equipment (UE) can use sensors and contextual data to predict a user's trajectory and define an area of interest (AoI). The user equipment (UE) may send an area of interest (AoI) request, including its display characteristics, to a network. A network function can then fetch map data, perform server-side optimizations like tiling and layer selection, and computationally crop the data to a shape corresponding to the user equipment (UE) display before transmission. This approach …

Mobile User Interface With An Ergonomic, Bottom-Placed Primary Action And An Explicit, Non-Iconic Dismissal Control, Lin Chen, Yanhao Tan, Andrew Campbell, Onur Turna, Tom Woudenberg, Boriana Viljoen, Jaclyn Einis

Defensive Publications Series

A user interface for a mobile device, such as a smartphone or tablet, can present promotional content in a bottom-sheet container to address potential user engagement issues. The interface may utilize a vertical stack of call to action buttons, placing a primary action at the bottom of the stack to improve ergonomics by being within an easily accessible thumb zone. A conventional x close icon may be omitted. In its place, a secondary dismissive button can use explicit, descriptive text, for example, "Stay on website," to articulate the action's consequence. This configuration of ergonomic placement and an explicit, non-iconic dismissal …

Agentic System For Long-Form Audio Generation Using Recursive Synthesis And Multi-Dimensional Profiling, Ismail Haritaoglu, Leonid Kuligin

Defensive Publications Series

The automated generation of long-form audio may present challenges related to vocal consistency and quality control efficiency. A computer-implemented agentic system, which could operate on a server or cloud computing platform, can be used to address such challenges through a recursive review-and-synthesis loop managed by a multi-agent framework. The system may create a multi-dimensional profile for each text segment, defining instructions for attributes such as style, context, and emotion. It can then generate multiple audio candidates from one or more synthesis providers, and an auto-rater agent may score and select candidates based on their alignment with the profile. Low-scoring segments …

Fabrication Of Ewod Devices On Altic Substrate, Randi Walz, Christine Gan, Kshitija Mirkale, Akhil Namboori, Angela Moore, Jing Li, Marcus Mooney

Defensive Publications Series

Digital microfluidics represents a rapidly advancing domain offering precise control over small fluid volumes, spanning from microliters down to picoliters. Recognizing the limitations of conventional fragile substrates such as glass or silicon in extreme operational environments, this study introduces AlTiC substrates as a robust and effective alternative. In this work, we systematically detail the fabrication methodology for digital microfluidic devices utilizing AlTiC substrates, outlining the key processes needed to achieve reliable device functionality, thereby demonstrating fabrication of durable devices, and the specific fabrication nuances discussed may support further advancements in this technology.

Ai-Driven Dynamic Network Access From Communication Context, Nikhil Alampalli Ramu, Kamakshi R Devisetty

Defensive Publications Series

Proposed herein is an artificial intelligence (AI)-driven system that analyzes operational communication context to automatically identify network connectivity needs between systems. The system extracts network access intent from these communications, performs automated security validation, and dynamically generates temporary, policy-compliant network access permissions. By integrating contextual analysis with automated access control, the approach reduces manual configuration delays, improves operational efficiency, and optimizes the handling of legitimate connectivity requests in modern enterprise networks.

Adaptive Evaluation Coverage Alignment For Ai Capabilities In Enterprise Platforms, Jithu Paulose, Ashly Joseph, David C White

Defensive Publications Series

Proposed herein is a system that includes a control plane that measures whether current artificial intelligence (AI) evaluation coverage in a platform still matches real production usage. The system can provide such features by capturing structured live usage, mapping both production interactions and benchmark items into a shared workflow-intent graph, identifying uncovered usage regions, and surfacing those gaps so teams can decide whether to add or remove benchmark coverage or add or remove guardrails. The technical distinction of the system proposed herein is not ordinary drift monitoring or release gating. Rather, the technical distinction is the continuous alignment of real-world …

Intelligent Network Remediation System With Adaptive Prescriptions, Rohit Bahl, Mrulay Sureshbhai Mistry, Ronnit Roy Burman, Hemang Lavana, David C White

Defensive Publications Series

An artificial intelligence (AI)-driven self-adapting remediation system is proposed herein. The proposed system can facilitate iterative, validated remediation with real-time state adaptation, as well as learning-enhanced prescription ranking under multi-dimensional constraints and intent-centric, standards-based multi-modal translation. Thus, the proposed system represents a fundamental advancement in network remediation technology by combining AI-driven intelligence with deterministic safety validation through the Model Context Protocol (MCP) framework. The integration of Large Language Model (LLM)-based command generation, risk-aware planning, adaptive configuration intelligence, and MCP-based validation creates a novel solution that addresses the critical limitations of traditional network remediation approaches while providing robust safety guarantees through …

