Abstract: A method, apparatus, non-transitory computer readable medium, apparatus, and system for image generation include obtaining a sketch input depicting an object, processing the sketch input to obtain sketch guidance, and generating a synthesized image based on the sketch guidance using an image generation model, where the synthesized image depicts the object from the sketch input.

Abstract: A method includes receiving an input including a target style and a glyph. The method further includes masking the glyph. The method further includes generating a stylized glyph by a glyph generative model using the masked glyph. The method further includes rendering the stylized glyph as a unicode stylized glyph.

Abstract: In one aspect, a query management module executing on a processor receives, from a large language model (LLM), a graph query generated by the LLM based on a natural language query (NLQ). A validation module identifies an error in the graph query. The query management module provides an indication of the error to the LLM. The query management module receives a modified graph query from the LLM. The validation module validates the modified graph query. Based on the validation of the modified graph query, the query management module executes the modified graph query against a knowledge graph to return a result as a response to the NLQ.

Abstract: Embodiments of the present disclosure include extracting structured text from a source document. The structured text comprises a plurality of source sections. Some embodiments generate a semantic outline based on the structured text. In some examples, the semantic outline comprises a plurality of output headings. Some embodiments generate text content corresponding to each of the plurality of output headings. An image is selected from the source document for each of the plurality of output headings by computing a similarity score between the image and the text content. Then, an output document is generated based on the semantic outline, where the output document comprises a plurality of output sections corresponding to the plurality of output headings, respectively.

Abstract: Face anonymization techniques are described that overcome conventional challenges to generate an anonymized face. In one example, a digital object editing system is configured to generate an anonymized face based on a target face and a reference face. As part of this, the digital object editing system employs an encoder as part of machine learning to extract a target encoding of the target face image and a reference encoding of the reference face. The digital object editing system then generates a mixed encoding from the target and reference encodings. The mixed encoding is employed by a machine-learning model of the digital object editing system to generate a mixed face. An object replacement module is used by the digital object editing system to replace the target face in the target digital image with the mixed face.

Abstract: An improved electronic communication system schedules transmission of electronic communications based on a predicted open time and click time. The open and click times are predicted from a machine learning model that is trained to optimize for both tasks. Additionally, when training the machine learning model, the loss used for adjusting the system to achieve a desired accuracy may be a biased loss determined from a function that penalizes overpredicting the open time. As such, the loss value may be determined by different set of rules depending on whether the predicted time is greater than the actual time or not.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media that utilize machine learning to generate subject lines from subject line keywords. In one or more embodiments, the disclosed systems receive, from a client device, one or more subject line keywords. Additionally, the disclosed systems generate, utilizing a subject generation machine-learning model having learned parameters, a subject line by selecting one or more words for the subject line from a word distribution based on the one or more subject line keywords. The disclosed systems further provide, for display on the client device, the subject line.

Abstract: Digital content coediting techniques are described, including detecting a change in state of digital content that is maintained as part of a coediting session involving a plurality of client devices. One or more elements of the digital content are located corresponding to the change in state and a fingerprint is generated responsive to the detecting, which is based on a hash of a serialization of the one or more elements. The fingerprint is communicated for receipt by at least one of the plurality of client devices, the fingerprint configured to cause local synchronization of the digital content.

Abstract: Embodiments are disclosed for music generation. The method may include receiving a text input describing music to be generated by a neural network and obtaining an initial noise laten. The neural network generates a music spectrogram based on the initial noise latent and the text input. One or more features are extracted from the music spectrogram. A loss is determined based on the one or more features from the music spectrogram and one or more target features of a target output. An optimized noise latent is obtained based on the loss and a new music spectrogram is generated using the optimized noise latent.

Abstract: Digital content coediting techniques are described, including detecting a change in state of digital content that is maintained as part of a coediting session involving a plurality of client devices. One or more elements of the digital content are located corresponding to the change in state and a fingerprint is generated responsive to the detecting, which is based on a hash of a serialization of the one or more elements. The fingerprint is communicated for receipt by at least one of the plurality of client devices, the fingerprint configured to cause local synchronization of the digital content.

Abstract: A method, apparatus, non-transitory computer readable medium, and system for generating sound effects for a video includes obtaining a selection input indicating an element of a video. Embodiments then generate, using an audio generation model, a synthetic audio clip based on the selection input. Embodiments subsequently generate a multimedia file including the video and the synthetic audio clip.

