Digital Commons Network^™

Color-Coded Lane Guidance System For Dynamic Traffic Flow Optimization, Jacques Mouton

Defensive Publications Series

Traffic flow can be sub-optimal when drivers lack real-time, lane-specific information. This disclosure describes a system for dynamic lane guidance, for example, within a navigation application on a computing device (e.g., a smartphone, smart watch, wearable device, augmented reality glasses, an in-vehicle infotainment system, etc.). In some implementations, the system can utilize a probabilistic inference engine, such as a Bayesian inference engine, to analyze multiple data sources. These sources may include real-time traffic data, historical data, predictive models, and crowdsourced information. This analysis can be used to calculate a probability of congestion, delay, and/or accident risk for individual lanes. These …

Automatic Detection Of Multi-Customer Power Outages Using Aggregated Device Connectivity, Daniel Klein, Mel Bass

Defensive Publications Series

Some existing methods for detecting power outages may rely on manual customer reports, which can be slow, or inaccurate, and may be unfeasible when communication networks are overwhelmed or damaged. A method is disclosed that can enable automatic outage detection by aggregating and analyzing the connectivity status of a large number of internet-connected devices (e.g., smart speakers, IoT appliances, network routers) across multiple locations. A simultaneous loss of communication from a high density of geolocated devices in a particular area may be used to infer an outage. By correlating signals from diverse devices and service providers, including, for example, those …

Verifying Ownership Of Online Profiles And Web Sites For Creators And Publishers, Nick Krasney, Yi Li, Govind Kaushal

Defensive Publications Series

Verifying that a single creator or publisher controls multiple disparate online properties, such as websites and social media profiles, can be a challenge. Existing methods may rely on bilateral agreements between platforms, which may not be scalable and can create friction. This disclosure describes a decentralized cryptographic method for verifying ownership. A creator may generate a unique public-private key pair and can use the private key to sign the canonical URL of each online property, creating a digital signature. The creator can then provide the public key and the corresponding digital signature to each platform, which may embed this information …

Automatic Accessibility Verification Using Multimodal Data, Yash Dave, Jaydeep Untwal, Rakhi Parekh

Defensive Publications Series

Accessibility information on mapping platforms can be self-reported by businesses, which may lead to uncertainty for users with accessibility needs. This disclosure describes a method for automatically verifying accessibility claims using multimodal data. A system can analyze street-level imagery to detect physical features, such as ramps, and can process user-generated text reviews for accessibility-related keywords. This combined data may be input into a fine-tuned AI model to assess the presence of accessibility features at a given location. The outcome can be an indicator, for example a badge on a business profile, which can provide users with verifiable information, assisting them …

Continuous Private Location Tracking Of Mobile Devices Using Satellite Communications And Other Infrastructure, N/A

Defensive Publications Series

Some existing location tracking methods may not provide reliable, real-time data for mobile devices in environments with limited or no cellular connectivity, such as over air and sea. The disclosed system can address this by using a hybrid communication method that may dynamically switch between networks such as satellite (satcom), wireless local area networks (WLAN), cellular, and other networks. The system may employ virtual geofences to trigger location updates, which can help optimize power consumption. During periods of no connectivity, a predictive analytics model may estimate the device’s position by integrating device sensor data, historical travel patterns, and external data …

System For Modeling Local Inventory From Multimodal, Unstructured Data Signals, Cathal D'Arcy, Nicolas Garnier, Yiqing Zhao, Alisa Khatipova, Anton Minashkin, Jack Collins, Kathrin Schuler, Pawan Kumar, Richard Ledley

Defensive Publications Series

The disclosed technology describes a method for programmatically predicting conceptual-level product availability, such as for brands, categories, and product lines, for stores that can lack detailed inventory feeds. A machine learning model can ingest and analyze multimodal signals, including unstructured data from user-generated content like in-store photos, crawled website, and text/images. The model processes this evidence to predict the level of confidence that a given product concept is available at a specific local store. This method can increase local product data coverage without requiring formal inventory feeds from merchants, enabling a system to surface more relevant in-store product information to …