Protocol For Coordinated Distributed Computing Within Communication Networks, Bhavya Bhardwaj, Lokesh Kumar, Uma S Mohanty

Defensive Publications Series

Proposed herein is a technique, referred to as a Network-Integrated Distributed Computing (NIDC) protocol, which allows network devices to collaboratively execute distributed computational workloads by sharing embedded computing resources and routing computations based on real-time availability. The proposed NIDC protocol unifies compute orchestration, routing, and failure recovery directly within the network, eliminating reliance on external compute clusters. This enables scalable, resilient, and low-latency processing using existing network infrastructure.

Cooperative Management Of Cellular Parameters For Power Optimization During Wlan Connectivity, Vid Adiraju, Arunoday Bhattacharya

Defensive Publications Series

A system for cooperative power management can be implemented in a user equipment (UE), such as a smartphone, smart watch, or other wearable device, with concurrent cellular and wireless local area network (WLAN) connections. A UE can experience power consumption from its cellular radio, which may remain active with certain parameters even when data traffic is offloaded to a WLAN. After determining that a WLAN connection is stable, for example, the UE can signal its status to the cellular network. In response, the network can reconfigure the UE's cellular radio parameters, which can include applying a longer discontinuous reception cycle …

Machine Learning-Based System For Combining Behavioral Scores To Assess Advertising Inventory Value, Shuyang Sheng, Dongjing He, Richard Zippel, Xing Tan, Sahil Dhuper, Zachary Loebel Fried

Defensive Publications Series

Evaluating advertising inventory value can be difficult, as methods that assess individual signals in isolation may not fully capture the quality of web traffic. This technology provides a method for generating an unified metric to assess the overall value of ad inventory. The system aggregates multiple behavioral scores, such as user engagement, ad load, and content quality, along with other signals like spam fraction. A machine learning model is trained to combine these inputs into a single, non-binary “combined behavioral score” for each website or application. The model is trained using a dataset of manually evaluated websites, with labels derived …

A Generative Framework For Evaluating And Optimizing Geographic Experimentation Methodologies, Sam Bailey, Wincy Liu

Defensive Publications Series

A generative framework with supporting code library evaluates and optimizes disparate geo-experimentation methodologies on a finite sample of historical data. The evaluation process utilizes a multivariate moving block bootstrapping algorithm that generates synthetic histories while preserving cross-sectional dependencies to augment data in order to run hundreds of simulations prior to experiment launch. By simulating experiments across various generated datasets, the framework estimates the statistical validity, bias, and sensitivity of different experimental designs allowing users to compare outcomes from analysis methods (e.g. time-based-regression vs. synthetic controls) and runtime parameters. This approach provides a consistent basis for optimizing experimental designs across a …

System And Method For Assessing User Engagement And Ad Placement Quality, Xing Tan, Sahil Dhuper, Dongjing He, Yan Chen, Richard Zippel, Rui Zhang, Zachary Loebel Fried

Defensive Publications Series

The described technology provides a system for generating behavioral scores to assess digital publisher properties. The system calculates scores indicating the quality of content placements and user engagement to identify anomalous or low-quality traffic patterns. The evaluation is based on behavioral metrics derived from user interactions with the publisher content and promotional elements. This approach transforms varied interaction signals into normalized, ensembled scores that provide actionable insights into the underlying quality of the digital inventory. Keywords: behavioral score, user engagement, placement evaluation, anomalous traffic, digital advertising, content quality.

Measurement And Evaluation Of User-Perceived Advertising Load Across Digital Platforms, Dongjing He, Shuyang Sheng, Richard Zippel, Daniel Chang, Sahil Dhuper, Zachary Loebel Fried

Defensive Publications Series

This publication describes a technology for evaluating and scoring the advertising load presented to users across digital platforms. The system calculates a user-perceived advertising load by aggregating data at a user-session level and applying percentile-based statistical methods to determine a publisher-level score. This approach provides a robust mechanism for identifying digital environments with excessive advertising that degrades the user experience. The technology utilizes both logged event data and automated crawling techniques to construct a comprehensive view of the advertising landscape. Keywords: advertising load, user session, behavioral score, digital inventory, automated crawling, viewability.

