Facebook, Inc.

Microsoft Technology Licensing, LLC

Meta Platforms, Inc.

Mylio, LLC

Mohsen M Agsen

Jason Curtis Jenks

Jeremy David Fein

Philippe Vincent Ajoux

Michael Robert Hanson

Joshua Scott Evenson

Emmanouil Koukoumidis

Xiaohu Liu

Fuchun Peng

Benoit F Dumoulin

Patrick James Helland

Rajen Subba

Rodney Thomas Limprecht

Craig Symonds

Jiedan Zhu

Yasser Baseer Asmi

Momin Mahmoud Al-Ghosien

Jean-Pierre Duplessis

Deon Brewis

Scott Robert Marison

Jean Pierre Duplessis

David De Regt

In one embodiment, a method includes receiving a first user input from a first user, wherein the first user input comprises a partial request, presenting one or more suggested intent auto-completions corresponding to the partial request, receiving a selection by the first user of a first suggested intent auto-completion of the suggested intent auto-completions and a second user input, presenting one or more suggested slot auto-completions corresponding to one or more candidate slot-hypotheses corresponding to the second user input, respectively, wherein each of the candidate slot-hypotheses comprise a slot-suggestion, and wherein each suggested slot auto-completion comprises the second user input and the corresponding candidate slot-hypothesis, receiving a selection by the first user of a first suggested slot auto-completion of the suggested slot auto-completions, and presenting execution results of one or more tasks corresponding to the first suggested intent auto-completion and the first suggested slot auto-completion.

Contextual auto-completion for assistant systems

In one embodiment, a method includes, by a client system, receiving, at the client system, a first user input, processing by the client system, the first user input to provide an initial response by identifying one or more entities referenced by the first user input and providing, by the client system, the initial response, where the initial response includes a conversational filler referencing at least one of the one or more identified entities, processing the first user input to provide a complete response by identifying, by the client system, one or more intents and one or more slots associated with the first user input based on a semantic analysis by a natural-language understanding module, and providing, by the client system, the complete response subsequent to the initial response, where the complete response is based on the one or more intents and the one or more slots.

Predictive injection of conversation fillers for assistant systems

Techniques for directive-based messaging synchronization are described. In one embodiment, an apparatus may comprise a local network component operative to receive a directive package at a messaging client on a client device; and a local database synchronization component operative to execute the directive package with a messaging-sync virtual machine to modify a local messaging database of the messaging client; and refresh a user interface component of the messaging client in response to modifying the local messaging database of the messaging client. Other embodiments are described and claimed.

Techniques for directive-based messaging synchronization

In one embodiment, a method includes, by one or more computing systems, receiving, from a client system associated with a first user, a first user input from the first user, identifying one or more entities referenced by the first user input, determining a classification of the first user input based on a machine-learning classifier model, generating several candidate conversational fillers based on the classification of the first user input and the one or more identified entities, wherein each candidate conversational filler references at least one of the one or more identified entities, ranking the candidate conversational fillers based on a relevancy of the candidate conversational filler to the first user input and a decay model hysteresis, and sending instructions for presenting a top-ranked candidate conversational filler as an initial response to the first user.

In one embodiment, a method includes, by one or more computing systems, receiving, from a client system associated with a user, an initial portion of a user input, wherein the initial portion comprises a partial request, and wherein the initial portion is received while the user is continuing to provide further input, generating, responsive to receiving the initial portion of the user input, one or more speculative queries based on the partial request and a machine-learning predictive model, wherein each speculative query is a predicted complete request based on the partial request, calculating a confidence score for each speculative query based on the predictive model, ranking the one or more speculative queries based on their respective confidence scores and associated costs, executing one or more of the speculative queries based on their ranks, and caching one or more results of the executed one or more speculative queries.

Context-based utterance prediction for assistant systems

Techniques for a database-driven messaging user interface are described. In one embodiment, an apparatus may comprise a database access component operative to determine that a local messaging database of a messaging client on a client device has been updated; and examine the local messaging database to determine a set of database queries to rerun for a plurality of local user interface components; and the local user interface components operative to refresh based on running the set of database queries against the local messaging database. Other embodiments are described and claimed.

