Abstract: In an example embodiment, a system includes an inverter configured to operate in at least one of a charging mode or a drive mode, a cascaded direct current (DC)-DC converter, the DC-DC converter including a first portion of the inverter and at least one controller configured to selectively couple the first portion of the inverter to a first portion of the cascaded DC-DC converter during the charging mode, and selectively couple the inverter to a second portion of the cascaded DC-DC converter during the drive mode.

Abstract: A text dialogue between a first user and a second user is obtained. Based on an analysis of the text dialogue, workflows that are relevant to the text dialogue are identified. A list of the workflows is provided to a client device associated with the first user. An indication of a selection of a particular workflow from the list of the workflow is received from the client device associated with the first user. Based on the particular workflow, a communication session is established between the second user and an automated chatbot associated with the particular workflow.

Abstract: Techniques for updating a machine learning model based on user interactions are described. In particular, in some examples, user interactions with a chatbot provide aspects of a data set to be used to train or fine-tune a ML model. In some examples, this is accomplished by collecting data from a first plurality of interactions with a machine learning (ML) model; generating a variant of the ML model using the collected data by: filtering the collected data to create a first data set, training the ML model based on the first data set to generate an adapted ML model, and fine-tuning the adapted ML model on a second data set, different than the first data set to generate the variant of the ML model.

Abstract: An indication from a live agent in a communication session with a user to invoke a specified automated agent interactive conversation flow is received. The communication session with the user is transitioned from the live agent to the specified automated agent interactive conversation flow. During the specified automated agent interactive conversation flow, the live agent retains control over the communication session. An indication of an interaction of the user with the specified automated agent interactive conversation flow is provided. The communication session with the user is returned back to the live agent.

Abstract: Aspects of the subject technology provide for suggested home screens of an electronic device. The suggested home screens can include suggested background images, user interface elements, colors, styles, photos, and the like. Foreground interface elements can be positioned on the suggested home screen so as to not obscure a feature of the background image of the suggested home screen. The home screens may be generated and/or suggested specifically for various operational modes of the electronic device.

Abstract: A method includes receiving a permission request from a user to perform an operation in a computer environment. Data elements of operations previously performed by a plurality of users are received. A set of operation types are generated by identifying data elements of operations previously performed by other users of the plurality of users having operation types performed by the user. The data elements in the set are mapped into a graph database. Permission messages are generated with a correct response identified from a data attribute from a specific operation type previously performed by the user, and wrong responses identified from the specific operation type previously performed by the other users. An accumulated probability based on the permission respond message is computed using the graph database. The user is tagged based on the accumulated probability as a permissioned user for performing the operation, or a non-permissioned violator user.

Abstract: Generating granular clusters for real-time processing is provided. The systems can identify tokens based on aggregating input from computing devices over a time interval. The systems can identify, based on metrics, a subset of tokens for cluster generation. The systems can generate, via a clustering technique, token clusters from the subset of the tokens, each of the token clusters comprising two or more tokens from the subset of the tokens. The systems can apply a de-duplication technique to each of the token clusters. The systems can apply a filtering technique to the token clusters to remove tokens erroneously grouped in a token cluster. The systems can assign, based on a selection process, a label for each of the token clusters. The systems can activate, based on a number of remaining tokens in each of the token clusters, a subset of the token clusters for real-time content selection.

Abstract: In an example embodiment, a system includes a plurality of inverters configured to operate in at least one of a charging mode or a drive mode, a plurality of motors and at least one controller configured to selectively couple the plurality of inverters to the plurality of motors during a drive mode, and selectively couple at least one of the inverters to an alternating current (AC) source during the charging mode and decouple the at least one of the inverters from the plurality of motors during the charging mode.

Abstract: Aspects of the subject technology provide for suggested home screens and/or lock screens of an electronic device. The suggested home screens and/or lock screens can include suggested backgrounds, user interface elements, colors, styles, photos, and the like, and can be generated and/or suggested for a user of an electronic device based on activity history information for the user and/or the electronic device. In one or more implementations, the home screens and/or lock screens may be generated and/or suggested specifically for various operational modes of the electronic device.

Abstract: Methods, systems and computer products disclosed herein are directed to an O.C. Engine that detects a first request sent from a requesting account during a target account session, the target account associated with multiple sessions. The O.C. Engine generates a first request parent node active for a range of time. The O.C. Engine detects a subsequent second request sent from the requesting account and generates a second request node. The O.C. Engine verifies that the subsequent second request is not an initial request by the requesting account during a range of time that has not yet expired. Upon verification, the O.C. Engine identifies the second request node as a child node by generating a link between the parent node and the child node.

Abstract: A method includes receiving a permission request from a user to perform an operation in a computer environment. Data elements of operations previously performed by a plurality of users are received. A set of operation types are generated by identifying data elements of operations previously performed by other users of the plurality of users having operation types performed by the user. The data elements in the set are mapped into a graph database. Permission messages are generated with a correct response identified from a data attribute from a specific operation type previously performed by the user, and wrong responses identified from the specific operation type previously performed by the other users. An accumulated probability based on the permission respond message is computed using the graph database. The user is tagged based on the accumulated probability as a permissioned user for performing the operation, or a non-permissioned violator user.

