Abstract: A system and method for generating task-specific text by processing multimodal inputs using machine-learning models is provided. The method may include accessing first sets of tokens associated with a desired task and one or more modalities associated with a context of the desired task. The method may further include determining a second set of tokens for each of the one or more modalities using a classifier network associated with the modality. The method may further include generating a number of embedding vectors by mapping the first sets of tokens and the second set of tokens associated with each of the one or more modalities to an embedding space. The method may further include producing a sequence of words addressing the desired task by processing the number of embedding vectors with an encoder-decoder network.

Abstract: Systems, methods, and non-transitory computer-readable media are disclosed for selectively rendering augmented reality content based on predictions regarding a user's ability to visually process the augmented reality content. For instance, the disclosed systems can identify eye tracking information for a user at an initial time. Moreover, the disclosed systems can predict a change in an ability of the user to visually process an augmented reality element at a future time based on the eye tracking information. Additionally, the disclosed systems can selectively render the augmented reality element at the future time based on the predicted change in the ability of the user to visually process the augmented reality element.

Abstract: In one embodiment, a method includes accessing a first document, accessing a plurality of second documents, calculating a relevance score for each of the plurality of second documents indicating a degree of relevance of the second document to the first document using an encoder of a machine-learning model, selecting a subset of the second documents based on their corresponding relevance scores, generating a target document by using the machine-learning model to process the subset of second documents and their corresponding relevance scores, and updating parameters of the machine-learning model based on a comparison between the first document and the generated target document.

Abstract: The disclosed systems for suspending cable (e.g., fiber optic cable) from an overhead powerline may include a payload subsystem for housing and dispensing a cable along an overhead powerline, a rotation subsystem for winging the cable from the payload subsystem around the powerline, an extension subsystem for moving the payload subsystem to avoid obstacles, an obstacle detection subsystem for automatically detecting obstacles encountered along the powerline, a drive subsystem for driving the system along the powerline, and at least one processor for controlling the payload subsystem, rotation subsystem, extension subsystem, obstacle detection subsystem, and drive subsystem in a manner that avoids obstacles as the system moves along the powerline. Various other related systems, devices, components, and methods are also disclosed.

Abstract: Methods, systems, and storage media for augmenting a video are disclosed. Exemplary implementations may: receive a selection of an effect; receive user-generated content comprising video data and audio data; detect a characteristic of the audio data comprising at least a volume and/or a pitch of the audio data during a period of time; determine a series of numeric values based on the characteristic of the audio data during the period of time, individual numeric values of the series of numeric values being correlated with an amplitude of the volume and/or pitch at a discrete point within the period of time; and augment at least one of the video data and/or the audio data to include the effect based on the series of numeric values at discrete points in time within the period of time.

Abstract: Methods, systems, and apparatuses for disguising or otherwise hiding big data access patterns and frequencies may include a replica being directly obtained from a key and a replica index, based on applying a pseudorandom permutation or pseudorandom function. Data object access frequencies may be quantized so that the ratios of quantized data object access frequencies are rational numbers of integers, which may allow for a uniform distribution of data object accesses. Fake accesses may be implemented by simulating an access schedule produced by a cryptographic primitive and by performing a greedy mapping between the real data object accesses and simulated data object accesses.

Abstract: The present disclosure is directed toward systems and methods for optimizing video conferences. For instance, systems and methods described herein optimize both the transmission and display of one or more video conference data streams. Systems and methods described herein optimize the transmission and display of one or more video conference data streams by identifying a context associated with the one or more video conference data streams and optimizing the one or more video conference data streams based on the identified context.

Abstract: According to examples, a system for implementing image modification functions via use of variable scanning orders is described. The system may include a processor and a memory storing instructions. The processor, when executing the instructions, may cause the system to partition an image into a plurality of image blocks, identify one or more image blocks of the plurality of image blocks associated with a region of interest (ROI), and scan the one or more image blocks in an image modification order. The processor, when executing the instructions, may then arrange the one or more image blocks according to the image modification order to form a modified image including the region of interest (ROI) and crop the region of interest (ROI) in the modified image to form a new image.

Abstract: A semiconductor design that uses high refractive index material between low refractive index material. This structure may act as an optical waveguide.

Abstract: A system for facilitating RDMA transmit flow scheduling and pacing is disclosed. The system may determine a set of groups. The groups may be associated with a range of transmission rates. The system may determine a transmission rate of a queue pair. The system may assign the QP to a first group of the set of groups. The transmission rate of the QP may be within the range of transmission rates of the first group. The system may determine an available QP of at least one group from the set of groups. The system may schedule transmission, by an arbiter, of the available QP. The system may transmit, by the arbiter, a message associated with the available QP.

