Abstract: Methods, systems, and apparatuses, including computer programs encoded on computer storage media, for training a target generative neural network over a plurality of training iterations. At each iteration, a first data item is generated by processing a conditioning input using the target generative neural network. An improvement generative neural network then processes the first data item and the conditioning input to generate a second, preferred data item. A training example is generated that includes the first and second data items and indicates that the second data item is preferred over the first. The target generative neural network is then trained on this training example. By using this iterative process to dynamically generate preference data, the described techniques improve the performance of the generative neural network beyond the limitations of static, offline datasets without requiring computationally expensive reward models or external human annotation.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a diffusion neural network using a region-aware fine-tuning process. After training, the diffusion neural network can be used to generate an image conditioned on a conditioning input.

Abstract: Techniques are disclosed that enable the generation of candidate endorsements for recommended items of content using an ensemble of nominators. Various implementations include each nominator in the ensemble providing a candidate endorsement for each recommended item of content. Additionally or alternatively, an endorsement is selected to present to the user based on a score determined for each candidate endorsement.

Abstract: Aspects of the disclosed technology include computer-implemented systems and methods for machine-learned multimodal models for feedback predictions for synthetic content. A machine-learned multimodal model is configured to generate a feature map based at least in part on fusion of image information and text information from a synthetic image and a text prompt. The model is configured to generate a set of text tokens based at least in part on fusion of the image information and the text information. The model is configured to generate at least one misalignment or implausibility heatmap based at least in part on the at least one feature map. The model is configured to generate at least one predicted misalignment sequence based at least in part on the set of text tokens.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating images. In one aspect, a method comprises: obtaining an input query comprising text; processing, using prompt expansion model, a model input comprising at least the input query to generate a set of expanded prompts of the input query, wherein each of the expanded prompts describes an image in more detail than the input query; and for one or more of the expanded prompts in the set, generating, using an image generation model, a respective image that represents the expanded prompt.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for analyzing embedding spaces using large language models. In one aspect, a method performed by one or more computers for analyzing a target embedding space using a neural network configured to perform a set of machine learning tasks is described. The method includes: obtaining, for each of one or more entities, a respective domain embedding representing the entity in the target embedding space; receiving a text prompt including a sequence of input tokens describing a particular machine learning task in the set to be performed on the one or more entities; preparing, for the neural network, an input sequence including each input token in the text prompt and each domain embedding; and processing the input sequence, using the neural network, to generate a sequence of output tokens describing a result of the particular machine learning task.

Abstract: Techniques are disclosed that enable the generation of candidate endorsements for recommended items of content using an ensemble of nominators. Various implementations include each nominator in the ensemble providing a candidate endorsement for each recommended item of content. Additionally or alternatively, an endorsement is selected to present to the user based on a score determined for each candidate endorsement.

Abstract: Techniques are disclosed that enable the generation of candidate endorsements for recommended items of content using an ensemble of nominators. Various implementations include each nominator in the ensemble providing a candidate endorsement for each recommended item of content. Additionally or alternatively, an endorsement is selected to present to the user based on a score determined for each candidate endorsement.

Abstract: Techniques are disclosed that enable the generation of candidate endorsements for recommended items of content using an ensemble of nominators. Various implementations include each nominator in the ensemble providing a candidate endorsement for each recommended item of content. Additionally or alternatively, an endorsement is selected to present to the user based on a score determined for each candidate endorsement.

Abstract: In some natural language understanding (NLU) applications, results may not be tailored to the user's query. In an embodiment of the present invention, a method includes tagging elements of automated speech recognition (ASR) data based on an ontology stored in a memory. The method further includes indexing tagged elements to an entity of the ontology. The method further includes generating a logical form of the ASR data based on the tagged elements and the indexed entities. The method further includes mapping the logical form to a query to a respective corresponding database stored in the memory. The method further includes issuing the query to the respective corresponding databases. The method further includes presenting results of the query to the user via a display or a voice response system.

Abstract: Presenting natural-language-understanding (NLU) results can include redundancies and awkward sentence structures. In an embodiment of the present invention, a method includes, responsive to receiving a result to a NLU query, loading a matching template of a plurality of templates stored in a memory. Each template has mask fields associated with at least one property. The method compares the properties of the mask fields of each of the templates to properties of the query and properties of the result, and selects the matching template. The method further completes the matching template by inserting fields of the result into corresponding mask fields of the matching template. The method may further suppress certain mask fields of the matching template to increase brevity and improve the naturalness of the response when appropriate based on the results of the NLU query. The method further presents the completed matching template to a user via a display.

