Abstract: In a general aspect, a processor can include an instruction processing unit configured to execute a plurality of machine-readable instructions of a program in accordance with a control flow of the program and, in response to completing execution of respective instructions of the machine-readable instructions, produce respective instruction retirement information. The processor can also include a selectively enablable trace encoder that is configured, when enabled, to receive the respective instruction retirement information, and encode the instruction retirement information in an instruction trace corresponding with the control flow. The processor can also include a circular buffer that is configured to, when the trace encoder is enabled, receive and store the instruction trace, and, in response to a flush, communicate the stored instruction trace via a memory interface, where the memory interface is selectively enabled in response to the flush request.

Abstract: A system and method for mitigating sharp increases in power consumption associated with integrated circuits that may compute large workloads (e.g., computations). A processor may receive a signal associated with a workload. When the workload is determined to be above a workload threshold, a first signal may be sent to a compute engine to initiate a background computation for a ramp up period to increase the power consumption to a threshold. The workload may be performed following the ramp up period. The processor may send a second signal in response to the end of performing the workload. The second signal may initiate a background computation for a ramp down period, such that the power consumption in a ramp down period does not rapidly decrease to zero following a workloads computation.

Abstract: Methods, systems, or apparatuses for storage and access of micro-scaling numerics (MX) format data in memory. The approach may involve organizing MX data into tiled formats that match the read capabilities of consuming hardware blocks, interleaving scaling data with floating-point data, storing floating-point data in raster scan order, and storing scaling data in sub-tiles packed in multiples of a memory line size. A constant product of tile x-size and format size may be maintained across different MX format sizes, ensuring consistent memory utilization. This method may affect memory management for MX format data.

Abstract: Methods, systems, or computing devices are disclosed for enabling shared social media sessions that may allow multiple users to synchronously experience personalized content across different devices. The system may leverage artificial intelligence to create dynamic, constraint-free content streams that incorporate preferences from multiple users while maintaining individual privacy controls. Users may initiate shared sessions through messaging interfaces or standalone applications, with content sourced from multiple platforms and tailored to the collective preferences of session participants.

Abstract: Systems, apparatuses and methods provide technology that compresses first data based on a first compression scheme to generate second data, where the first data is associated with a first machine learning model. The technology stores the second data into a memory, adjusts a first entry of a lookup table to correspond to the first compression scheme based on the first data being compressed based on the first compression scheme, provide the second data from the memory to processing elements of a processing array during execution of the first machine learning model, and decompresses, at the processing array, the second data based on the lookup table to obtain the first data.

Abstract: Three-dimensional application specific integrated circuit (IC) architecture is described herein. In one aspect, an IC may include a first die including: a first semiconductor layer; a plurality of processing elements (PEs) located on the first semiconductor layer; and a first interface region of the first semiconductor layer, electrically coupled to the plurality of PEs and configured to communicate electrical signals with the plurality of PEs; a second die including: a second semiconductor layer; a plurality of IC elements located on the second semiconductor layer; and a second interface region of the second semiconductor layer, electrically coupled to the plurality of IC elements and configured to communicate electrical signals to the plurality of IC elements, where the first interface region and the second interface region are electrically coupled to each other and configured to transmit electrical signals between the plurality of PEs and the plurality of IC elements.

Abstract: Systems or methods disclosed herein may provide artificial intelligence-based aggregation of social media content across multiple platforms. In an example, a computing system may receive content queries and may retrieve relevant content from various social media platforms. The computing system may generate consolidated content feeds that include first-party content from platforms owned by a primary entity and third-party content from external messaging platforms. The consolidated feeds may be arranged into content slots with sponsored content intelligently placed between organic content slots based on user engagement metrics. The computing system may implement machine learning models to optimize content selection or arrangement based on user interaction patterns.

Abstract: A system and method for determining an identity verification of a user(s) are provided. The system may implement a machine learning model including training data pre-trained, trained in real-time, or periodically trained based on one or more faces of users. The system may determine whether a plurality of embeddings of a communication device associated with an image or a video of a face of at least one user includes a first similarity to a plurality of randomly generated embeddings generated by a network device. The system may receive, by the network device, a first plurality of similarity scores associated with the first similarity without receiving the plurality of embeddings of the communication device. The system may determine a second plurality of similarity scores denoting whether a second plurality of embeddings generated by the network device.

Abstract: According to examples, a system for using to processing of audio data using a plurality of distributed computer manner is described. The system may include a processor and a memory storing instructions. The processor may cause the system to receive audio data associated with a content item in an initial format, process the audio data to generate one or more audio segments for distributed processing, and decode the one or more audio segments from the audio data in the initial format to generate decoded audio data in a decoding format. The processor may then encode the decoded audio data in a decoding format to encoded audio data in an encoding format and trim a segment of the encoded audio data in the encoded format to generate a trimmed segment of audio data that may be utilized to enable continuous playback by a receiving device.

Abstract: In one embodiment, a method includes presenting a suggestion to a user of a head-mounted device by the head-mounted device via an assistant xbot during a dialog session between the user and the assistant xbot, wherein the suggestion is associated with a plurality of actions to be performed by an assistant system associated with the assistant xbot, accessing signals from inertial measurement unit (IMU) sensors of the head-mounted device by the head-mounted device during the dialog session, determining a head gesture performed by the user during the dialog session by an on-device head-gesture detection model and based only on the signals from the IMU sensors, and executing a first action from multiple actions by the assistant system executing on the head-mounted device, wherein the first action is selected based on the determined head gesture during the dialog session.

