Digital Commons Network^™

Interactive Photo Exploration Agent With Chained Language Model Calls, D Shin

Defensive Publications Series

Some photo library applications provide a conversational interface that enables users to retrieve photos that match their query. However, current interfaces are not multi-turn and are limited to retrieval and cannot automatically perform generation tasks. This disclosure describes an interactive, conversational interface for a digital photo library that allows users to explore their photos using natural language and receive generative, insightful answers. The exploration is multi-turn and multimodal, and is powered by a multimodal large language model (LLM). With user permission, the LLM is employed to interpret both the user's query (e.g., entered as free-form text) and the content of …

Generating Interactive, Llm-Driven Annotations For Inclusion In Digital Image Files, Ilia Labzovsky, Danny Karmon

Defensive Publications Series

While photographs capture visual information, they lack depth and interactivity. Contextual information, memories, stories, and related data associated with a photograph are external to the image file itself, and are scattered across applications, cloud services, or within the user’s mind. This disclosure describes techniques to enhance a digital image by automatically generating interactive annotations using a large language model (LLM) and incorporating the annotations into the image. Entities within a photo are identified and an LLM is instructed to generate an annotation object for the recognized entities. The annotation object is a self-contained package of data and instructions that specifies …

Integrating Generative Animations Of Photographs Into A Coherent Video Journey, Ilia Labzovsky, Danny Karmon

Defensive Publications Series

Users enjoy creating slideshows, animated videos, etc. that capture their trips. Current techniques to prepare such creations are limited by the content of the captured photographs and user knowledge of such features. This disclosure describes the use of artificial intelligence to automatically generate a dynamic video journey from a collection of user photographs. With user permission, an artificial intelligence model performs semantic analysis of user photos to infer the user’s geographic path during the photo capture journey. A generative video model accesses street-level images along the path and synthesizes a base video that simulates first-person movement along the path. The …

On-Device Measurement Of Impact Of Artificial Intelligence Tools On Productivity, Navid Kagalwalla, Tom Piacentini, Jacques Mouton

Defensive Publications Series

This disclosure describes techniques to measure the net productivity impact of AI-assisted workflows on user productivity. Metadata regarding user activity on a user device is obtained with user permission, using a private, on-device architecture. The data is analyzed on-device to determine whether an AI-assisted workflow is in process and the impact of the use of AI on productivity. In an example implementation, an Activity Sensing and Filtration Layer (ASFL) and a Cognitive Workflow Engine (CWE) are implemented and used to calculate the true net time saved using AI assistance. This metric takes into account the hidden cost of the user …

A Distance Based Embedding Method Using Nlp Methods To Identify Proximity To A Network, Shuhan Zhang, Anurag Tangri, Chinmay Narendra Sonar, Lin Meng, Can Liu, Chiranjeet Chetia

Defensive Publications Series

The present disclosure relates to the field of Artificial Intelligence (AI), in particular to AI methods and systems for analyzing transactional data using Natural Language Processing (NLP) techniques to identify latent relationships and proximity within networks. The disclosed system utilizes text fields within datasets to categorize input data into appropriate classifications. The input comprises a dataset containing text data that provides contextual information related to transactions. Feature embeddings are generated using a sentence-based Language Learning Model (LLM), which captures semantic meaning of sentences. The LLM converts input data of varying lengths into fixed-dimension embeddings. The sentence-based LLM is fine-tuned based …

Llm-Augmented Dynamic Selective Blocking Of Service Calls For Application Control, Johannes Start, John Lunney

Defensive Publications Series

Software applications face critical challenges in rapidly and precisely mitigating emergent threats such as safety incidents or query-of-death attacks originating from or affecting backend calls. This disclosure describes techniques for dynamic, real-time, intelligent blocking or modification of harmful or problematic backend service call responses. The techniques combine traditional filtering with semantic analysis using a large language model (LLM). A control plane enables operators such as site reliability engineers (SREs) to define and deploy mitigation rules on-the-fly, without code changes or lengthy propagation delays. This enables rapid and precise response to emergent threats, such as safety incidents involving AI-generated content or …

Enhancing Hardware Scan Coverage By Selecting A Data Streaming Interface Unconnected To The Failure, Kaushal Purohit, Victor Wong