Techniques for a database-driven messaging user interface

Techniques for backend-specific cursor tracking are described. In one embodiment, an apparatus may comprise a local database synchronization component operative to initiate a client update at a messaging client on a client device, the client update associated with a specific backend service for a messaging system; retrieve an opaque backend-specific update cursor for the specific backend service; and store an updated opaque backend-specific update cursor for the messaging client; and a local network component operative to send the opaque backend-specific update cursor to the messaging system in association with a client update request; and receive an update package at the messaging client on the client device, the update package comprising the updated opaque backend-specific update cursor. Other embodiments are described and claimed.

Techniques for backend-specific cursor tracking

In one embodiment, a method includes receiving a user input including a partial request from a client system of a first user, analyzing the user input to generate one or more candidate hypotheses based on a personalized language model where each of the candidate hypotheses includes one or more of an intent-suggestion or a slot-suggestion, sending instructions for presenting one or more suggested auto-completions corresponding to one or more of the candidate hypotheses, respectively, to the client system, where each suggested auto-completion comprises the partial request and the corresponding candidate hypothesis, receiving an indication of a selection by the first user of a first suggested auto-completion of the suggested auto-completions from the client system, and executing one or more tasks based on the first suggested auto-completion selected by the first user via one or more agents.

Techniques for in-place directive execution are described. In one embodiment, an apparatus may comprise a local network component operative to receive a directive package at a messaging client on a client device, the directive package formatted according to a serialized directive schema; and store the directive package in a memory of the client device in the serialized directive schema; and a local database synchronization component operative to execute the directive package using in-place access of the directive package in the memory of the client device. Other embodiments are described and claimed.

Techniques for in-place directive execution

Systems and methods for providing data replication and/or synchronization across multiple processor-based devices. Data objects (e.g., image files, video files, audio files) may exist in a plurality of forms (e.g., original, preview, thumbnail). The systems and methods utilize object device links (ODLs) and needs/has flags which are replicated to multiple processor-based devices to provide partial or full object replication between multiple processor-based devices. Some implementations provide complete replica backups of objects. Some implementations provide temporary protection of original forms of objects, for example, when a user is traveling. Some implementations provide for relaying and shuttling of objects across multiple devices via one or more networks and/or via peer-to-peer wired or wireless communications also allowing for offsite and remote location backup of objects. Some implementations provide support for container and individual object replication policies with inheritance. Some implementations can provide load balancing storage of objects. Some implementations can provide for efficient space management of objects on storage-constrained devices.

Object replication using object device links and flags

Ink lag compensation techniques are described. In implementations, an input is recognized by a computing device as an ink stroke to be rendered on a display device of the computing device. An extension is predicted, by the computing device, to be applied to the ink stroke that is likely to be subsequently input in conjunction with the ink stroke. The ink stroke and the extension are output for concurrent display on the display device of the computing device.

Ink lag compensation techniques

Exception-based component models can be generated from non-exception-based models. For example, a non-exception-based interface contract (such as a non-exception-based interface contract in a header file) for a consumable software component can be parsed. Additionally, a wrapper for the consumable software component can be generated from the non-exception-based interface contract. The wrapper can have an exception-based interface contract implementation, and the exception-based interface contract implementation can be a mirror equivalent of the non-exception-based interface contract. If the non-exception-based interface contract hints at one or more arguments as being out argument(s) for a function in the consumable software component, then predefined heuristics can be used to select one of those argument(s) to be used by the exception-based interface contract implementation as a return value for the function.

Generating exception-based component models from non-exception-based models

A component-based framework includes system services and component integration interfaces that provide a run-time environment for component-based server application in which server application components have control over the duration of their state independent of client action. The framework provides functions that a server application component calls to indicate its work for a client is complete, and its state is to be destroyed. On the component's return from the client's method call, the framework destroys the component's state thus releasing any server computer resources held by the component. This allows the server application components to be programmed so as to minimize the duration that their state remains in memory consuming server computer resources, and which enhances scalability of the server application.

Growing community of inventors

Honolulu, HI, United States of America

Mohsen M Agsen