Abstract: This document relates to natural language processing using a framework such as a neural network. One example method involves obtaining a first document and a second document and propagating attention from the first document to the second document. The example method also involves producing contextualized semantic representations of individual words in the second document based at least on the propagating. The contextualized semantic representations can provide a basis for performing one or more natural language processing operations.

Abstract: A computer-implemented technique is described herein for assisting a user in advancing a task objective. The technique uses a suggestion-generating system (SGS) to provide one or more suggestions to a user in response to at least a last-submitted query provided by the user. The SGS may correspond to a classification-type or generative-type neural network. The SGS uses a machine-trained model that is trained using a multi-task training framework based on plural groups of training examples, which, in turn, are produced using different respective example-generating methods. One such example-generating method constructs a training example from queries in a search session. It operates by identifying the task-related intent the queries, and then identifying at least one sequence of queries in the search session that exhibits a coherent task-related intent. A training example is constructed based on queries in such a sequence.

Abstract: This document relates to natural language processing using a framework such as a neural network. One example method involves obtaining a first document and a second document and propagating attention from the first document to the second document. The example method also involves producing contextualized semantic representations of individual words in the second document based at least on the propagating. The contextualized semantic representations can provide a basis for performing one or more natural language processing operations.

Abstract: A computer-implemented technique is described herein for assisting a user in advancing a task objective. The technique uses a suggestion-generating system (SGS) to provide one or more suggestions to a user in response to at least a last-submitted query provided by the user. The SGS may correspond to a classification-type or generative-type neural network. The SGS uses a machine-trained model that is trained using a multi-task training framework based on plural groups of training examples, which, in turn, are produced using different respective example-generating methods. One such example-generating method constructs a training example from queries in a search session. It operates by identifying the task-related intent the queries, and then identifying at least one sequence of queries in the search session that exhibits a coherent task-related intent. A training example is constructed based on queries in such a sequence.

Abstract: A computer-implemented technique is described herein for assisting a user in advancing a task objective. The technique uses a suggestion-generating system (SGS) to provide one or more suggestions to a user in response to at least a last-submitted query provided by the user. The SGS may correspond to a classification-type or generative-type neural network. The SGS uses a machine-trained model that is trained using a multi-task training framework based on plural groups of training examples, which, in turn, are produced using different respective example-generating methods. One such example-generating method constructs a training example from queries in a search session. It operates by identifying the task-related intent the queries, and then identifying at least one sequence of queries in the search session that exhibits a coherent task-related intent. A training example is constructed based on queries in such a sequence.

Abstract: The present concepts relate to a differentiable user-item co-clustering (“DUICC”) model for recommendation and co-clustering. Users' interaction with items (e.g., content) may be centered around information co-clusters—groups of items and users that exhibit common consumption behavior. The DUICC model may learn fine-grained co-cluster structures of items and users based on their interaction data. The DUICC model can then leverage the learned latent co-cluster structures to calculate preference stores of the items for a user. The top scoring items may be presented to the user as recommendations.

Abstract: Disclosed are techniques for extracting, identifying, and consuming imprecise temporal elements (“ITEs”). A user input may be received from a client device. A prediction may be generated of one or more time intervals to which the user input refers based upon an ITE model. The user input may be associated with the prediction, and provided to the client device.

Abstract: This document relates to natural language processing using a framework such as a neural network. One example method involves obtaining a first document and a second document and propagating attention from the first document to the second document. The example method also involves producing contextualized semantic representations of individual words in the second document based at least on the propagating. The contextualized semantic representations can provide a basis for performing one or more natural language processing operations.

Abstract: This document relates to natural language processing using a framework such as a neural network. One example method involves obtaining a first document and a second document and propagating attention from the first document to the second document. The example method also involves producing contextualized semantic representations of individual words in the second document based at least on the propagating. The contextualized semantic representations can provide a basis for performing one or more natural language processing operations.

Abstract: The present concepts relate to a differentiable user-item co-clustering (“DUICC”) model for recommendation and co-clustering. Users' interaction with items (e.g., content) may be centered around information co-clusters—groups of items and users that exhibit common consumption behavior. The DUICC model may learn fine-grained co-cluster structures of items and users based on their interaction data. The DUICC model can then leverage the learned latent co-cluster structures to calculate preference stores of the items for a user. The top scoring items may be presented to the user as recommendations.