Defensive Publications Series

Hardware scanning helps debug problems related to internal components within a device by capturing hardware states as a stream of bits, known as a scan chain. However, employing on-chip or off-chip memory for storage of scan data requires that the path to/from the memory be excluded from the hardware scan, creating a critical gap in debugging. Moreover, wired debugging interfaces are typically unavailable or inaccessible on remote production devices, which makes it challenging to retrieve the scan data quickly. This disclosure describes techniques to optimize the streaming of hardware scan data from on-chip crash manager applications employed to detect hardware …

Optimizing Data Transfer From Smart Glasses To Paired Device Based On Eye Tracking, Willy Horng Jean Cheung, Shenil Navin Dodhia, Gordon Wan

Defensive Publications Series

Smart glasses and other wearable devices that include cameras can capture the user’s environment as memories and can also answer questions or surface assistance proactively. However, smart glasses have limited compute and storage capabilities, and enabling continuous perception is therefore infeasible. A paired device such as a smartphone can enable smart glasses to operate in continuous perception mode and also provide compute power for artificial intelligence models. However, a key bottleneck is the data transfer between the smart glasses and the paired device, which is constrained by wireless data transfer rates. Even with aggressive data-compression schemes, the payload over the …

Dynamic Management Of Virtual Lanes And Buffer Zones For Autonomous Traffic, Sumukh Shevde, Kamran Mustafa, Abhijit Adsule

Defensive Publications Series

Static assignment of roadway lanes can result in inefficient traffic flow, where congestion may develop in one direction while opposing lanes remain underutilized. To address this, systems and methods for dynamic traffic management are described. A computational system, such as a central server or an in-vehicle computer, can treat a roadway as a single, partitionable surface, creating and modifying virtual lanes in near real-time. The number, width, and direction of these lanes can be adjusted based on current and predicted traffic demand. Autonomous or semi-autonomous vehicles can be assigned to these virtual lanes, and a dynamic virtual buffer zone may …

A Real-Time Computer Vision System For Procedural Verification In Manual Assembly, Frank Mao, Sun Nee Chan, George Liu

Defensive Publications Series

Manual assembly processes can be susceptible to error, potentially leading to incorrect component handling or procedural deviations. A system for real-time procedural verification can use computer vision to address this. An inspection node, which may comprise an imaging device (e.g., a camera) and a processing unit, can capture and analyze an operator's physical actions at a workstation. These actions may be compared against a predefined digital blueprint, such as a standard assembly model (SAM), that codifies a specified sequence of events. Upon detection of a deviation between the operator's actions and the SAM, the system can provide feedback to the …

Method And System For Automated Incident Lifecycle Management With Self-Learning Agents, Soumendra Kumar Mishra, Amitabh Ranjan

Defensive Publications Series

The present disclosure relates to a system and method for automating the resolution of incidents in enterprise environments through a modular multi-agent orchestration framework. The system includes an intelligent log processing engine that analyzes and cleanses enterprise logs to extract relevant error contexts, a repository intelligence agent that maps error stack traces to corresponding code files, and an incident analysis agent that identifies root causes while generating actionable insights. A code generation agent produces targeted code fixes based on these insights, while a conversational assistant provides real-time guidance and validation for developers. The framework implements a standardized communication protocol for …

System For Llm-Driven Orchestration And Live Migration Of Isolated Software Agents, Johannes Start, John Lunney

Defensive Publications Series

Systems for managing autonomous software agents in distributed environments can be limited by a lack of runtime mobility, rigid communication protocols, and poor observability. An integrated platform for executing software agents can combine isolated agent execution environments for fault containment with a live agent migration module for stateful mobility between host computing devices, such as physical servers or virtual machines. A central communication fabric, which can utilize a large language model (LLM), may mediate inter-agent communication. The LLM can analyze the semantic intent of a request to dynamically generate and execute a multi-step orchestration plan. This approach can support resilient …

Llm-Powered Agents For Context-Aware Validation Of Code And Configurations, Peng Chen