Abstract: Systems, methods, client devices, and non-transitory computer-readable media are disclosed for utilizing video data and video processing data to enable shared augmented reality scenes having video textures depicting participants of video calls as augmented reality (AR) effects during the video calls. For instance, the disclosed systems can establish a video call between client devices that include streaming channels (e.g., a video and audio data channel). In one or more implementations, the disclosed systems enable the client devices to transmit video processing data and video data of a participant through the streaming channel during a video call. Indeed, in one or more embodiments, the disclosed systems cause the client devices to utilize video data streams and video processing data to render videos as video textures within AR effects in a shared AR scene (or AR space) of the video call (e.g., to depict participants within the AR scene).

Abstract: In one embodiment, a method includes, determining that a bmm operation between a first activation tensor and a second activation tensor needs to be performed, collecting the second activation tensor in two blocks from activation buffers of N tensor processor units, splitting each of the two blocks of the second activation tensor into an MSB tile and an LSB tile, loading the second activation tensor to weight buffers of the N tensor processor units by filling a first entry of each weight buffer of each of the N tensor processor units with contents of the MSB tiles of the two blocks and filling a second entry of the weight buffer with contents of the LSB tiles of the two blocks, and generating a bmm result using the first activation tensor distributed in the activation buffers and the second activation tensor in the weight buffers.

Abstract: Systems and methods for social media video streaming may include (1) presenting, within a short-form video being streamed as part of a short-form video feed, a selectable theme element, indicating a theme associated with the short-form video, (2) receiving user input selecting the selectable theme element, and (3) in response to receiving the user input, presenting a digest interface of short-form videos, relating to the theme, which includes (i) sets of tiles, in which each tile corresponds to a different short-form video, organized into subthemes, and (ii) a digital social engagement element, which provides an entry point for digital social engagement relating to the theme.

Abstract: In one embodiment, a method includes receiving, from a client system associated with a user, a first user request that includes a reference to a target object and one or more of an attribute or a relationship of the target object. Visual data including one or more images portraying the target object may then be accessed, and the reference may be resolved to the target object portrayed in the one or more images. Object information of the target object that corresponds to the referenced attribute or relationship of the first user request may be determined based on a visual analysis of the one or more images. Finally, responsive to receiving the first user request, the object information of the target object may be stored in a multimodal dialog state.

Abstract: In one embodiment, a method includes, by one or more computing systems, receiving, by an assistant xbot associated with the one or more computing systems, a user request for a content digest, retrieving one or more content objects corresponding to the user request, generating one or more slides for the one or more retrieved content objects, respectively, and providing, by the assistant xbot, instructions for presenting the content digest responsive to the request from the first user, wherein the content digest comprises the one or more slides, and wherein one or more of the slides of the content digest are removed from the content digest after a predetermined time period.

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods for generating an association between a graphic element and text for a messaging thread and providing an overlay including the graphic element based on providing a message to the messaging thread including the text. In particular, in one or more embodiments, the disclosed systems receive user selection of the text and the graphic element and provide the association between the text and the element to various client devices associated with the messaging thread. Accordingly, the disclosed systems provide instructions for each client device associated with the messaging thread to generate and provide an overlay for the messaging thread.

Abstract: A user provides video data to an online system for presentation to additional users of the online system. Additionally, the user provides information to the online system identifying positions in the information capable of presenting supplemental content items, which may be sponsored, and may provide information describing a context of the video data at the various positions capable of presenting the supplemental content items. For example, the information identifies one or more times within the video data at which an advertisement is capable of being presented via the video data. Based on the information provided by the user, the online system selects supplemental content items for presentation via the video data at the identified positions and presents the selected supplemental content items to additional users viewing the video data at the identified positions.

Abstract: In some embodiments, a computer-implemented method includes obtaining a without-eyeglasses face scan of a subject, the without-eyeglasses face scan being a three-dimensional (3D) model of a face of the subject without eyeglasses; obtaining a with-eyeglasses face scan of the subject, the with-eyeglasses face scan being a 3D model of the subject with eyeglasses; and using the without-eyeglasses face scan and the with-eyeglasses face scan to place a 3D eyeglasses model on a face model of the subject. In some embodiments of the computer-implemented method, the 3D eyeglasses model is placed on the face model of the subject using frame placement information generated using the without-eyeglasses face scan and the with-eyeglasses face scan.

Abstract: The disclosed device may include a radio frequency (RF) component and a thermal management component establishing a thermal path with the RF component. The device may also include an RF filter electrically that is coupled to the RF component and arranged away from the thermal path. The device may further include a chassis for housing the RF component. The thermal management component and the RF filter may be positioned near an exterior of the chassis. Various other devices, apparatuses, and systems are also disclosed.