Abstract: An apparatus includes a host interface; a network interface; hardware storage to store a flow table; and programmable circuitry comprising processors to implement network interface functionality and to: implement a hash table and an age context table, wherein the hash table and the age context table are to reference flow rules maintained in the flow table; process a synchronization packet for a flow by adding a flow rule for the flow to the flow table, adding a hash entry corresponding to the flow rule to the hash table, and adding an age context entry for the flow to the age context table; and process subsequent packets for the flow by performing a first lookup at the hash table to access the flow rule at the flow table and by performing a second lookup at the age context table to apply aging rules to the flow rule in the flow table.

Abstract: Techniques described herein address the above challenges that arise when using host executed software to manage vector databases by providing a vector database accelerator and shard management offload logic that is implemented within hardware and by software executed on device processors and programmable data planes of a programmable network interface device. In one embodiment, a programmable network interface device includes infrastructure management circuitry configured to facilitate data access for a neural network inference engine having a distributed data model via dynamic management of a node associated with the neural network inference engine, the node including a database shard of a vector database.

Abstract: An exemplary method of generating content for a user of a social networking service includes, in an application of a social networking service, receiving a classification model from a server system associated with the social networking service. The method further includes generating a collage for a user. The collage includes a set of content items stored on the client device. The method also includes, presenting the collage to the user if the collage satisfies a criterion included in the classification model. The method also includes tracking and providing user-interaction data for the collage to the server system, and receiving an updated classification model from the server system in accordance with the user-interaction data. The method also includes generating and presenting a new collage that includes a new set of content items stored on the client device, if the new collage satisfies the criterion included in the updated classification model.

Abstract: An exemplary method of generating content for a user of a social networking service includes, in an application of a social networking service, receiving a classification model from a server system associated with the social networking service. The method further includes generating a collage for a user. The collage includes a set of content items stored on the client device. The method also includes, presenting the collage to the user if the collage satisfies a criterion included in the classification model. The method also includes tracking and providing user-interaction data for the collage to the server system, and receiving an updated classification model from the server system in accordance with the user-interaction data. The method also includes generating and presenting a new collage that includes a new set of content items stored on the client device, if the new collage satisfies the criterion included in the updated classification model.

Abstract: An exemplary method of generating content for a user of a social networking service includes, in an application of a social networking service, receiving a classification model from a server system associated with the social networking service. The method further includes detecting content items stored on the client device independently of the application. The method further includes generating content for a user of the social networking service that includes a set of content items from the detected content items, based at least in part on metadata for the set of content items. The metadata is stored at the client device. The method further includes, in accordance with a determination that the generated content for the user satisfies a criterion included in the classification model, presenting the content to the user in the application.

Abstract: An online system customizes video conversations between users of the online system by providing graphics that are likely to interest the users. The online system may present composite views overlaying selected graphics on a video stream, which may be part of an augmented reality (AR) environment. The graphics include, for example, background graphics, masks, props, visual or particular effects, frames or borders, etc. The online system may use a machine learning model to predict whether a user is likely to select or interact with a particular graphic. The online system can also filter or rank graphics according to user affinities or user connections on the online system. Users may be encouraged to interact with the graphics that are customized to their specific interests, which can promote an engaging video conversation or AR experience.

Abstract: An online system customizes video conversations between users of the online system by providing graphics that are likely to interest the users. The online system may present composite views overlaying selected graphics on a video stream, which may be part of an augmented reality (AR) environment. The graphics include, for example, background graphics, masks, props, visual or particular effects, frames or borders, etc. The online system may use a machine learning model to predict whether a user is likely to select or interact with a particular graphic. The online system can also filter or rank graphics according to user affinities or user connections on the online system. Users may be encouraged to interact with the graphics that are customized to their specific interests, which can promote an engaging video conversation or AR experience.

Abstract: An online system customizes video conversations between users of the online system by providing graphics that are likely to interest the users. The online system may present composite views overlaying selected graphics on a video stream, which may be part of an augmented reality (AR) environment. The graphics include, for example, background graphics, masks, props, visual or particular effects, frames or borders, etc. The online system may use a machine learning model to predict whether a user is likely to select or interact with a particular graphic. The online system can also filter or rank graphics according to user affinities or user connections on the online system. Users may be encouraged to interact with the graphics that are customized to their specific interests, which can promote an engaging video conversation or AR experience.

Abstract: An exemplary method of generating content for a user of a social networking service includes, in an application of a social networking service, receiving a classification model from a server system associated with the social networking service. The method further includes detecting content items stored on the client device independently of the application. The method further includes generating content for a user of the social networking service that includes a set of content items from the detected content items, based at least in part on metadata for the set of content items. The metadata is stored at the client device. The method further includes, in accordance with a determination that the generated content for the user satisfies a criterion included in the classification model, presenting the content to the user in the application.

Abstract: Methods and systems for configuring the display of items in a network-based merchandising environment are provided. In one embodiment, the present invention identifies items within a catalog of items, computes a browse relevance score for each item, and configures the display of a Web page such that items with the highest browse relevance score are prominently displayed. In at least one embodiment, browse data, such as category fit data, popularity data, newness data, etc., are used to compute the browse relevance score for each item.

Abstract: Methods and systems for configuring the display of items in a network-based merchandising environment are provided. In one embodiment, the present invention identifies items within a catalog of items, computes a browse relevance score for each item, and configures the display of a Web page such that items with the highest browse relevance score are prominently displayed. In at least one embodiment, browse data, such as category fit data, popularity data, newness data, etc., are used to compute the browse relevance score for each item.