Textvantage, Palakh Shangle

Defensive Publications Series

The present disclosure discloses TextVantage is AI-powered accessibility framework designed to enhance digital finance platform for dyslexic users and merchants. The Present disclosure includes UI Enhancements include Dyslexic-friendly fonts (e.g., OpenDyslexic), text simplification, increased line spacing, and text-to-speech narration. Further, hybrid inputs include Text queries, voice commands (via Web Speech API), autodetection of complex text, and visual toggles for font/spacing/narration. Further, RAG Pipeline retrieves dyslexia research, WCAG guidelines, and text simplification strategies to power an LLM that generates tailored responses (e.g., simplified text, merchant code). The present disclosure discloses merchant support includes Auto-generates dyslexic-friendly code or content via file …

Rank-Based Secure Multi-Ai Agent Communication System For Misinformation Prevention, Sree Gowri Addepalli, Sushma Mahadevaswamy, Arica Schivonne Jackson, Barry Qi Yuan

Defensive Publications Series

This submission introduces a novel approach to prevent the generation of artificial intelligence (AI) misinformation by proposing rank-based multi-AI agent communication with novel dynamic parameter adaptive re-ranking technique based on various security behavioral metrics and data recency. This approach provides the ability to minimize the spread of outdated information or misinformation, ensuring more trustworthy outputs from AI systems.

System And Method For Managing Invoice-On-Behalf Within Open-Loop Transaction Networks, Anna Grau Galvan, Harrison Robert Sylvester, Seani Ramabulana, Adi Mistry, David Blower

Defensive Publications Series

The present disclosure provides a method and a system for managing Invoice-on-Behalf (IoB) within open-loop transaction networks. The method comprises identifying a portable financial device enabled for IoB at a merchant system and generating an authorization request for an IoB transaction. The authorization request is then approved at an issuer system, followed by transmitting an authorization response back to the merchant system. At a Point-Of-Sale (POS) system associated with the merchant system, issuance of a VAT invoice is suppressed. Line-item data and tax data related to the transaction are transmitted, and the issuer system generates IoB documents at predetermined intervals …

Analog Channel Fault Detection, Anonymous

Defensive Publications Series

The present disclosure relates to analog fault detection in amplified spontaneous emission (ASE)-loaded optical line systems, specifically addressing channel power degradation scenarios that are not detectable by conventional digital methods. The disclosed solution initiates fault detection based on integrated, filtered channel power measurements taken at the line-out location of a section multiplexer (section-mux). These measurements are back-scaled to the corresponding section-mux input and compared against forward-scaled measurements and expected channel input power. By comparing these scaled power values, the solution accurately distinguishes between upstream faults (such as incorrect transmitter frequency tuning, silent pixel failures in wavelength selective switches, or fiber …

Smart Splitter In Passive Optical Network, Anonymous

Defensive Publications Series

A smart splitter system is disclosed for enhancing security, reliability, and remote management capabilities in Passive Optical Networks (PONs). The smart splitter integrates a dedicated controller, controllable mirrors, and internal optical light absorbers, leveraging the ONU Management and Control Interface (OMCI) protocol for dynamic control. The Optical Line Terminal (OLT) utilizes OMCI Managed Entities (MEs) to remotely enable or disable splitter output ports, swiftly isolating rogue Optical Network Units (ONUs) upon detection. This OMCI-driven splitter addresses security vulnerabilities posed by unused splitter ports, rogue ONUs transmitting outside allocated time slots, and malicious upstream optical interference. By automating port isolation via …

Techniques To Provide Accurate, Real-Time Power Supply Unit (Psu) Efficiency Reporting, Tina Jiang, Karen Wang

Defensive Publications Series

With the development of artificial intelligence (AI) infrastructure, power consumption for networking equipment continues to be a concern. Notably, the power supply units (PSUs) of such equipment can consume considerable power and it can be critical to track PSU consumption in network deployments. Techniques proposed herein provide for the ability to synchronize input and output power sampling and calculations through the use of primary and secondary zero-crossings of an Inductor-Inductor-Capacitor (LLC) resonant converter. The proposed techniques can enable real-time, accurate efficiency reporting for a PSU without requiring extra hardware synchronization circuits, thereby making efficiency monitoring more reliable and accessible.

