Technical Disclosure Common
Not a member yet
9157 research outputs found
Sort by
Adaptive Fast Charging for Hearable Devices Based on State of Charge
No full textDisclosed is an adaptive, conditional fast-charging method for hearable devices. In the method, a high charge rate (C-rate) is applied only when the battery state of charge (SOC) is below a low threshold (e.g., 10-30%) and desired conditions (e.g., temperature, age) are met. Further, charging immediately reverts to a standard rate once an SOC target or a temperature limit is reached. This selective, conditional approach provides a rapid initial charge that significantly enhances the user experience, reduces the risk of lithium plating, and avoids the cost and size constraints associated with batteries designed for sustained high C-rates
Scan to Email Summary
No full textThis 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
Automated Execution of Manufacturing Flowcharts Using GenAI
No full textThis 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
A Forensic Logging Framework for LLMs
No full textLarge language models (LLMs) are increasingly embedded in enterprise, operational, and security-sensitive workflows. These include automated coding assistants, knowledge management systems, and multi-agent orchestration platforms where LLMs act as intermediaries between users and external tools. While significant research has been directed toward preventive safeguards such as input filtering, refusal mechanisms, and adversarial robustness, far less attention has been paid to forensic readiness. When incidents occur, ranging from malicious prompt injections to unauthorized tool usage or exfiltration of sensitive data, existing LLM implementations fail to provide investigators with reliable records for reconstruction, containment, or attribution.
To address this deficiency, a comprehensive system and method are proposed herein that enable forensic-grade logging for LLM-driven interactions. The system provides a detailed record of specific tampered-evident, chain-of-custody logs, recording artefacts such as prompts, responses, and tool invocations in a secure, cryptographically verifiable format. The design incorporates provenance metadata, including timestamps, model version identifiers, and execution context, to ensure that each recorded element can be reconstructed in its original sequence. In parallel, the system integrates privacy-preserving mechanisms that apply selective redaction and retention policies, balancing evidentiary completeness with compliance obligations under legal and regulatory frameworks. This dual emphasis on forensic trustworthiness and privacy protection enables organizations to not only prevent but also investigate and attribute incidents within LLM ecosystems
MeshTalk: A Decentralized, Identity-Agnostic Messaging Protocol over KTKN-OS
No full textDisclosed is MeshTalk, a decentralized, E2E encrypted messaging protocol utilizing KTKN-OS that eliminates reliance on central servers and phone numbers. The core innovation is the replacement of centralized identity providers with cryptographic identities managed by KTKN-OS . Users create self-sovereign identities and establish connections by sharing keys directly (e.g., QR codes). Each conversation (direct or group) is implemented as an independent KTKN-OS Space, with messages stored as encrypted CRDT documents within the Space\u27s database. This ensures E2E encryption and P2P replication directly between participants. The protocol is transport-agnostic, supporting integration with low-bandwidth mesh networks (e.g., LoRa)
The Neo-NLE Hybrid Architecture: Unifying Generative AI Synthesis and Deterministic Media Assembly
No full textDisclosed is the Neo-NLE (Next-Generation Non-Linear Editor) architecture, a hybrid system unifying traditional deterministic video assembly workflows with generative AI synthesis within a decentralized, local-first framework. The core innovation is the seamless integration of Assembly Workspaces (optimized for deterministic playback) and Generative Workspaces (for AI synthesis and code execution) within a single application. Both layers operate on a shared, decentralized data fabric (KTKN-OS/Automerge), enabling real-time interaction between generated assets and the assembly timeline. The architecture mandates a Local-First implementation, ensuring zero-latency interaction and data sovereignty, with P2P synchronization eliminating centralized cloud reliance
METHOD AND SYSTEM FOR AUTOMATED INCIDENT LIFECYCLE MANAGEMENT WITH SELF-LEARNING AGENTS
No full textThe 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 integration with external tools and incorporates a self-learning knowledge base that evolves based on historical resolution data, enhancing overall efficiency in incident management processes
LLM-augmented Dynamic Selective Blocking of Service Calls for Application Control
No full textSoftware 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 query-of-death issues, thereby minimizing service disruption and user impact
Model-Based Vision System for Evaluation and Improvement of Gaits in Legged Locomotion
No full textThis disclosure relates to the evaluation and improvement of gaits in legged locomotion of humans, animals, and robots by superimposing a biomechanical figure selected from a spectrum of models onto the recorded motion of a test subject. For gait evaluation, a model is selected from a database to closely match the observed gait and existing performance parameters are obtained. For gait improvement, an incrementally superior model is selected from the database to provide enhanced performance goals. This Model Vision Evaluation (MoVE) system thus includes spectrums of biomechanical models over a range of inferior to superior levels of skill and performance for evaluating and improving all types of gaits in legged locomotion
End-to-End Differentiable GNSS Positioning with Neural Network-Derived Weights
No full textThere is a fundamental limitation in accuracy and robustness of traditional Global Navigation Satellite System (GNSS) positioning techniques that rely on the Weighted Least Squares (WLS) algorithm. Traditional techniques use static heuristic functions for assigning weights to pseudorange measurements in the WLS algorithm. This disclosure addresses these limitations by replacing the static heuristic functions that are used for assigning weights to pseudorange measurements in the WLS algorithm with a trained Neural Network (NN). This NN acts as a dynamic Satellite Weight Prediction Model. Per the techniques described herein, the entire positioning pipeline—from satellite features to final positional error—is treated as a single, differentiable system. This allows the parameters of the NN to be optimized directly to minimize the final positional error (loss function of calculated position vs. ground truth position). This is a key advantage over conventional techniques