Abstract: Techniques for generation of compressed representations for appearance of fiber-based digital assets are described that support computationally efficient and high fidelity rendering of digital assets that include fiber primitives under a variety of lighting conditions and view directions. A processing device, for instance, receives a digital asset that includes fiber primitives to be included in a three-dimensional digital scene. The processing device generates a compressed representation of the digital asset that maintains a geometry of the digital asset and includes a precomputed light transport. The processing device then inserts the compressed representation into the digital scene, such as at a location relative to one or more light sources. The content processing system applies one or more lighting effects to the compressed representation based on the precomputed light transport and the location relative to the one or more light sources.

Abstract: Dialogue skeleton assisted prompt transfer for dialogue summarization techniques are described that support training of a language model to perform dialogue summarization in a few-shot scenario. A processing device, for instance, receives a training dataset that includes training dialogues. The processing device then generates dialogue skeletons based on the training dialogues using one or more perturbation-based probes. The processing device trains a language model using prompt transfer between a source task, e.g., dialogue state tracking, and a target task, e.g., dialogue summarization, using the dialogue skeletons as supervision. The processing device then receives an input dialogue and uses the trained language model to generate a summary of the input dialogue.

Abstract: Embodiments are disclosed for performing universal segmentation to mask objects across multiple frames of a video. The method may include determining an image segmentation mask which masks an object of a frame of a video sequence using the frame and an image segmentation module of a segmentation system. The method further includes determining a mask propagation mask which masks the object of the frame of the video sequence using the frame, a representation of a previous frame of the video sequence, and a mask propagation module of the segmentation system. The method further includes determining a frame mask which masks the object of the frame of the video sequence based on a comparison of the image segmentation mask and the mask propagation mask. The method further includes displaying the frame mask of the video sequence.

Abstract: In implementation of techniques for upsampling a digital material model based on radiances, a computing device implements an upsampling system to receive an input digital material model having a first resolution. The upsampling system generates a bilinearly upsampled texel based on the input digital material model having the first resolution. The upsampling system then generates a texel having a second resolution that is higher than the first resolution based on the bilinearly upsampled texel using a machine learning model trained on training data to generate texels. Based on the texel having the second resolution, the upsampling system generates an output digital material model having a resolution that is higher than the first resolution.

Abstract: In implementation of techniques for progressively generating fine polygon meshes, a computing device implements a mesh progression system to receive a coarse polygon mesh. The mesh progression system generates a fine polygon mesh that has a higher level of resolution than the coarse polygon mesh by decoding the coarse polygon mesh using a machine learning model. The mesh progression system then receives additional data describing a residual feature of a polygon mesh. Based on the additional data, the mesh progression system generates an adjusted fine polygon mesh that has a higher level of resolution than the fine polygon mesh.

Abstract: Decentralized metadata registry techniques are described. In one or more examples, registration data is received from an originating device at a metadata registry system. The registration data identifies a nonfungible token (NFT), metadata describing one or more characteristics associated with entities that receive the nonfungible token, and access rules to control access to the metadata. The nonfungible token, the metadata, and the access rules are registered as part of a metadata registry. An access request is then received from an originating device. The access request identifies the nonfungible token and requests access to metadata associated with the nonfungible token. A determination is made by the metadata registry system that access to the metadata is permitted based on the access rule and access to the metadata by the originating device is permitted.

Abstract: Interactions between a user and an e-commerce platform are automatically guided to increase the chances of a conversion. Previous sequences of interactions (e.g., conversion journeys and non-conversion journeys) with the e-commerce platform are collected, an artificial neural network (ANN) learns how to estimate a safety value a current user state by learning from previous user interactions (e.g., conversion and non-conversion journeys), a software agent of the e-commerce platform applies a current user state of the user to the ANN to determine a current safety value, and the software agent provides content to the user based on the current safety value and the current user state.

Abstract: Systems and methods for content customization are provided. One aspect of the systems and methods includes receiving dynamic characteristics for a plurality of users, wherein the dynamic characteristics include interactions between the plurality of users and a digital content channel; clustering the plurality of users in a plurality of segments based on the dynamic characteristics using a machine learning model; assigning a user to a segment of the plurality of segments based on static characteristics of the user; and providing customized digital content for the user based on the segment.

Abstract: Embodiments of the present disclosure provide, a method, a system, and a computer storage media that provide mechanisms for multimedia effect addition and editing support for text-based video editing tools. The method includes generating a user interface (UI) displaying a transcript of an audio track of a video and receiving, via the UI, input identifying selection of a text segment from the transcript. The method also includes in response to receiving, via the UI, input identifying selection of a particular type of text stylization or layout for application to the text segment. The method further includes identifying a video effect corresponding to the particular type of text stylization or layout, applying the video effect to a video segment corresponding to the text segment, and applying the particular type of text stylization or layout to the text segment to visually represent the video effect in the transcript.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media for training a generative inpainting neural network to accurately generate inpainted digital images via object-aware training and/or masked regularization. For example, the disclosed systems utilize an object-aware training technique to learn parameters for a generative inpainting neural network based on masking individual object instances depicted within sample digital images of a training dataset. In some embodiments, the disclosed systems also (or alternatively) utilize a masked regularization technique as part of training to prevent overfitting by penalizing a discriminator neural network utilizing a regularization term that is based on an object mask. In certain cases, the disclosed systems further generate an inpainted digital image utilizing a trained generative inpainting model with parameters learned via the object-aware training and/or the masked regularization.