Abstract: Methods, systems and computer products disclosed herein are directed to an O.C. Engine that detects a first request sent from a requesting account during a target account session, the target account associated with multiple sessions. The O.C. Engine generates a first request parent node active for a range of time. The O.C. Engine detects a subsequent second request sent from the requesting account and generates a second request node. The O.C. Engine verifies that the subsequent second request is not an initial request by the requesting account during a range of time that has not yet expired. Upon verification, the O.C. Engine identifies the second request node as a child node by generating a link between the parent node and the child node.

Abstract: Certain aspects of the present disclosure provide techniques for providing sentiment analysis of content. In order to determine the overall sentiment of content, a request is received by a sentiment analyzer, which then identifies a content identification number and retrieves comments associated with the content identification number. The sentiment analyzer pre-processes the comments, which includes removing all personal identifying information from the comments. The sentiment analyzer sends the pre-processed comments to a natural language processing service, and in turn, receives sentiment indications corresponding to the comments provided. Based on the sentiment scores, the sentiment analyzer generates a sentiment analysis and displays the sentiment analysis in the graphical user interface generated by the sentiment analyzer.

Abstract: Methods, systems and computer products as described herein are directed to a Live Listing Engine. Live Listing Engine sends a live video stream to a viewer computer device and a host computer device. The live video stream includes a view of an item from a perspective of the host computer device. The Live Listing Engine receives an upload of an image of the view of the item captured from a preview window rendered on the host computer device. The Live Listing Engine receives listing information associated with the image from the host computer device. The Live Listing Engine generates a listing thumbnail based on the image and the listing information and inserts the listing thumbnail into the live video stream for display at the viewer computer device and the host computer device.

Abstract: Certain aspects of the disclosure provide systems and methods for training an information extraction transformer model architecture directed to pre-training a first multimodal transformer model on an unlabeled dataset, training a second multimodal transformer model on a first labeled dataset to perform a key information extraction task processing the unlabeled dataset with the second multimodal transformer model to generate pseudo-labels for the unlabeled dataset, training the first multimodal transformer model based on a second labeled dataset comprising one or more labels, the pseudo-labels generated, or combinations thereof to generate a third multimodal transformer model, generating updated pseudo-labels based on label completion predictions from the third multimodal transformer model, and training the third multimodal transformer model using a noise-aware loss function and the updated pseudo-labels to generate an updated third multimodal transformer model.

Abstract: Methods, systems and computer products as described herein are directed to a Live Listing Engine. Live Listing Engine sends a live video stream to a viewer computer device and a host computer device. The live video stream includes a view of an item from a perspective of the host computer device. The Live Listing Engine receives an upload of an image of the view of the item captured from a preview window rendered on the host computer device. The Live Listing Engine receives listing information associated with the image from the host computer device. The Live Listing Engine generates a listing thumbnail based on the image and the listing information and inserts the listing thumbnail into the live video stream for display at the viewer computer device and the host computer device.

Abstract: Aspects of the subject technology provide for suggested home screens and/or lock screens of an electronic device. The suggested home screens and/or lock screens can include suggested backgrounds, user interface elements, colors, styles, photos, and the like, and can be generated and/or suggested for a user of an electronic device based on activity history information for the user and/or the electronic device. In one or more implementations, the home screens and/or lock screens may be generated and/or suggested specifically for various operational modes of the electronic device.

Abstract: Systems and methods are described for generating dynamic web clippings. A web clipping service can obtain from a client computing device metadata for a dynamic web clipping, which includes an identifier of a web document and state information of a web browser that has rendered the web document on the client computing device, the state information enabling replication of a browsing session of the client computing device. The web clipping service can use the metadata to recreate the browsing session of the client device on the web clipping service, and capture an image of the web document as rendered within he recreated browsing session. The web clipping service can then provide the image of the web document as the dynamic web clipping.

Abstract: Generating granular clusters for real-time processing is provided. The systems can identify tokens based on aggregating input from computing devices over a time interval. The systems can identify, based on metrics, a subset of tokens for cluster generation. The systems can generate, via a clustering technique, token clusters from the subset of the tokens, each of the token clusters comprising two or more tokens from the subset of the tokens. The systems can apply a de-duplication technique to each of the token clusters. The systems can apply a filtering technique to the token clusters to remove tokens erroneously grouped in a token cluster. The systems can assign, based on a selection process, a label for each of the token clusters. The systems can activate, based on a number of remaining tokens in each of the token clusters, a subset of the token clusters for real-time content selection.

Abstract: A control module comprises software instructions for execution by an electronic data processor. An inductance estimator is configured to estimate direct inductance and quadrature inductance associated with the electric machine based on a first set of equations that assume constant rotor resistance and constant magnetic flux. A torque estimator is configured to estimate an observed torque associated with a rotor of the electric machine based a commanded direct-axis voltage and a commanded quadrature-axis voltage, where the observed torque is proportional to the observed power consumption of the electric machine. A pulse-width-modulation (PWM) module is configured to provide pulse-width modulated signals to the electric machine based on the commanded direct-axis voltage and a commanded quadrature-axis voltage. A back-electromotive force adjustment is derived from an estimated, observed torque and a commanded torque.