Proactive Reservation Of Network Slices Based On Manual Or Contextual Triggers, Arunoday Bhattacharya, Amruth Ramachandran, Jayachandran C, Vid Adiraju, Pavan Nuggehalli

Defensive Publications Series

Systems and methods are described for a mechanism that supports the proactive reservation of network resources for a specified future time period. This reservation can be initiated manually by a user or automatically by an on-device agent that analyzes contextual signals, such as calendar events or location history, to anticipate a future need for enhanced connectivity. Following a pre-negotiation phase, the network can reserve the resources and deliver an updated routing policy to a user device, for example, a smartphone, smart watch, or augmented reality glasses. This can allow the device to activate the pre-provisioned network slice at the scheduled …

Solid State Forms Of 2-(3,5-Bis(Trifluoromethyl)Phenyl)-N-(4-(4-Fluoro-2-Methylphenyl)- 6-((7s,9as)-7-(Hydroxymethyl)Hexahydropyrazino(2,1-C)(1,4)Oxazin-8(1h)-Yl)-3- Pyridinyl)-N,2-Dimethylpropanamide And Processes For Preparation Thereof, Msn Laboratories Private Limited, R&D Center; Srinivasan Thirumalai Rajan, Sagyam Rajeshwar Reddy, Kammari Bal Raju & Garai Abhijit.

Defensive Publications Series

The present invention relates to solid state forms of 2-(3,5-bis(trifluoromethyl) phenyl)-N-(4-(4-fluoro-2-methylphenyl)-6-((7S,9aS)-7-(hydroxymethyl)hexahydropyrazino (2,1-c)(1,4)oxazin-8(1H)-yl)-3-pyridinyl)-N,2-dimethylpropanamide represented by the following structural formula-1 and processes for preparation thereof.

Formula-1

Audio-Synchronized Personal Media Playback In A Streaming Interface, Devin Carder

Defensive Publications Series

A technique is proposed for audio-synchronized personal media playback in a streaming interface. A user interface associated with a content streaming platform is provided for presentation via a client device. The content streaming platform provides a current audiovisual track for playback. A user interaction with one or more user interface elements is detected. A determination is made of an emotional tone and a temporal characteristic associated with the current audiovisual track. A set of content items corresponding to the determined emotional tone is identified from a personal content repository associated with the user. The user interface is updated to include …

A Process For The Preparation Of Crystalline Form I Of Fruquintinib, Msn Laboratories Private Limited, R&D Center; Srinivasan Thirumalai Rajan; Gade Srinivas Reddy; Bingi Hariprasad Reddy; And Matapathi Mahesh.

Defensive Publications Series

The present disclosure provides a process for the preparation of crystalline form I of Fruquintinib.

Systems And Methods For Detecting Silent Backward-Incompatible Api Contract Drift Using Runtime Structural-Semantic Fingerprints, Harsh Agrawal, Isha Navanath Shinde

Defensive Publications Series

A computer-implemented method for detecting silent backward-incompatible application programming interface (API) contract drift includes receiving API interaction data from one or more sources in a production environment, where the API interaction data includes request and response payloads for one or more API endpoints. The method includes normalizing the request and response payloads into structured representations and generating runtime structural-semantic fingerprints from the normalized payloads, where each fingerprint encodes field presence, field absence, hierarchical path structure, type signatures, nullability indicators, enumeration membership, and cross-field relationships. The method includes learning endpoint-specific behavioral invariants from historical API traffic, where the behavioral invariants capture …

Graph Based Context Ingestion And Automated Risk Identification Engine, Kautilay Siingh, Sanjeev Murthy, Shreyas Kc, Brianna Rezler, Aishwarya Hariharan Athreya

Defensive Publications Series

Systems and methods are provided for a risk identification and generation engine that performs security architecture analysis in a standardized manner. The engine automatically populates security questionnaires within a security architecture workflow and determines whether a feature requires a minor assessment or a detailed analysis. The engine generates dataflow and architecture diagrams, provides editing capabilities through drag‑and‑drop operations, and enables creation of additional diagrams. Threat models are generated using application context to identify security flaws at the design stage. The system stores design and technical security requirements, application details including summaries, APIs, and implemented controls, as well as records of …

Elastomeric Cowling For Pcb Connectors, Robert Hrabchak, Alberto Cavallaro, Ethan Hoying

