Abstract: Some aspects relate to technologies for dynamically generating digital content for events using event data and content intent descriptors. In some aspects, when a content server identifies an event for digital content creation, the content server provides data to a user device that is based on event data for the event and a content intent descriptor. The user device generates a prompt using the received data and provides the prompt to a generative model on the user device, causing the generative model to generate a digital content item using the prompt as input. The user device then presents the digital content item.

Abstract: Some aspects relate to technologies for generating custom digital content using a content intent descriptor from a content server and on-device contextual data maintained on a user device. In some aspects, a user device receives a content intent descriptor communicated over a network from a content server. The user device generates a prompt using the content intent descriptor and on-device contextual data maintained on the user device. A generative model is caused to generate a digital content item using the prompt, and the digital content item is presented on the user device.

Abstract: Some aspects relate to technologies for generating and/or presenting digital content using on-device subscription data maintained on a user device. In some aspects, a user device receives a content intent descriptor communicated over a network from a content server. The user device performs a comparison of on-device subscription data with the content intent descriptor. Based on the comparison, the user device generates a prompt using the content intent descriptor. A generative model is caused to generate a digital content item using the prompt, and the digital content item is presented on the user device.

Abstract: Methods, computer systems, computer storage media, and graphical user interfaces are provided for facilitating identification of relevant data using data embeddings. In one implementation, a query embedding representing a query is generated. Using the query embedding, a data embedding representing data in a hyperspace that is similar to the query embedding is identified. Thereafter, the data, represented by the data embedding identified to be similar to the query embedding, is identified as relevant to the query. Content may then be generated via one or more generative artificial intelligence (AI) models based on at least a portion of the query and the data identified as relevant to the query. Such content may be displayed via a graphical user interface.

Abstract: Methods, computer systems, computer storage media, and graphical user interfaces are provided for facilitating identification of relevant data using a set of hierarchical knowledge graphs. In one implementation, a data source relevant to a query is identified using a root knowledge graph. Thereafter, a data knowledge graph associated with the data source identified as relevant to the query is identified from among a plurality of data knowledge graphs. Such a data knowledge graph is used to identify a set of data relevant to the query. In embodiments, content may be generated, via one or more generative artificial intelligence (AI) models, based on at least a portion of the query and at least a portion of the set of data identified as relevant to the query. Such content may be provided for display via a graphical user interface.

Abstract: Methods and systems are provided for transcreation of textual content using a language model. In embodiments described herein, a user inputs content and selects a target locale to transcreate the content for the target locale. A language model is prompted to classify textual portions of the content into popular-topic categories of a popular-topic taxonomy and determine a subject of each of popular-topic categories based on the classified textual portions. The language model is prompted to determine a substitution subject relevant to the target locale based on the subject of each popular-topic category. Refined content is generated by replacing the classified textual portions of the content with text corresponding to the substitution subject. Transcreated content is generated by prompting the language model to generate the transcreated content based on the refined content and communication-interpretation rules relevant to the target locale. The transcreated content is displayed to the user.

Abstract: Methods and systems are provided for transcreation of textual content using a language model. In embodiments described herein, a user inputs content and selects a target locale to transcreate the content for the target locale. A language model is prompted to classify textual portions of the content into popular-topic categories of a popular-topic taxonomy and determine a subject of each of popular-topic categories based on the classified textual portions. The language model is prompted to determine a substitution subject relevant to the target locale based on the subject of each popular-topic category. Refined content is generated by replacing the classified textual portions of the content with text corresponding to the substitution subject. Transcreated content is generated by prompting the language model to generate the transcreated content based on the refined content and communication-interpretation rules relevant to the target locale. The transcreated content is displayed to the user.

Abstract: Some aspects relate to technologies for employing generative models for content generation and interactive content editing. In accordance with some aspects, user input is received for generating a content item of a content type having a number of fragments. A model set for the content type is identified. The model set comprises generative models for the fragments and an execution order specifying an order for generating the fragments. A root generative model from the model set is caused to generate text for a root fragment in the execution order based on the user input. Each subsequent generative model in the model set is sequentially caused to generate text for each subsequent fragment in the execution order for the model set, wherein input for each subsequent generative model includes text of any previous fragments in the execution order. The content item is generated by combining the text of the fragments.

Abstract: Some aspects relate to technologies for generating time series of sketches and using the time series of sketches for approximate query processing. In accordance with some aspects, tabular data is accessed that has a number of columns. Responsive to identifying a first column as comprising numerical data, sketches are generated for the numerical data for each of a number of time steps, and the sketches for the numerical data are stored as a first time series of sketches. Responsive to identifying a second column as comprising categorical data, sketches are generated for the categorical data for each of the time steps, and the sketches for the categorical data are stored as a second time series of sketches. When a query is received, a response to the query is provided using sketches from the first time series of sketches and/or the second time series of sketches.

Abstract: In accordance with the described techniques, an image transformation system receives an input image and a text prompt, and leverages a generator network to edit the input image based on the text prompt. The generator network includes a plurality of layers configured to perform respective edits. A plurality of masks are generated based on the text prompt that define local edit regions, respectively, of the input image for respective layers of the generator network. Further, the generator network generates an edited image by editing the input image based on the plurality of masks, the respective edits of the respective layers, and the text prompt.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media that control bias in machine learning models by utilizing a fairness deviation constraint to learn a decision matrix that modifies machine learning model predictions. In one or more embodiments, the disclosed systems generate, utilizing a machine learning model, predicted classification probabilities from a plurality of samples comprising a plurality of values for a data attribute. Moreover, the disclosed systems determine utilizing a decision matrix and the predicted classification probabilities, that the machine learning model fails to satisfy a fairness deviation constraint with respect to a value of the data attribute. In addition, the disclosed systems generate a modified decision matrix for the machine learning model to satisfy the fairness deviation constraint by selecting a modified decision threshold for the value of the data attribute.