Defensive Publications Series

Enforcing complex, team-specific software development rules that are beyond the scope of some static analysis tools can present challenges, and manual review for such rules may be slow and error-prone. The described techniques may address these challenges by employing automated agents powered by large language models within a code review workflow. These agents can be configured with natural language prompts that codify domain-specific knowledge and complex validation logic. When triggered by a code change, an agent can analyze the modification and its relationship to other relevant files in the codebase. This method can provide automated, inline feedback on adherence to …

Churchmesh Communicator: Dedicated E-Ink Mesh Messaging Device Architecture, Matthew D. Judson

Defensive Publications Series

Disclosed is the hardware architecture for the ChurchMesh Communicator, a dedicated, ultra-low-power mesh messaging device. The invention details the specific integration of components optimized for long-range, resilient communication: an ESP32-S3 MCU, an E-Ink display (for sunlight readability and zero static power draw), a physical QWERTY keyboard, and an SX1262 LoRa module (10km+ range). The architecture is optimized for 2+ week battery life. The firmware (ESP-IDF + LVGL) utilizes novel power management techniques, including E-Ink partial refresh optimization and aggressive deep sleep modes, waking only on keyboard interrupt or incoming LoRa message. The device integrates "KTKN-Lite" for decentralized synchronization.

Deterministic Data Synthesis Pipeline For Multimodal Ai Training (The Trueos-Ktkn Data Engine), Matthew D. Judson

Defensive Publications Series

Disclosed is a system for generating "Golden Master" training datasets for Multimodal Vision-Language-Action (VLA) models. The core innovation is a "Deterministic Render-Capture Loop" that decouples simulation time (simTime) from wall-clock time. The system advances the simulation by a fixed delta ($\Delta t$), renders the state, and captures the frame using WebCodecs, timestamping it with simTime regardless of actual render duration. This guarantees a mathematically perfect framerate and perfect synchronization between video frames and telemetry logs, eliminating the frame drops and jitter inherent in real-time screen recording. The pipeline is orchestrated within the TrueOS/Electron shell, managing the simulation backend (e.g., Python …

Stratospheric-Terrestrial Hybrid Mesh Network (Churchmesh & Winch Data Mule Protocol), Matthew D. Judson

Defensive Publications Series

Disclosed is a hybrid network topology combining terrestrial LoRa mesh nodes with stratospheric Aerostat Data Mules. The terrestrial layer (ChurchMesh) utilizes ultra-efficient hardware (RAK4631/nRF52840) for 10-14 day autonomy and the Meshtastic ROUTER_LATE configuration to prevent packet storms. The core innovation is the "Winch Data Mule Protocol" (the "Yo-Yo" maneuver) for bridging Line-of-Sight gaps using Tethered Aerostats. The aerostat collects data at low altitude (~200ft), then the winch releases it to high altitude (~2,000ft) to gain LOS ("Virtual Ionosphere") and transmit stored bundles. The aerostat then winches back down. This mechanical data muling conserves airborne battery and minimizes weather exposure and …

The Trueos/Aether Shell: A Sovereign "App Browser" Architecture For Local-First Orchestration, Matthew D. Judson

Defensive Publications Series

Disclosed is the architecture for TrueOS/Aether, a desktop-class Integrated Development Environment (IDE) designed as a sovereign "App Browser" for orchestrating complex, local-first applications. The architecture utilizes the Electron framework, prioritizing rendering consistency and robust local process management. The core innovation is the "Native Child Process Orchestration" mechanism, where the Electron Main Process natively spawns heavy local backends (e.g., Python AI environments like ComfyUI) as direct OS child processes. This bypasses the serialization bottlenecks of traditional IPC bridges, enabling high-throughput communication via mechanisms like ZeroMQ or shared memory. The architecture guarantees a deterministic rendering environment by bundling a specific Chromium version, …

Wireless, Fully-Sealed Coxswain Audio Transmission System With Modular Boat Electronics, Wireless Charging, And No Exposed User-Side Electrical Connections, Glenn R. Engel

Defensive Publications Series

This defensive publication describes a fully sealed, wireless coxswain audio communication system designed for rowing shells, eliminating all exposed electrical connectors commonly found in traditional cox box systems. The solution integrates a waterproof wearable headset with digital MEMS microphone(s), a Bluetooth LE Audio transmitter, and Qi-based inductive charging within a connector-free housing. A boat-mounted receiver/amplifier module wirelessly receives audio, drives existing speaker wiring, and supplies inductive power to portable components. An optional detachable user-interface module provides visual feedback, system control, and additional charge-sharing capability through wireless power exchange. The system architecture enables complete wireless communication among all user-handled components, reducing …