Defensive Publications Series

Traditional metal cowlings for board‑to‑board connectors in electronic devices may add mass and occupy internal volume. These assemblies often rely on foam compression and multiple screws, which may lead to unpredictable retention forces and increased assembly complexity. Elastomeric retention assembly hardware is disclosed to secure connectors between a flexible circuit and a printed circuit board. The assembly includes an elastomeric band anchored by rigid bars to retention brackets on the circuit board. A stiffener with a convex, parabolic profile is positioned between the elastomer and the connector. This curved geometry facilitates a uniform distribution of downward force across the connector …

System And Method For Multi-Parcel Hoa Covenant Violation Detection From Spherical Imagery, Brice Fenger

Defensive Publications Series

Disclosed is a system and method for automated detection of homeowner-association (HOA) covenant violations from spherical (360°) imagery captured during a single drivethrough of a planned community. The novel aspects of the disclosure are: (1) the treatment of a single spherical image as a multi-parcel artifact, with violations attributed across all parcels whose geocoded locations fall within a radius of the image’s capture coordinates; (2) the per-parcel retrieval and merging of covenant rule sets - keyed by the specific parcel and, where applicable, by development phase - into a single multimodal large language model (LLM) prompt that performs both spatial …

System And Method For End-To-End Agentic Log Analysis, Verification, And Automated Remediation In Content Serving, Indu Ramamurthi, Raghav Lagisetty, Ajay Joshi, Srinath Narayanan, Sarvjeet Singh, Pradyumna Narayana

Defensive Publications Series

The present disclosure relates to an end-to-end agentic artificial intelligence system capable of autonomously ingesting live serving logs, isolating performance issues, mathematically verifying hypotheses using deterministic tool-calls, and deploying safe, low-latency fixes to a serving stack. The system chunks data into semantic problem slices and generates natural-language hypotheses. To prevent hallucination, the system executes deterministic queries against live data to verify its own hypotheses. Verified insights are translated into simple, non-machine-learning programmable rules that execute rapidly in production. These rules are pushed to a limited fraction of live traffic, where a real-time monitor compares performance against a baseline, automatically shutting …

Brand-Specific Conversational Agent With Lead Qualification, Balaji Pattabhiraman, Vijay Garg, Derrick Isoka, Deepak Yadav, Paneendra Ba, Teresa Batten, Teresa Chen, Lohitaksh Trehan, Shen Huang, Shruti Bhutada, Rina Jensen, Sudheer Sahu, John Amir-Abbassi, Ujjwal Gupta, Maneesha Nanda, Swati Mohanty

Defensive Publications Series

Keywords: conversational agent, brand-specific agent, lead qualification, search results integration, user intent, advertiser data aggregation, personalized user experience, dynamic call to action, AI agent training, lead prioritization

Abstract

Generating qualified sales leads for complex user journeys can be inefficient, often resulting in user drop-off and providing advertisers with little insight into user intent. This technique addresses these issues by integrating a brand-specific conversational agent directly into a content provider’s owned and operated surfaces, starting with the search results page. Trained on comprehensive advertiser data such as websites, business profiles, and user reviews, the agent provides detailed, real-time information, allowing users …

Dynamic Context-Aware And Personalized Insertion Of Sponsored Elements Into Llm-Generated Lists, Tim Whalin, Gagan Goel, Chris Monkman, Shrey Gupta, Hanny Kamal, Jason Galep, Shreya Kothaneth

Defensive Publications Series

Generative artificial intelligence (AI) systems frequently produce structured lists by synthesizing web data. Inserting ads into such lists can lead to jarring disruptions or redundant entries when integrated into these dynamic outputs. This creates a challenge in incorporating sponsored content without compromising user experience or trust, particularly when a relevant advertisement duplicates an organically generated item. The technology introduces a system that dynamically integrates sponsored elements into AI-generated lists by first employing a large language model to reverse-engineer the list’s context and derive one or more granular synthetic queries. This allows for the selection of sponsored candidates based on an …

Real-Time Context Engine For Dynamic Content Generation And Interface Adaptation, Tim Whalin, Otty Soemitro

Defensive Publications Series

A Real-Time Context Engine (RTCE) dynamically generates and adapts conversational content based on a user’s interactions with a webpage. The RTCE monitors client-side telemetry, such as scroll velocity and cursor position, and feeds these behavioral signals into a large language model (LLM) that functions as a real-time semantic orchestrator. For sponsored items, the system generates both an impartial “Objective Explainer” and a distinct “Advertiser Voice” through dual-voice orchestration, synchronizing them with the user’s expectations to maintain transparency. The technology also performs predictive mode-shifting by analyzing micro-interactions to infer shifts in user intent, automatically altering the user interface from one modality …