Abstract: Examples provide a self-supervised language model for document-to-document similarity scoring and ranking long documents of arbitrary length in an absence of similarity labels. In a first stage of a two-staged hierarchical scoring, a sentence similarity matrix is created for each paragraph in the candidate document. A sentence similarity score is calculated based on the sentence similarity matrix. In the second stage, a paragraph similarity matrix is constructed based on aggregated sentence similarity scores associated with the first candidate document. A total similarity score for the document is calculated based on the normalize the paragraph similarity matrix for each candidate document in a collection of documents. The model is trained using a masked language model and intra-and-inter document sampling. The documents are ranked based on the similarity scores for the documents.

Abstract: The disclosure herein describes a system for interpreting text-based similarity between a seed item and a recommended item selected by a pre-trained language model from a plurality of candidate items based on semantic similarities between the seed item and the recommended item. The system analyzes similarity scores and contextual paragraph representations representing text-based descriptions of the seed item and recommended item to generate gradient maps and word scores representing the text-based descriptions. A model for interpreting text-based similarity utilizes the calculated gradients and word scores to match words from the seed item description with words in the recommended item description having similar semantic meaning. The word-pairs having the highest weight are identified by the system as the word-pairs having the greatest influence over the selection of the recommended item from the candidate items by the original pre-trained language model.

Abstract: Semantic search techniques via focused summarizations are described. For example, a search query is received for a text-based content item in a data set comprising a plurality of text-based content items. A first feature vector representative of the search query is obtained. A respective semantic similarity score is determined between the first feature vector and each of a plurality of second feature vectors. Each of the second feature vectors is representative of a machine-generated summarization of a respective text-based content item. The machine-generated summarization comprises a plurality of multi-word fragments that are selected from the respective text-based content item via a transformer-based machine learning model. A search result is provided responsive to the search query.

Abstract: A comparison engine performs item similarity comparisons. A source item and one or more candidate items are input into a triplet-trained machine learning model trained using training data including triplets of anchor elements, positive elements, and negative elements. Each triplet corresponds to an item included in the training data. The anchor elements and the positive elements are included in the corresponding item. The negative element is included in a different item in the training data. A similarity score between the source item and each of the one or more candidate items is generated from the triplet-trained machine learning model.

Abstract: K-space data obtained from a magnetic resonance imaging scan where motion was detected is split into two parts in accordance with the timing of the motion to produce first and second sets of k-space data corresponding to different poses. Sub-images are reconstructed from the k first and second sets of k-space data, which are used as inputs to a deep neural network which transforms them into a motion-corrected image.

Abstract: A computing system scores importance of a number of tokens in an input token sequence to one or more prediction scores computed by a neural network model on the input token sequence. The neural network model includes multiple encoding layers. Self-attention matrices of the neural network model are received into an importance evaluator. The self-attention matrices are generated by the neural network model while computing the one or more prediction scores based on the input token sequence. Each self-attention matrix corresponds to one of the multiple encoding layers. The importance evaluator generates an importance score for one or more of the tokens in the input token sequence. Each importance score is based on a summation as a function of the self-attention matrices, the summation being computed across the tokens in the input token sequence, across the self-attention matrices, and across the multiple encoding layers in the neural network model.

Abstract: A relational similarity determination engine receives as input a dataset including a set of entities and co-occurrence data that defines co-occurrence relations for pairs of the entities. The relational similarity determination engine also receives as input side information defining explicit relations between the entities. The relational similarity determination engine jointly models the co-occurrence relations and the explicit relations for the entities to compute a similarity metric for each different pair of entities within the dataset. Based on the computed similarity metrics, the relational similarity determination engine identifies a most similar replacement entity from the dataset for each of the entities within the dataset. For a select entity received as an input, the relational similarity determination engine outputs the identified most similar replacement entity.

Abstract: A method of generating an aggregate saliency map using a convolutional neural network. Convolutional activation maps of the convolutional neural network model are received into a saliency map generator, the convolutional activation maps being generated by the neural network model while computing the one or more prediction scores based on unlabeled input data. Each convolutional activation map corresponds to one of the multiple encoding layers. The saliency map generator generates a layer-dependent saliency map for each encoding layer of the unlabeled input data, each layer-dependent saliency map being based on a summation of element-wise products of the convolutional activation maps and their corresponding gradients. The layer-dependent saliency maps are combined into the aggregate saliency map indicating the relative contributions of individual components of the unlabeled input data to the one or more prediction scores computed by the convolutional neural network model on the unlabeled input data.

Abstract: Examples provide a self-supervised language model for document-to-document similarity scoring and ranking long documents of arbitrary length in an absence of similarity labels. In a first stage of a two-staged hierarchical scoring, a sentence similarity matrix is created for each paragraph in the candidate document. A sentence similarity score is calculated based on the sentence similarity matrix. In the second stage, a paragraph similarity matrix is constructed based on aggregated sentence similarity scores associated with the first candidate document. A total similarity score for the document is calculated based on the normalize the paragraph similarity matrix for each candidate document in a collection of documents. The model is trained using a masked language model and intra-and-inter document sampling. The documents are ranked based on the similarity scores for the documents.

Abstract: Examples provide a self-supervised language model for document-to-document similarity scoring and ranking long documents of arbitrary length in an absence of similarity labels. In a first stage of a two-staged hierarchical scoring, a sentence similarity matrix is created for each paragraph in the candidate document. A sentence similarity score is calculated based on the sentence similarity matrix. In the second stage, a paragraph similarity matrix is constructed based on aggregated sentence similarity scores associated with the first candidate document. A total similarity score for the document is calculated based on the normalize the paragraph similarity matrix for each candidate document in a collection of documents. The model is trained using a masked language model and intra-and-inter document sampling. The documents are ranked based on the similarity scores for the documents.