Abstract: A method, apparatus, non-transitory computer readable medium, and system of training a domain-specific language model are described. One or more aspects of the method, apparatus, non-transitory computer readable medium, and system include obtaining domain-specific training data including a plurality of domain-specific documents having a document structure corresponding to a domain, and obtaining domain-agnostic training data including a plurality of documents outside of the domain. The domain-specific training data and the domain-agnostic training data are used to train a language model to perform a domain-specific task based on the domain-specific training data and to perform a domain agnostic task based on the domain-agnostic training data.

Abstract: Systems and methods for generating images using hybrid sampling include obtaining a noisy image and generating a first denoised image during a first reverse diffusion phase using a diffusion neural network. The first denoised image is generated based on a first sampler that uses a first sampling density during at least a portion of the first reverse diffusion phase. Subsequently, a second denoised image is generated based on the first denoised image during a second reverse diffusion phase using the diffusion neural network. The second denoised image is generated based on a second sampler that uses a second sampling density different from the first sampling density during at least a portion of the second reverse diffusion phase.

Abstract: Systems and methods for image tagging are provided. One aspect of the systems and methods includes encoding an image and a tag of the image using a multimodal encoder to obtain an image embedding and a text embedding, respectively. Another aspect of the systems and methods includes generating training data for a machine learning model by filtering a plurality of image-tag pairs based on a similarity between the image embedding and the text embedding. Another aspect of the systems and methods includes training the machine learning model using the training data.

Abstract: Grid structure control techniques and systems are described that support use of a flexible grid structure and layout control. In an implementation, an input is received via a user interface, the user interface displaying a grid having a plurality of grid cells. A determination is made as to whether the input corresponds to a first said grid cell of the grid. Responsive to determining by a processing device the input corresponds to the first said grid cell, a first edge of the first grid cell and a second edge of a second grid cell is moved as following the input. The second edge is disposed opposite and proximal to the first edge.

Abstract: Systems and methods for document classification are described. Embodiments of the present disclosure generate classification data for a plurality of samples using a neural network trained to identify a plurality of known classes; select a set of samples for annotation from the plurality of samples using an open-set metric based on the classification data, wherein the annotation includes an unknown class; and train the neural network to identify the unknown class based on the annotation of the set of samples.

Abstract: A method, apparatus, non-transitory computer readable medium, apparatus, and system for scene re-lighting using direct shading control include obtaining an input image and a lighting direction indicator that describes a lighting direction. A direct shading map is generated based on the input image and the lighting direction indicator and a shaded image is generated depicting an object from the input image with shading consistent with the lighting direction based on the shading map.

Abstract: In some examples, a computing system accesses a field of view (FOV) image that has a field of view less than 360 degrees and has low dynamic range (LDR) values. The computing system estimates lighting parameters from a scene depicted in the FOV image and generates a lighting image based on the lighting parameters. The computing system further generates lighting features generated the lighting image and image features generated from the FOV image. These features are aggregated into aggregated features and a machine learning model is applied to the image features and the aggregated features to generate a panorama image having high dynamic range (HDR) values.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer-readable media that modifies parameters of a fused feature extractor. In particular, the disclosed systems generate inferred masks from digital images and digital text prompts using a fused feature extractor. Furthermore, the disclosed systems identify a subset of the inferred masks that satisfy a validity threshold. Moreover, the disclosed systems generate an augmented training set by combining the subset of the inferred masks with a training set that includes the ground truth masks. Further, the disclosed systems generate object mask predictions from the augmented training set and determine ground truth and pseudo measures of loss by comparing the object mask predictions with the inferred masks and the ground truth masks. From the ground truth and pseudo measures of loss, the disclosed systems modify parameters of the fused feature extractors.

Abstract: A method, apparatus, non-transitory computer readable medium, and system for generating suggested prompts include obtaining a sequence of text prompts associated with a user and determining a session concept for the user based on the sequence of text prompts. Embodiments then generate, using a prompt generation model, an image generation prompt based on the sequence of text prompts and the session concept. Subsequently, embodiments generate, using an image generation model, a synthetic image based on the image generation prompt.