Abstract: Methods and systems are provided for facilitating implementation of machine learning models in embedded software. In embodiments, a lean machine learning model, having a limited number of layers, is trained in association with a complex machine learning model, having a greater number of layers. To this end, a complex machine learning model, having a first number of layers, can be trained based on an output generated from a lean machine learning model used as input to the complex machine learning model. Further, the lean machine learning model, having a second number of layers less than the first number of layers, is trained using a loss value generated in association with training the complex machine learning model. Thereafter, the trained lean machine learning model can be provided for implementation in an embedded software.

Abstract: In implementation of techniques for connecting paths based on primitives, a computing device implements a path connection system to receive a first path and a second path displayed in a user interface. The path connection system determines an end section of the first path and a corresponding end section of the second path. Based on the on the end section of the first path, the path connection system identifies a first primitive. Based on the corresponding end section of the second path, the path connection system identifies a second primitive. The path connection system then generates a connection path for display relative to the first path and the second path in the user interface by generating a Bezier curve based on the first primitive and the second primitive.

Abstract: The disclosed method generates helpful training data for a language model, for example, a model implementing a punctuation restoration task, for real-world ASR texts. The method uses a reinforcement learning method using a generative AI model to generate additional data to train the language model. The method allows the generative AI model to learn from real-world ASR text to generate more effective training examples based on gradient feedback from the language model.

Abstract: A method, a system, and a computer program product for analyzing data collected during a randomized controlled experiment to determine an effect of variations of digital content. Determination of the effect includes execution of first and second testing sequences that prompt responses to first and second digital contents, respectively, from users. The testing sequences execute during a predetermined duration of time. Responses to the first and second testing sequences generate first and second test data, respectively. One or more confidence intervals for each first and second test data are generated at a randomly selected time during the predetermined duration of time. A testing metric indicating the effect of the second digital content over the first digital content is determined at the randomly selected time. The testing metric is determined at any time before expiration of the predetermined duration of time.

Abstract: In some embodiments, a multimodal computing system receives a query and identifies, from source documents, text passages and images that are relevant to the query. The multimodal computing system accesses a multimodal question-answering model that includes a textual stream of language models and a visual stream of language models. Each of the textual stream and the visual stream contains a set of transformer-based models and each transformer-based model includes a cross-attention layer using data generated by both the textual stream and visual stream of language models as an input. The multimodal computing system identifies text relevant to the query by applying the textual stream to the text passages and computes, using the visual stream, relevance scores of the images to the query, respectively. The multimodal computing system further generates a response to the query by including the text and/or an image according to the relevance scores.

Abstract: In implementations of systems for searching for images using generated images, a computing device implements a search system to receive a natural language search query for digital images included in a digital image repository. The search system generates a set of digital images using a machine learning model based on the natural language search query. The machine learning model is trained on training data to generate digital images based on natural language inputs. The search system performs an image-based search for digital images included in the digital image repository using the set of digital images. An indication of the search result is generated for display in a user interface based on performing the image-based search.

Abstract: In implementations of systems for generating and applying editing presets, a computing device implements a preset system to detect objects depicted in a digital image that is displayed in a user interface of an application for editing digital content. Input data is received describing an edited region of the digital image and properties of an editing operation performed in the edited region. The preset system identifies a particular detected object of the detected objects based on a bounding box of the particular detected object and an area of the edited region. An additional digital image is edited by applying the properties of the editing operation to a detected object that is depicted in the additional digital image based on a classification of the detected object and a classification of the particular detected object.

Abstract: The present disclosure relates to using end-to-end differentiable pipeline for optimizing parameters of a base procedural material to generate a procedural material corresponding to a target physical material. For example, the disclosed systems can receive a digital image of a target physical material. In response, the disclosed systems can retrieve a differentiable procedural material for use as a base procedural material in response. The disclosed systems can compare a digital image of the base procedural material with the digital image of the target physical material using a loss function, such as a style loss function that compares visual appearance. Based on the determined loss, the disclosed systems can modify the parameters of the base procedural material to determine procedural material parameters for the target physical material. The disclosed systems can generate a procedural material corresponding to the base procedural material using the determined procedural material parameters.

Abstract: Self-consumable portions generation techniques from a digital document are described. The self-consumable portions are generated based on a determination of an amount of resources available at a receiver device that is to receive the digital document. Examples of the resources include an amount of memory resources, processing resources, and/or network resources associated with the receiver device. The self-consumable portions, once generated, are separately renderable at the receiver device.

Abstract: In implementations of systems for generating indications of relationships between electronic documents, a processing device implements a relationship system to segment text of electronic documents included in a document corpus into segments. The relationship system determines a subset of the electronic documents that includes electronic document pairs having a number of similar segments that is greater than a threshold number. The similar segments are identified using locality sensitive hashing. The electronic document pairs are classified as related documents or unrelated documents using a machine learning model that receives a pair of electronic documents as an input and generates an indication of a classification for the pair of electronic documents as an output. Indications of relationships between particular electronic documents included in the subset are generated based at least partially on the electronic document pairs that are classified as related documents.