Abstract: In implementations of systems for generating image metadata using a compact color space, a computing device implements a color system to receive input data describing pixels of a digital image and corresponding RGB values of the pixels. The color system assigns a color of a compact color space to each of the pixels based on the corresponding RGB values of the pixels. The compact color space includes a subset of colors included in an RGB color space. The color system computes a histogram of colors of the compact color space and determines a particular color of the compact color space based on the histogram. The color system generates color metadata for the digital image describing a natural language name of the particular color of the compact color space.

Abstract: Some aspects relate to technologies for generating time series of sketches and using the time series of sketches for approximate query processing. In accordance with some aspects, tabular data is accessed that has a number of columns. Responsive to identifying a first column as comprising numerical data, sketches are generated for the numerical data for each of a number of time steps, and the sketches for the numerical data are stored as a first time series of sketches. Responsive to identifying a second column as comprising categorical data, sketches are generated for the categorical data for each of the time steps, and the sketches for the categorical data are stored as a second time series of sketches. When a query is received, a response to the query is provided using sketches from the first time series of sketches and/or the second time series of sketches.

Abstract: Methods and systems are provided for using Shapley values to evaluate prompt generation parameters. In embodiments described herein, a selection of prompt parameters are accessed. A plurality of prompts are generated as a function of a combination of the prompt parameters. A corresponding quality metric is determined for each of the prompts. Prompt parameter contribution metrics are determined using a Shapley-value-based determination corresponding to a contribution of each of the prompt parameters to the corresponding content quality metric for each of the prompts. The prompt parameter contribution metrics are then displayed.

Abstract: In accordance with the described techniques, an image transformation system receives an input image and a text prompt, and leverages a generator network to edit the input image based on the text prompt. The generator network includes a plurality of layers configured to perform respective edits. A plurality of masks are generated based on the text prompt that define local edit regions, respectively, of the input image for respective layers of the generator network. Further, the generator network generates an edited image by editing the input image based on the plurality of masks, the respective edits of the respective layers, and the text prompt.

Abstract: The present disclosure is directed toward systems, methods, and non-transitory computer readable media that provide a contextual content generation system that trains and implements a unique machine learning architecture to generate context-specific digital content items based on a digital guideline document. In particular, the disclosed systems select a content generation method from among prompt engineering and/or updating one or more machine learning models to generate digital content. For example, the disclosed systems utilize machine learning models to extract key elements from a digital guideline document comprising context-specific guidelines for digital content. Further, the disclosed systems generate an augmented prompt comprising indications of key elements from the digital guideline document.

Abstract: Methods, computer systems, computer-storage media, and graphical user interfaces are provided for determining user affinities by tracking historical user interactions with tagged digital content and using the user affinities in content generation applications. Accordingly, the system may track user interactions with published digital content in order to generate user interaction reports whenever a user engages with the digital content. The system may aggregate the interaction reports to generate an affinity profile for a user or audience of users. A marketer may then generate digital content for a target user or audience of users and the system may process the digital content to generate a set of tags for the digital content. Based on the set of tags, the system may then evaluate the digital content in view of the affinity profile for the target user/audience to determine similarities or differences between the digital content and the affinity profile.

Abstract: Embodiments are generally directed to extending artificial intelligence (AI) and machine learning (ML) techniques to generate content predicted to elicit a performance response from an intended recipient of a target audience. One method of generating content includes determining content generation information from a user prompt, the content generation information comprising a subject, an audience segment, and a performance indicator; and providing the content generation information to a content generation model to generate at least one item of audience-targeted content corresponding to the subject targeted to the audience segment to elicit a response defined by the performance indicator, wherein the content generation module comprises a natural language processing (NLP) model trained, via a content generation training module, using reinforcement learning based on a reward of a performance prediction determined by a performance prediction model based on historical performance data.

Abstract: Techniques for prompt-based image relighting and editing are described that support automatic generation of an edited digital image with high-fidelity and realistic lighting effects and background features. A processing device, for instance, receives as input a digital image that depicts a digital object, a lighting prompt, and a background prompt. The processing device generates a relit digital object that has a lighting condition specified by the lighting prompt applied to the digital object. The processing device further generates a background that includes a feature specified by the background prompt and the lighting condition. The processing device generates an edited digital object for output that includes the relit digital object and the background. The processing device further leverages a shadow synthesis model to edit shadows in the edited digital image. In this way, the techniques described herein preserve content details of the digital object when applying background and lighting effects.

Abstract: Some aspects relate to technologies for generating time series of sketches and using the time series of sketches for approximate query processing. In accordance with some aspects, tabular data is accessed that has a number of columns. Responsive to identifying a first column as comprising numerical data, sketches are generated for the numerical data for each of a number of time steps, and the sketches for the numerical data are stored as a first time series of sketches. Responsive to identifying a second column as comprising categorical data, sketches are generated for the categorical data for each of the time steps, and the sketches for the categorical data are stored as a second time series of sketches. When a query is received, a response to the query is provided using sketches from the first time series of sketches and/or the second time series of sketches.