Abstract: A computer-implemented technique is described herein for assisting a user in advancing a task objective. The technique uses a suggestion-generating system (SGS) to provide one or more suggestions to a user in response to at least a last-submitted query provided by the user. The SGS may correspond to a classification-type or generative-type neural network. The SGS uses a machine-trained model that is trained using a multi-task training framework based on plural groups of training examples, which, in turn, are produced using different respective example-generating methods. One such example-generating method constructs a training example from queries in a search session. It operates by identifying the task-related intent the queries, and then identifying at least one sequence of queries in the search session that exhibits a coherent task-related intent. A training example is constructed based on queries in such a sequence.

Abstract: A computer-implemented technique receives an input expression that a user submits with an intent to accomplish some objective. The technique then uses a machine-trained intent encoder component to map the input expression into an input expression intent vector (IEIV). The IEIV corresponds to a distributed representation of the intent associated with the input expression, within a vector intent vector space. The technique then leverages the intent vector to facilitate some downstream application task, such as the retrieval of information. Some application tasks also use a neighbor search component to find expressions that express an intent similar to that of the input expression. A training system trains the intent encoder component based on the nexus between queries and user clicks, as recorded in a search engine's search log.

Abstract: A digital task document can include instructions for performing a task, and a task state data structure can indicate a state of completion of the task. A first update of the data structure can be performed in response to visual user input received from a user profile via a first computer application/device. A second update of the data structure can be performed in response to natural language input received from the user profile via the second computer application/device. A first set of task guidance can be provided to the user profile via the first application/device in a visual format by displaying the task document on a computer display. A second set of task guidance can be provided to the user profile via the second application/device in a natural language format. The first and second sets of task guidance can be provided using the task document and the data structure.

Abstract: The present disclosure relates to systems and methods for discovering relatedness between entities from a corpora of information by automatically extracting attributes from the plurality of heterogeneous entities in a graph. A standardized representation of the extracted attributes from the plurality of heterogeneous entities are propagated across the graph and these propagated attributes are used to find a degree to which the plurality of heterogeneous entities are associated with the extracted attributes. The degree to which the plurality of heterogeneous entities are associated with the extracted attributes is used to create a representation space illustrating a level of relatedness of an entity to another entity of the plurality of heterogeneous entities.

Abstract: This document relates to natural language processing using a framework such as a neural network. One example method involves obtaining a first document and a second document and propagating attention from the first document to the second document. The example method also involves producing contextualized semantic representations of individual words in the second document based at least on the propagating. The contextualized semantic representations can provide a basis for performing one or more natural language processing operations.

Abstract: A digital task document can include instructions for performing a task, and a task state data structure can indicate a state of completion of the task. A first update of the data structure can be performed in response to visual user input received from a user profile via a first computer application/device. A second update of the data structure can be performed in response to natural language input received from the user profile via the second computer application/device. A first set of task guidance can be provided to the user profile via the first application/device in a visual format by displaying the task document on a computer display. A second set of task guidance can be provided to the user profile via the second application/device in a natural language format. The first and second sets of task guidance can be provided using the task document and the data structure.

Abstract: Disclosed are techniques for extracting, identifying, and consuming imprecise temporal elements (“ITEs”). A user input may be received from a client device. A prediction may be generated of one or more time intervals to which the user input refers based upon an ITE model. The user input may be associated with the prediction, and provided to the client device.

Abstract: A computer-implemented technique is described herein that receives an input expression that a user submits with an intent to accomplish some objective. The technique then uses a machine-trained intent encoder component to map the input expression into an input expression intent vector (IEIV). The IEIV corresponds to a distributed representation of the intent associated with the input expression, within a vector intent vector space. The technique then leverages the intent vector to facilitate some downstream application task, such as the retrieval of information. Some application tasks also use a neighbor search component to find expressions that express an intent similar to that of the input expression. A training system trains the intent encoder component based on the nexus between queries and user clicks, as recorded in a search engine's search log.

Abstract: A digital task document can include instructions for performing a task, and a task state data structure can indicate a state of completion of the task. A first update of the data structure can be performed in response to visual user input received from a user profile via a first computer application/device. A second update of the data structure can be performed in response to natural language input received from the user profile via the second computer application/device. A first set of task guidance can be provided to the user profile via the first application/device in a visual format by displaying the task document on a computer display. A second set of task guidance can be provided to the user profile via the second application/device in a natural language format. The first and second sets of task guidance can be provided using the task document and the data structure.

Abstract: Disclosed are techniques for extracting, identifying, and consuming imprecise temporal elements (“ITEs”). A user input may be received from a client device. A prediction may be generated of one or more time intervals to which the user input refers based upon an ITE model. The user input may be associated with the prediction, and provided to the client device.