Abstract: A method is provided for representing and updating the state of a dialog involving a series of queries and commands to an artificial intelligence system. Each statement within the dialogue may be modeled as a relational tree spanning nodes corresponding to named entities within the statement. A data structure may be used to store each of these trees and to modify them as the dialog progresses. A subsequent statement in the dialog may be parsed and its contents used to update an ongoing search initiated within that dialog. Statements may be used for the update process despite being fragmentary or not corresponding to any predetermined grammar. An algorithm is disclosed for updating the trees within the data structure after a new statement is parsed.

Abstract: A method includes, with a physical computing system, executing a navigational application for a printer interface of a printer connected to the physical computing system through a network, with the physical computing system, using a content model for the navigational application, the content model defining attributes and a category for a screen. The method further includes, with the physical computing system, receiving content associated with the category, and with the physical computing system, providing to the printer the screen and the content in a format that is executable by the printer.

Abstract: In some natural language understanding (NLU) applications, results may not be tailored to the user's query. In an embodiment of the present invention, a method includes tagging elements of automated speech recognition (ASR) data based on an ontology stored in a memory. The method further includes indexing tagged elements to an entity of the ontology. The method further includes generating a logical form of the ASR data based on the tagged elements and the indexed entities. The method further includes mapping the logical form to a query to a respective corresponding database stored in the memory. The method further includes issuing the query to the respective corresponding databases. The method further includes presenting results of the query to the user via a display or a voice response system.

Abstract: Presenting natural-language-understanding (NLU) results can include redundancies and awkward sentence structures. In an embodiment of the present invention, a method includes, responsive to receiving a result to a NLU query, loading a matching template of a plurality of templates stored in a memory. Each template has mask fields associated with at least one property. The method compares the properties of the mask fields of each of the templates to properties of the query and properties of the result, and selects the matching template. The method further completes the matching template by inserting fields of the result into corresponding mask fields of the matching template. The method may further suppress certain mask fields of the matching template to increase brevity and improve the naturalness of the response when appropriate based on the results of the NLU query. The method further presents the completed matching template to a user via a display.

Abstract: One or more embodiments of techniques or systems for modeling familiarity for a traveler are provided herein. Familiarity evidence can be received, indicative of how familiar a traveler is with an area or road segment, and based on a number of visits the traveler has made to that area. The familiarity evidence can be used to generate one or more familiarity models indicative of a predicted familiarity of locations around the area. Familiarity models can be based on kernels, graph distances, Markov random fields (MRFs), etc. When route directions are generated from an origin location to a destination location, one or more of the directions can be provided based on one or more of the familiarity models. For example, if a familiarity model indicates that a traveler is familiar with a route, driving directions of the route can be adapted to be more succinct.

Abstract: A method is provided for representing and updating the state of a dialog involving a series of queries and commands to an artificial intelligence system. Each statement within the dialogue may be modeled as a relational tree spanning nodes corresponding to named entities within the statement. A data structure may be used to store each of these trees and to modify them as the dialog progresses. A subsequent statement in the dialog may be parsed and its contents used to update an ongoing search initiated within that dialog. Statements may be used for the update process despite being fragmentary or not corresponding to any predetermined grammar. An algorithm is disclosed for updating the trees within the data structure after a new statement is parsed.

Abstract: An utterance is received from a user specifying a location attribute and a landmark. A set of candidate locations is identified based on the specified location attribute, and a confidence score can be determined for each candidate location. A set of landmarks is identified based on the specified landmark, and confidence scores can be determined for the landmarks. An associated kernel model is generated for each landmark. Each kernel model is centered at the location of the associated landmark on a map, and the amplitude of the kernel model can be based on landmark attributes, landmark confidence scores, characteristics of the user, and the like. The candidate locations are ranked based on the amplitudes of overlapping kernel models at the candidate locations, and can also be ranked based on confidence scores associated with the candidate locations. A candidate location is selected and presented to the user based on the candidate location ranking.

Abstract: A powertrain of a hybrid electric vehicle (HEV) is controlled. A first value ?1 and a second value ?2 are determined. ?1 represents a proportion of an instantaneous power requirement (Preq) supplied by an engine of the HEV. ?2 controls a recharging rate of a battery of the HEV. A determination is performed, based on ?1 and ?2, regarding how much engine power to use (Peng) and how much battery power to use (Pbatt). Peng and Pbatt are sent to the powertrain.

Abstract: Systems and methods of establishing a secure connection between a proxy server and a base station device are disclosed. An example of a method includes providing a proxy server with a session token for a mobile device from a cloud service. The method also includes providing a session code directly to the mobile device from the cloud service based on the session token. The method also includes providing access by the proxy server to the base station device if the proxy server provides the session code to the cloud service.