The Tech Audit Revolution: Navigating The Next 50 Years For Industry And Technology Leadership And An End-To-End Audit Framework For Ads, Search, And Content Platforms, Ramkumar Bharathan

Defensive Publications Series

The internal audit function stands at a pivotal juncture, undergoing a profound transformation from its traditional role in compliance to a strategic enabler of business value. For leading technology companies, the ability to safely deploy cutting-edge innovations is a core competitive differentiator, making a proactive and technologically advanced audit function essential for secure and rapid market entry. This paper presents a comprehensive, revolutionary audit framework designed to navigate the complex risk environment of the next five decades, addressing both current challenges and future technological frontiers. This framework provides a deep dive into auditing emerging technologies such as Quantum Computing, …

Exploring And Improving Initialization For Deep Graph Neural Networks: A Signal Propagation Perspective

Defensive Publications Series

Graph Neural Networks (GNNs) often suffer from performance degradation as the network depth increases. This paper addresses this issue by introducing initialization methods that enhance signal propagation (SP) within GNNs. We propose three key metrics for effective SP in GNNs: forward propagation, backward propagation, and graph embedding variation (GEV). While the first two metrics derive from classical SP theory, the third is specifically designed for GNNs. We theoretically demonstrate that a broad range of commonly used initialization methods for GNNs, which exhibit performance degradation with increasing depth, fail to control these three metrics simultaneously. To deal with this limitation, a …

Llm-Based Navigation Guidance Revisions Based On Conversational Context, Florian Hartmann, Matthew Sharifi

Defensive Publications Series

When a vehicle has multiple occupants engaged in a conversation, audio instructions from a digital map/ navigation application can be disruptive and/or unnecessary. This disclosure describes techniques, implemented with specific user permission to access and analyze data related to the presence of multiple users in a vehicle and the conversational context. With user permission, the ongoing conversation in the vehicle is analyzed locally using a large language model and the analysis is used to enhance the audio delivery of dynamic navigational guidance. A transcript of the conversation related to the ongoing drive is provided to an LLM along with a …

Alphafold Database Debiasing For Robust Inverse Folding

Defensive Publications Series

The AlphaFold Protein Structure Database (AFDB) offers unparalleled structural coverage at near-experimental accuracy, positioning it as a valuable resource for data-driven protein design. However, its direct use in training deep models that are sensitive to fine-grained atomic geometry—such as inverse folding—exposes a critical limitation. Comparative analysis of structural feature distributions reveals that AFDB structures exhibit distinct statistical regularities, reflecting a systematic geometric bias that deviates from the conformational diversity found in experimentally determined structures from the Protein Data Bank (PDB). While AFDB structures are cleaner and more idealized, PDB structures capture the intrinsic variability and physical realism essential for generalization …

Contextual Explanation Of Local Differences On Digital Map User Interface, Florian Hartmann, Matthew Sharifi

Defensive Publications Series

Different regions of the world have different regulations, e.g., related to driving, parking, etc. as well as different social norms, e.g., tipping, acceptable use of a mobile device on public transport, etc. Such differences can trip up travelers when they are in unfamiliar locations. This disclosure describes techniques to automatically determine via offline processing by a large language model (LLM). The differences are stored in a database and are used to surface proactive contextual alerts to users at appropriate times by explaining relevant differences between rules or norms at their current location and their typical location. The LLM can be …