Abstract: Examples provide a self-supervised language model for document-to-document similarity scoring and ranking long documents of arbitrary length in an absence of similarity labels. In a first stage of a two-staged hierarchical scoring, a sentence similarity matrix is created for each paragraph in the candidate document. A sentence similarity score is calculated based on the sentence similarity matrix. In the second stage, a paragraph similarity matrix is constructed based on aggregated sentence similarity scores associated with the first candidate document. A total similarity score for the document is calculated based on the normalize the paragraph similarity matrix for each candidate document in a collection of documents. The model is trained using a masked language model and intra-and-inter document sampling. The documents are ranked based on the similarity scores for the documents.

Abstract: The disclosure herein describes a system for interpreting text-based similarity between a seed item and a recommended item selected by a pre-trained language model from a plurality of candidate items based on semantic similarities between the seed item and the recommended item. The system analyzes similarity scores and contextual paragraph representations representing text-based descriptions of the seed item and recommended item to generate gradient maps and word scores representing the text-based descriptions. A model for interpreting text-based similarity utilizes the calculated gradients and word scores to match words from the seed item description with words in the recommended item description having similar semantic meaning. The word-pairs having the highest weight are identified by the system as the word-pairs having the greatest influence over the selection of the recommended item from the candidate items by the original pre-trained language model.

Abstract: The disclosure herein describes a recommendation system utilizing a specialized domain-specific language model for generating cold-start recommendations in an absence of user-specific data based on a user-selection of a seed item. A generalized language model is trained using a domain-specific corpus of training data, including title and description pairs associated with candidate items in a domain-specific catalog. The language model is trained to distinguish between real title-description pairs and fake title-description pairs. The trained language model analyzes the title and description of the seed item with the title and description of each candidate item in the catalog to create a hybrid set of scores. The set of scores includes similarity scores and classification scores for the seed item title with each candidate item description and title. The scores are utilized by the model to identify candidate items maximizing similarity with the seed item for cold-start recommendation to a user.

Abstract: A method is provided that includes acquiring coil data from a magnetic resonance imaging device. The coil data includes undersampled k-space data. The method includes processing the coil data using an image reconstruction technique to generate an initial undersampled image. The method includes generating a reconstructed image based on the coil data, the initial undersampled image, and multiple iterative blocks of a residual deep-learning image reconstruction network. A first iterative block of the residual deep-learning image reconstruction network receives the initial undersampled image. Each of the multiple iterative blocks includes a data-consistency unit that preserves the fidelity of the coil data in a respective output of a respective iterative block utilizing zeroed data consistency. The initial undersampled image is added to an output of the last iterative block via a residual connection.

Abstract: A method of reconstructing imaging data into a reconstructed image may include training a generative adversarial network (GAN) to reconstruct the imaging data. The GAN may include a generator and a discriminator. Training the GAN may include determining a combined loss by adaptively adjusting an adversarial loss based at least in part on a difference between the adversarial loss and a pixel-wise loss. Additionally, the combined loss may be a combination of the adversarial loss and the pixel-wise loss. Training the GAN may also include updating the generator based at least in part on the combined loss. The method may also include receiving, into the generator, the imaging data and reconstructing, via the generator, the imaging data into a reconstructed image.

Abstract: The disclosure herein describes a recommendation system utilizing a specialized domain-specific language model for generating cold-start recommendations in an absence of user-specific data based on a user-selection of a seed item. A generalized language model is trained using a domain-specific corpus of training data, including title and description pairs associated with candidate items in a domain-specific catalog. The language model is trained to distinguish between real title-description pairs and fake title-description pairs. The trained language model analyzes the title and description of the seed item with the title and description of each candidate item in the catalog to create a hybrid set of scores. The set of scores includes similarity scores and classification scores for the seed item title with each candidate item description and title. The scores are utilized by the model to identify candidate items maximizing similarity with the seed item for cold-start recommendation to a user.

Abstract: A magnetic resonance imaging (MRI) system includes control and analysis circuitry having programming to acquire magnetic resonance (MR) data using coil elements of the MRI system, analyze the MR data, and reconstruct the MR data into MR sub-images. The system also includes a trained neural network associated with the control and analysis circuitry to transform the MR sub-images into a prediction relating to a presence and extent of motion corruption in the MR sub-images. The programming of the control and analysis circuitry includes instructions to control operations of the MRI system based at least in part on the prediction of the trained neural network.

Abstract: K-space data obtained from a magnetic resonance imaging scan where motion was detected is split into two parts in accordance with the timing of the motion to produce first and second sets of k-space data corresponding to different poses. Sub-images are reconstructed from the k first and second sets of k-space data, which are used as inputs to a deep neural network which transforms them into a motion-corrected image.

Abstract: A system and method for detecting, timing, and adapting to patient motion during an MR scan includes using the inconsistencies between calculated images from different coil-array elements to detect the presence of patient motion and, together with the k-space scan-order information, determine the timing of the motion during the scan. Once the timing is known, various actions may be taken, including restarting the scan, reacquiring those portions of k-space acquired before the movement, or correcting for the motion using the existing data and reconstructing a motion-corrected image from the data.

Abstract: A magnetic resonance imaging (MRI) system includes control and analysis circuitry having programming to acquire magnetic resonance (MR) data using coil elements of the MRI system, analyze the MR data, and reconstruct the MR data into MR sub-images. The system also includes a trained neural network associated with the control and analysis circuitry to transform the MR sub-images into a prediction relating to a presence and extent of motion corruption in the MR sub-images. The programming of the control and analysis circuitry includes instructions to control operations of the MRI system based at least in part on the prediction of the trained neural network.