Abstract: Methods, systems, and apparatuses for risk analysis of web pages using a machine learning model are described herein. A computing device may receive a risk detection machine learning model trained to receive input corresponding to a web page and output an indication of risk associated with the web page. The computing device may execute a web browser application and collect user activity data by monitoring user activity associated with the web browser application. The computing device may access, via the web browser application, a first web page, and collect page data associated with the first web page. The computing device may calculate a risk level of the first web page. The risk level may be calculated by processing, using the risk detection machine learning model, both the user activity data and the page data. A security recommendation may be output based on the risk level.

Abstract: A valve assembly for integration within a wellbore string disposed within a hydrocarbon-containing reservoir is provided. The valve assembly includes a valve housing having a plurality of frac ports for establishing fluid communication between a central passage and the reservoir. The valve assembly includes a bottom sleeve operatively mounted within the valve housing configured to selectively open the frac ports, and a top sleeve operatively mounted within the valve housing slidable between (i) a first position defining a first fluid pathway whereby fluid is flowable down into the central passage and into the reservoir via the frac ports, and (ii) a tracing position defining a second fluid pathway whereby fluid is flowable from the reservoir into an annulus defined between the top sleeve and the housing. The valve assembly also has a tracer compartment defined within the annulus forming part of the second fluid pathway and accommodating a tracer material.

Abstract: The present disclosure is directed to systems and methods that include and/or leverage one or more machine-learned language models that generate intermediate textual analysis (e.g., including usage of structural tools such as APIs) in service of contextual text generation. For example, a computing system can obtain a contextual text string that includes one or more contextual text tokens. The computing system can process the contextual text string with the machine-learned language model to generate one or more intermediate text strings that include one or more intermediate text tokens. The computing system can process the one or more intermediate text strings with the machine-learned language model to generate an output text string comprising one or more output text tokens. The one or more intermediate text strings can include textual analysis of the contextual text string that supports the output text string.

Abstract: A torque security system for a vehicle is provided. The system receives a signal from a sensor coupled to a motor shaft of an electric motor and determines an acceleration of the electric motor, based on the signal from the sensor that indicates an amount of rotation of the motor shaft. The system determines an internal torque between the motor shaft and an input gear coupled to the motor shaft, based on the acceleration of the electric motor and an inertia of the electric motor and a gearbox. The powertrain of the vehicle comprises the gearbox and the electric motor, and the input gear couples the electric motor to the gearbox. The system determines whether the internal torque exceeds a threshold torque, and in response to determining that the internal torque exceeds the threshold torque, the system reduces power output to the electric motor. The system also diagnoses the health of the gearbox.

Abstract: A system for managing vehicle body and wheel motion control with a dual clutch differential includes sensors and actuators disposed on the vehicle, the sensors measuring real-time static and dynamic data and the actuators altering static and dynamic behavior of the motor vehicle. A control module executes program code portions stored in memory. The program code portions receive the real-time static and dynamic data; selectively prioritize torque output from a prime mover of the vehicle through the differential to driven wheels of the vehicle to control a body and the driven wheels; model and estimate clutch torque for each clutch of the dual clutch differential; model and estimate a joint clutch torque, a tire force, and corner torque; and generate a torque output for each clutch of the dual clutch differential that is selected to maintain one or more of body control, wheel control, and stability of the motor vehicle.

Abstract: A method for controlling a vehicle includes determining, via a processor disposed in communication with a sensory device disposed in an interior surface of a cargo bin of a delivery vehicle, that a first package is loaded in the cargo bin, causing to scan, via the processor, a bar code disposed on an exterior surface of the first package, causing to measure, via a RADAR sensory device disposed in the cargo bin, dimensional information associated with the first package, determining, based on the dimensional information, that the cargo bin is loaded at full capacity, causing to illuminate an output device indicative that the cargo bin is loaded at full capacity, determining, via the processor, a cargo bin door closed status and generating, via the processor, an output indicating whether the vehicle is ready for delivery.

Abstract: An incentive management system may receive a request for registration of a software application from a software developer and assign at least an API key and an API secret to the software application. The system may further provide, to the software developer, integration instructions including at least the API key and the API secret and provide a developer platform to the software developer, the developer platform configured to receive program information including at one or more actions that may be performed by a user of the software application. The system may monitor messages from one or more application servers, such as via webhooks embedded in the software application, regarding performance of one or more actions identified as actionable by the software developer. The system may periodically determine a cumulative incentive associated with each of the users of the software application and coordinate provision of the incentives.

Abstract: In one embodiment, a method includes accessing multiple 3D photos to be concurrently displayed through multiple frames positioned in a virtual space, each of the of 3D photos having an optimal viewing point in the virtual space and determining a reference point based on a head pose of a viewer within the virtual space. The method may further include adjusting each 3D photo by rotating the 3D photo so that the optimal viewing point of the 3D photo points at the reference point, translating the rotated 3D photo toward the reference point, and non-uniformly scaling the rotated and translated 3D photo based on a scaling factor determined using the reference point and a position of the frame through which the 3D photo is to be viewed. The method may further include rendering an image comprising the adjusted multiple 3D photos as seen through the multiple frames.

Abstract: Roofing shingles are disclosed that are capable of self-adhering to a roof deck or underlayment and/or other roofing shingles and that require few or no mechanical fasteners to remain attached to the roof. By appropriate positioning of sealant lines on the shingle, direct adhesion between the shingle and the roof deck or underlayment and/or other roofing shingles can be achieved. If the shingle is laminated, the layers may be mechanically attached with indentations in the common bond area. The nail zone of the shingle may be visually indicated with fines and/or one or more paint lines. A roofing system comprising a plurality of courses of the shingles is also disclosed.

Abstract: This application relates to apparatus and methods for natural language understanding in conversational systems using machine learning processes. In some examples, a computing device receives a request that identifies textual data. The computing device applies a natural language model to the textual data to generate first embeddings. In some examples, the natural language model is trained on retail data, such as item descriptions and chat session data. The computing device also applies a dependency based model to the textual data to generate second embeddings. Further, the computing device concatenates the first and second embeddings, and applies an intent and entity classifier to the concatenated embeddings to determine entities, and an intent, for the request. The computing device may generate a response to the request based on the determined intent and entities.

Abstract: A spindle nut retainer is used with a spindle with a flat portion on its threaded end. The flat portion interrupts the threads on the threaded end. A washer with a mating flat portion is slidably mounted over the threaded end and a nut is removably threaded to the threaded end of the spindle against the washer to set the preload on bearings. The nut retainer overlays a plurality of the corners of the nut and engages the threaded portion of the spindle with tangs. A majority of the tangs are flexed or bent to remain in biased contact with the threads on the spindle. Some of the tangs are located over the flat portion and remain unbent. The tangs resist rotation of the nut by engagement with the flat portion where it meets the threads.