Implicit Graph Neural Networks With Flexible Propagation Operators

Defensive Publications Series

Due to the capability to capture high-order information of nodes and reduce memory consumption, implicit graph neural networks have become an explored hotspot in recent years. However, these implicit graph neural networks are limited by the static topology, which makes it difficult to handle heterophilic graph-structured data. Furthermore, the existing methods inspired by optimization objectives are limited by the explicit structure of graph neural networks, which makes it difficult to set an appropriate number of network layers to solve optimization problems. To address these issues, we propose an implicit graph neural network with flexible propagation operators in this paper. From …

Space Socks: Quantum Entanglement Simulation Game, Suma Mallapragada, Celeste Gnits

Defensive Publications Series

Quantum teleportation and entanglement are foundational topics in quantum mechanics that are difficult to visualize and intuitively grasp. “Space Socks” is an interactive educational game that demystifies these quantum phenomena through playful simulation. Developed as part of the Celeste GNITS Space Research initiative, this game leverages quantum circuit logic and entangled metaphors to teach quantum teleportation, quantum internet, and entanglement swapping. This whitepaper outlines the scientific basis, game mechanics, implementation, and educational impact of the Space Socks project.

By publishing the work and its associated repository publicly, we are preventing the possibility of patent claims on the unique ideas …

Systems And Methods For Training Frameworks To Optimize Model Based On Business Goals, Chinamy Narendra Sonar, Shuhan Zhang, Keyi Li, Lin Meng, Can Liu, Shubham Agrawal, Chiranjeet Chetia

Defensive Publications Series

[0001] The present disclosure provides a system and method for providing training frameworks to optimize model based on business goals. The processor of the training framework system introduces novel model evaluation metrics which are directly aligned with business goals and capture most business use cases for A2A fraud detection models. Metrics can be used in model training and evaluation. In the first approach, the processor of the training framework system sets validation criteria to a new metric but training the model with traditional loss. In the second approach, the processor of the training framework system trains and validates the model …

Predictive Modeling Ai Agent, Madhusudan St

Defensive Publications Series

The present disclosure related to the field of Artificial Intelligence (AI), in particular to AI agents for analyzing data and generating insights in response to business transaction queries. The one or more user devices are associated with the one or more users, in which an AI agent is deployed in the one or more user devices. Further, the one or more users provide one or more requests such as plain text queries, and the AI agent interprets and understand the one or more requests accurately. Furthermore, the AI agent performs ad-hoc analysis on the one or more requests, to shed …

Managing International Remittance Transactions In A Closed Loop Transaction Network, Vinay Babu Chennareddy, Jaishankar Kumar Venkataraman, Anil Kumar, Richard Meszaros

Defensive Publications Series

The present disclosure provides a method and a system for managing international remittance transactions in a closed loop transaction network. The method includes receiving an initiation request for an international remittance transaction that is funded by a sender via a sender funding financial institution system. The initiation request may originate from a receiver and be transmitted to the sender for review, approval and funding. The initiation request is validated at the sender funding financial institution system and transmitted to a transaction service provider system. The validated transaction is then exclusively routed to a designated financial institution system that is pre-associated …

Method And System For Enabling Chromatic Accessibility Of User Interfaces Related To Inclusive Digital Finance, Palakh Sanghle

Defensive Publications Series

Methods and systems, for enabling chromatic accessibility of UIs related to inclusive digital finance is disclosed. The present disclosure includes collecting user input with diverse data from users of an inclusive digital finance platform. Further embedding the user input by converting the user input to vector representations. Followed by, retrieving documents which are relevant to the vector representation of the user input based on a similarity criteria. Processing the vector representation of the user input in accordance with the relevant documents using Large Language Model (LLM). The LLM configured to generate suggestions as a response to the user input. Further, …

Latency Reduction In Conversational User Interfaces By Selective Execution Of Multi-Pass Speech Recognition Based On Llm Evaluation, Petar Aleksic, Lillian Zhou