Abstract: In some embodiments, an integrated circuit (IC) chip package, includes a die; and die level diagnostic cell circuitry embedded in the die at a die level diagnostic testing location, wherein the die level diagnostic cell circuitry located at the die level diagnostic testing location incorporates a die level diagnostic stimulus to test a diagnostic trace of the die without requiring a decapsulation of the integrated circuit chip package. In some embodiments, the die level diagnostic cell circuitry is utilized to capture at least one of an inductance associated with the die level diagnostic cell circuitry stimulus driven diagnostic trace, a capacitance associated with the die level diagnostic cell circuitry stimulus driven diagnostic trace, an impedance associated with the die level diagnostic cell circuitry stimulus driven diagnostic trace, and a resistance associated with the die level diagnostic cell circuitry stimulus driven diagnostic trace.

Abstract: Systems, methods, and non-transitory computer readable media are configured to perform operations comprising providing, by a computing system, a currently playing content item in an interface presented through a screen to a group for collaborative online consumption; receiving, by the computing system, a command in a selected direction from a user in the group to select a new content item for the group to consume; and in response to the command, causing, by the computing system, scrolling of the interface such that the currently playing content item disappears and the new content item appears through the screen.

Abstract: A computer-implemented method for securely captioning video calls may include (i) detecting, by a messenger application executing on a client device, speech captured by a microphone of the client device and video of a speaker of the speech being captured by a camera of the client device, (ii) parsing, by the messenger application on the client device, the speech to create a transcript of the speech, and (iii) transmitting, by the messenger application on the client device, the transcript of the speech to an additional device for display to a user of the messenger application in combination with the video of the speaker. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: Techniques for enabling users logged-out of a social networking system to access content associated with the social networking system are described. For example, based at least in part on determining a user is unassociated with a social networking system, the social networking system may present content to the user. In some examples, the amount of content available to the user may be limited, in which the user may be prompted to create an account. In some examples, logged-out users may be prompted to pick topics associated with their interests, allowing the social networking system to curate content for logged-out users.

Abstract: Methods, systems, and storage media for synchronizing audio and video data are disclosed. Exemplary implementations may: receive at least one visual media file; receive audio data; determine a beat pattern associated with the audio data; generate a visual effect sequence comprising the at least one visual media file and at least one three-dimensional feature; and generate a resultant video file comprising the visual effect sequence and a transition sequence associated with the beat pattern.

Abstract: The disclosed systems and methods may include (1) receiving, from a broadcaster-user of a live video stream, a selection of (i) an interactive digital sticker and (ii) a position for the interactive digital sticker within an active stream frame of the live video stream, (2) in response to receiving the selections, rendering non-interactive image data, corresponding to a non-interactive component of the interactive digital sticker, on top of the active stream frame at the selected position, and (3) as part of transmitting the live video stream to a device of a viewer of the live video stream, (i) transmitting, to the viewer's device, the active stream frame with the non-interactive image data and (ii) delivering to the viewer's device, at the active stream frame, interactive digital sticker data corresponding to an interactive component of the interactive digital sticker. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: According to examples, a system for enabling communication between one or more disparate networks and network devices utilizing various connection technologies. The system may include a processor and a memory storing instructions. The processor, when executing the instructions, may cause the system to modify a security gateway to provide a metaverse security gateway by implementing a protocol associated with a 3rd Generation Partnership Project (3GPP) standard, modify a server to provide a metaverse server by implementing a protocol associated with the 3rd Generation Partnership Project (3GPP) standard, and utilize the metaverse security gateway and the metaverse server to facilitate a metaverse communication for a client device.

Abstract: A method for real time communication in a virtual environment. The method includes presenting, in the virtual environment, a virtual payment object associated with a first user and a virtual payment processing object associated with a second user. The virtual payment object is associated with a real payment object. The virtual payment processing object is associated with a real payment processing service. Responsive to detecting an interaction in the virtual environment between the virtual payment object and the virtual payment processing object, data describing the real payment object is transmitted to the real payment processing service via a real time communication protocol, causing the real payment processing service to process a real payment from the real payment object to an account associated with the second user in near real time.

Abstract: A disclosed computer-implemented method may include presenting, within a video creation interface, a list of video styles, each video style within the list of video styles representative of a trained model, the trained model trained to apply different aesthetic attributes to video content in accordance with a different video style. The method may also include receiving, via user input to the video creation interface, a selection of a particular video style from the list of video styles and, in response to receiving the selection, applying, to video content received via the video creation interface, a video editing technique corresponding to the particular video style. Various other methods, systems, and devices are also disclosed.

Abstract: Systems, apparatuses and methods provide technology that generates an interaction matrix from collaboration data, where the collaboration data represents user interactions with a plurality of items. The technology applies a matrix factorization process to the interaction matrix to generate user vectors and matrix item vectors, and executes a multi-task training process with a neural network. The multi-task training process includes processing the collaboration data with the neural network to generate latent item vectors, and training the neural network based on a comparison of the latent item vectors to the matrix item vectors.