Automated Segmentation Of Impersonators (Fake Participants) In Virtual Meetings Using Quarantine Breakrooms, Bhavik Pradeep Shah, Chidambaram Arunachalam

Defensive Publications Series

This submission proposes novel techniques to automatically segment impersonators (fake users) detected in virtual collaboration meetings by dynamically moving them to a quarantine virtual room. Security administrators of an organization can be notified about segmentation events and may have the ability to join a quarantine room, observe the activities of impersonated users, assess potential business impacts of such users, and use the learning to further enhance impersonator detection techniques. A virtual meeting controller that orchestrates a virtual meeting can dynamically inform network controller and management systems in collaboration cloud networks to automatically apply network policies to block new sessions or …

Automated Execution Of Manufacturing Flowcharts Using Genai, Marcus T. Heron

Defensive Publications Series

This invention is the process of taking any image of a flowchart and using multiple GenAI Agents to execute the depicted process. In manufacturing flowcharts are used prolifically to describe multiple business practices and workflows, therefore this invention has wide applicability across a range of use cases. Enabling full automation of the depicted process in a reliable and repeatable fashion without the risk of human error and in a much faster time.

Scan To Email Summary, Hp Inc

Defensive Publications Series

This innovation focuses on the automated use of imaging technologies to simplify a well-known user experience that is frustrating to many customers today - the ability to scan to email, and then easily retrieve the scanned content. Providing an automated, simplified experience that results in significantly improved customer satisfaction, being able to search and retrieve emails that they have scanned on a multi-function device through their regular email client. The system leverages Large Language Models (LLMs) with engineered prompts to generate contextually relevant filenames, titles, and summaries for scanned documents.

The Master Engineering Compliance Atlas: A Unified Architecture For Automating Global Regulatory Governance, Ai Safety, And Cyber Risk, Ramkumar Bharathan

Defensive Publications Series

This technical disclosure presents the "Master Engineering Compliance Atlas," a comprehensive framework designed to bridge the gap between complex global regulations and modern software engineering practices. As regulatory landscapes fragment across jurisdictions, this document provides a unified methodology for harmonizing requirements from major frameworks, including Sarbanes-Oxley (SOX), GDPR, the EU Digital Markets Act (DMA), the Digital Services Act (DSA), the EU AI Act, NIST SP 800-53, and ISO 27001.

Unlike traditional legal summaries, this framework focuses on technical implementation, offering specific architectural patterns, "L7" control designs, and automation strategies to achieve "Compliance as Code." It introduces a "Mega-Matrix" that maps …

Generating A Personalized Artificial Intelligence Assistant For Domain-Specific Professional Work, Jan - Marten Ahrend

Defensive Publications Series

Professionals in specialized fields rely on their deep, domain-specific understanding that includes documented and undocumented knowledge such as project history, team dynamics, product goals, established communication styles, etc. Artificial intelligence (AI) assistants that support professionals may not capture the deep context specific to the domain, brand, and/or organization of the professional. This disclosure describes techniques for generating a personalized artificial intelligence (AI) assistant that is deeply contextualized to a specific professional’s unique work environment. Designed for domain-specific research and professional work, the AI assistant deploys a structured framework that enables the professional user to integrate their unique context and professional …

Ai-Driven Autonomous Content Generation For Trending Topics, Steven Dropsho, Dhruv Bakshi

Defensive Publications Series

Generative artificial intelligence models are capable of generating content in different modalities such as text, image, audio, video, and combinations. While user-driven content generation, where a user enters a prompt about a topic of interest and receives AI-generated content in response, is popular, this requires users to identify topics and seek content. This disclosure describes techniques for autonomously generating high-engagement content in various modalities. A continuous loop of input (identifying topics of interest), output (content generation based on the topics of interest), and metrics (engagement with the generated content) is used to generate new content that provides users engaging online …

Real-Time Deepfake Detection During Video Conferences: Harnessing Bidirectional Audio-Video Features From Reflections And Echoes, Mahesh Godavarti, Ali Mouline, Kamil Wojcicki, Radhakrishna Achanta, Raul Casas

Defensive Publications Series

Liveliness detection is increasingly important in video conferencing due to the advent of deepfakes and their use in conducting online fraud. Currently, liveliness detection technologies lag behind deepfake creation technologies. This submission proposes the use of the bidirectional nature of video conferencing to counter deepfakes more reliably and more efficiently than the state-of-the-art.