Defensive Publications Series

Conversational interfaces enable users to interact with a virtual assistant, chatbot, or other software via spoken audio. In a cascaded conversational system architecture, an automatic speech recognition (ASR) model transcribes a user’s spoken query to text, a large language model (LLM) generates the text of a response, and a text-to-speech (TTS) model generates response audio from the response text. This configuration is subject to high latency due to the need to wait for the ASR model to generate a transcription before the LLM response can be calculated. For a high-quality, full-context transcription, the ASR model can take time. Per techniques …

Dynamic Device Configuration Based On User-Defined Context, Ant Oztaskent, Aiko Nakano

Defensive Publications Series

This document describes techniques that enable a computing device (e.g., a smart watch, ring, glasses, etc.) to support user-defined contextual information display and configuration through natural language. The techniques of this disclosure may address the problem of complicated menus and settings and unreliable predictive solutions for displaying personalized and/or general information (e.g., stock prices) by introducing voice controls and artificial intelligence (AI) assistance. The techniques of this disclosure leverage voice input and natural language processing to allow a user to specify what information the user wants to see and the conditions under which the information should be displayed. In one …

Genai For Merchant Category Identification, Neethu Prem, Vinay Gaonkar, Hazar Yuksel, Kavalpreet Kawatra, Siva Natarajan, Anubhav Kumar, Rupal Gakkhar, Shikhar Shukla, Anupam Chauhan

Defensive Publications Series

The present disclosure related to a system and method for merchant category identification. The method includes extracting historical data and merchant’s transactional information. The method further include detecting miscode merchant category code based on the statistical and deep learning models. Subsequently, the method include sharing list of miscoded merchants to the business.

Effective Method Of Encoding Data Into Images For Recovery After Dsc, Kellie Shaw

Defensive Publications Series

Systems and methods for embedding metadata within image content is provided. One of these methods comprises generating, by the computing device, an image frame that includes a plurality of pixels corresponding to multiple color channels defining color characteristics of an image. The generating includes assigning a common value to a pixel subset of the plurality of pixels located on a first row of the image frame. The pixel subset corresponds to the multiple color channels defining the color characteristics of the image. The method also includes converting, by the computing device, the image frame to a luminance chroma format.

Agentic Ai Framework For Dynamic Generation, Verification, And Presentation Of Interactive Content In Search Results, Tushar Chugh, Aditya Mone, Karthik Kumara

Defensive Publications Series

This disclosure describes techniques for enhancing search engine results pages (SERPs) by employing an agentic AI framework. A framework is presented that analyzes user queries to determine the optimal presentation format for the response. For certain predefined query types (e.g., real-time data like stocks, sports scores), the framework retrieves existing interactive widgets. For other queries deemed suitable, the framework dynamically generates new, bespoke interactive widgets or simple web applications ("micro-apps") tailored to the detected intent of the query. This generation process involves the use of specialized artificial intelligence (AI) agents collaborating to design, build, fetch data for, and verify the …

Generating Speaker Notes For Presentations Using Large Language Models, Charles Lai, Vinayak Reddy

Defensive Publications Series

This disclosure describes techniques for generating speaker notes for presentations using large language models. The contents of a slide deck are provided as input to a large language model. A prompt is given to the model, instructing it to create speaker notes from the slide content. Additional related content, if available, can also be provided as input. The prompt can be configured to generate notes that are specific to each slide, while also connecting ideas between slides to provide a natural narrative flow. The generated speaker notes can be tailored to the user's presentation context and personal style.

Enabling Privacy-Compliant And Context-Aware Tool Use By Artificial Intelligence Agents, Victor Carbune, Nevan Wichers, Oriana Riva

Defensive Publications Series

This disclosure describes techniques that enable an AI agent to learn the appropriate mode of access to user data. In a first mode, the agent can obtain user-related data by directly asking the user and obtaining explicit permission from the user. In a second mode, the AI agent determines situations where it is appropriate to obtain, with user permission and based on user interactions and settings, user data via authenticated apps. The techniques enable the agent to discern when to access a certain personalization tool based on inferring what the user may find appropriate. The techniques enable an AI agent …