Rakuten Group

Visualizations are generated from visualization data defined by columns of fields in a database. A user prompt for the visualization is received, then metadata of the fields in the database, one or more system prompts, and the user prompt are transmitted to a natural language query interface to a large language model generative artificial intelligence system. The system prompts include a request to generate a visualization specification and corresponding database query components based upon the metadata and the user prompt. These are received, and a database query including the database query components is composed, then executed to an intermediary data interface of the database that stores the visualization data. Selected visualization data responsive to the database query is received, and the visualization is generated with the visualization specification.

Natural language description based banner creation is implemented by extracting, from a natural language request, a plurality of attributes of a desired banner, selecting, from among a plurality of processes based on the plurality of attributes, a desired process to achieve the desired banner, planning, according to the desired process, a sequence of tasks, each task configured as an input for a predetermined task model among a plurality of task models to obtain a banner component of the desired banner, and executing the desired process to obtain, for each task, the banner component as output of the predetermined task model.

The present disclosure relates to method and system for automated, computationally efficient content analysis of text-based media, such as books or video transcripts. The system receives input text, partitions it, and performs keyword-based searches on the partitions to score them. High-scoring partitions are selected for AI analysis, resulting in a content rating or warning. The system significantly reduces the computational cost and time associated with large-scale content analysis while maintaining high accuracy and recall across multiple categories.

Language modelling with factorization memory is performed by calculating a topic affinity score for each topic vector based on an input token embedding and a topic affinity weight matrix, updating each topic vector based on a corresponding topic affinity score, and merging the updated topic vectors to produce an output token embedding.

Provided is a technology for improving accuracy of processing of a target image. A training data acquisition module (103) of a learning system (1) acquires training data including, as an input portion, a training target image and a training reference image, and including, as a ground truth portion, ground truth information for processing the training target image so that a training target pose of a training target object coincides with a training reference pose of a training reference object. A first model storage unit (100) stores a first model that calculates a first training target feature and a first training reference feature, and outputs first training processing information. A second model storage unit (102) stores a second model that calculates a second training target feature and a second training reference feature, and outputs second training processing information based on the first training processing information, the second training target feature, and the second training reference feature. A training module (104) executes training of at least one of the first model or the second model based on the training data.

At least one processor executes displaying an icon associated with at least one function in a first display area of a display area in an app that provides the at least one function, an image being displayable in the display area, acquiring a usage status of the at least one function, and displaying, based on the usage status of the at least one function, a usage status image in a second display area of the display area, the usage status image being related to the usage status of the at least one function, and the icon being not displayed in the second display area.

An information processing system includes at least one memory configured to store a program and at least one processor configured to execute a process based on the program. The at least one processor is configured to execute acquiring notifications to be summarized in response to an instruction from a user, acquiring a summary of the notifications, displaying the summary of the notifications, and processing, after displaying the summary of the notifications, the notifications in bulk in response to an instruction from the user.

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …

Zusammenfassung wird geladen …