Pet-Aware Navigation Using Aggregated And Verified Point-Of-Interest Data, Kshitij Suresh Gaikar, Pushkar Mittal

Defensive Publications Series

Digital mapping applications may lack integrated data on pet policies for points of interest (POIs), which can lead users to consult fragmented and potentially inaccurate external sources when planning travel with pets. A system and method are described for integrating pet-compatibility data into a navigation application. The system can utilize a multi-source data engine to aggregate and harmonize information from sources such as crowdsourced user reports, third-party data providers, and artificial intelligence analysis of public web data. A harmonization framework may use confidence scoring and a rules engine to resolve data conflicts and verify the pet policies of POIs, such …

Network Of Ai-Agents For Multi-Step Iterative Debugging, Akshay Jajoo, Timo Koehler, Josh Halley, Ramana Kompella, Myungjin Lee

Defensive Publications Series

Proposed herein is a system to facilitate fault detection and root cause analysis (RCA) in distributed systems using a network of artificial intelligence (AI) agents that can work together to perform distributed debugging. The proposed system employs a hybrid architecture of centralized and localized AI agents, each co-located with network devices, enabling near-real-time, context-aware analysis of logs, traces, ring-buffers and other telemetry at the point of generation.

Blind Mate Paddle Fabric Shuffle Cassette, Anonymous

Defensive Publications Series

The present disclosure describes a blind-mate cassette interconnect system that enables high-speed, high-density connectivity between modular switching and fabric elements while maintaining full compatibility with standard OSFP (Octal Small Form-factor Pluggable) cages and connectors. The cassette integrates pre-terminated twinax or optical links within a floating housing that allows precise, tool-less alignment and automatic mating during module insertion, replacing up to sixteen discrete OSFP cables. This architecture combines the serviceability of blind-mate connectors with the flexibility of standard OSFP ecosystems (DAC, ACC, AOC, coherent, WDM, etc.), reducing installation time, eliminating cabling errors, and simplifying expansion from single-rack to multi-rack systems. By …

Cloud-Managed System For Distributing Audio Content To Point-Of-Sale Devices, Ramprasad S, Gulmohar Khan, Ankul Batra, Gajen Agrawal, Malay Singh, Saurabh Mimani, Ajay Prasad, Venu Annamraju, Sailendra Bathi

Defensive Publications Series

The utility for merchants of point-of-sale (POS) audio devices can be expanded beyond announcing transaction details. A cloud-managed system can facilitate the creation and distribution of custom audio content to these devices. For example, a user can input text through a client application, which can be sent to a cloud backend. The backend can use services, such as generative artificial intelligence for content refinement and a text-to-speech engine, to generate an audio file. The backend can then orchestrate the distribution of the audio file and an associated playback schedule to one or more endpoint audio devices. This system can expand …

Client-Side Logic Forcing Use Of Reliable Voice Codec For Satellite Voice Communications, Dang Tu Nguyen, Sooraj Sasindran, Youngtae Cha, Hwangoo Park, Bodam Nam, Karthick Musuvathi Janakan, Soumyadeep Paul

Defensive Publications Series

This disclosure describes a network-aware codec negotiation technique for satellite voice communications. Devices connected to satellite networks face issues such as high latency and low bandwidth that can cause modern High-Definition (HD) voice codecs to fail. This leads to call dropouts and/or unintelligible audio. The techniques described herein utilize client-side logic on the user equipment (UE) that forces the use of the Adaptive Multi-Rate Narrowband (AMR-NB) codec via the session description protocol (SDP) offer that is sent by the UE when a satellite connection is detected. This ensures a stable and reliable call, with priority given to intelligibility and reliability. …