Abstract: Parameters of a first transformer are accessed, and size dimensions of a second transformer that is to be trained and is larger than the first transformer are received. The parameters of the first transformer are linearly transformed using a combination of a width-growth operator and a depth-growth operator, wherein the linear transformation produces a set of new parameters, the set corresponding to the size dimensions of the second transformer. The second transformer is initialized with the set of new parameters.

Abstract: A source task prompt of each of a plurality of source tasks is decomposed as a multiplication of a shared prompt matrix shared across source tasks and a low-rank task-specific matrix. Prompt distillation is performed to transfer multitask knowledge to the shared prompt matrix by distilling knowledge from the source task prompts. Low-rank multiplicative updates are performed to the shared prompt matrix to transfer the multitask knowledge to one or more target tasks. The one or more target tasks (e.g.

Abstract: Detecting data anomalies by receiving a query image, determining a query image viewpoint according to a trained neural radiance field model, generating a 2D reference image according to the neural radiance field model, determining a difference between the query image and the reference image, and highlighting the difference in a presentation of the query image.

Abstract: An example system includes a processor to automatically extract text and images from a document. The processor can automatically generate text bags including a number of nearest texts for each of the extracted images. The processor can then train a multi-modal model based on the automatically generated text bags using a CLIP-MIL loss that computes, for each of the extracted images, a correlation between each of the different texts in the texts bags using a CLIP feature space at each gradient step of the gradient descent-based multiple instance learning (MIL) algorithm.

Abstract: Aspects of the disclosure include methods, systems, and computer program products for generating semantically meaningful question-answer pairs for graph-like charts, such as flowcharts. In one example, a method of implementing a Question Answering (QA) system may comprise generating a synthetic dataset of graph-like chart images. The generating may comprise rendering a plurality of graph-like chart images from a plurality of associated graph data, generating a plurality of question-answer pairs for each of the graph-like chart images, and calculating a plurality of ground truth annotations for each of the plurality of question-answer pairs and associated graph-like chart images from the plurality of associated graph data. The method of implementing the QA system may further comprise training a vision-language architecture on the synthetic dataset to answer questions about the graph-like chart images.

Abstract: An example system includes a processor that can randomly mask tokens using different masks to generate different subsets of masked tokens. The processor can process the different sets of masked tokens via a pretrained masked auto-encoder (MAE) encoder to output intermediate representations. The processor can process the intermediate representations via a pretrained MAE decoder to output reconstructed images. The processor can further compare input image with the output reconstructed images to generate an anomaly score.

Abstract: An example a system includes a processor to receive a model that is a neural network and a number of training images. The processor can train the model using a bridge transform that converts the training images into a set of transformed images within a bridge domain. The model is trained using a contrastive loss to generate representations based on the transformed images.

Abstract: An example system includes a processor to receive detected chart regions in a page of a document. The processor is to produce, via a graphical elements detector, predicted heatmaps and bounding boxes for graphical objects in the detected chart regions. The processor is also to apply chart type specific analysis algorithm to the predicted heatmaps and bounding boxes, to extract tabular chart data. The processor can then generate an output data file and a visualization based on the predicted heatmap and the extracted tabular chart data.

Abstract: An example system includes a processor to receive, a randomly generated alpha-map, a pair of training images, and a pair of training texts associated with the pair of training images. The processor is to generate a blended image based on the randomly generated alpha-map and the pair of training images. The processor is to train a visual language grounding model to separate the blended image into a pair of heatmaps identifying portions of the blended image corresponding to each of the training images using a separation loss.

Abstract: A computing system, computer program product, and computer-implemented method for incremental learning without forgetting for a classification/detection model are provided. The method includes receiving, at a computing system, a classification/detection model including a base embedding space and corresponding base embedding vectors that are based on a base training dataset including base classes. The method also includes expanding the classification/detection model to account for a new training dataset including new classes by lifting the base embedding space to add an orthogonal subspace for the new classes, producing an expanded embedding space and corresponding expanded embedding vectors that are of a higher dimension than the base embedding vectors.

Abstract: Meta-training an artificial neural cell for use in a few-shot learner, wherein the meta-training includes: executing a Neural Architecture Search (NAS) to automatically learn an architecture of the artificial neural cell; training adaptive controllers that are comprised in the architecture of the artificial neural cell, wherein each of the adaptive controllers is configured to adapt the architecture of the artificial neural cell to a few-shot learning task; and regressing the architecture of the artificial neural cell from support data of the few-shot learning task, through the adaptive controllers. Generating the few-shot learner based on the meta-trained artificial neural cell, to form an Artificial Neural Network (ANN).

Abstract: A computerized system and method of training a deep neural network (DNN) is provided. The DNN is trained in a first training cycle using a first training set including first training samples. Each first training sample includes at least one first training image synthetically generated based on design data. Upon receiving a user feedback with respect to the DNN trained using the first training set, a second training cycle is adjusted based on the user feedback by obtaining a second training set including augmented training samples. The DNN is re-trained using the second training set. The augmented training samples are obtained by augmenting at least part of the first training samples using defect-related synthetic data. The trained DNN is usable for examination of a semiconductor specimen.

Abstract: Embodiments may provide learning and recognition of classifications using only one or a few examples of items. For example, in an embodiment, a method of computer vision processing may be implemented in a computer comprising a processor, memory accessible by the processor, and computer program instructions stored in the memory and executable by the processor, the method may comprise training a neural network system implemented in the computer system to classify images into a plurality of classes using one or a few training images for each class and a plurality of associated semantic information, wherein the plurality of associated semantic information is from a plurality of sources and comprises at least some of class/object labels, textual description, or attributes, and wherein the neural network is trained by modulating the training images by sequentially applying the plurality of associated semantic information and classifying query images using the trained neural network system.

Abstract: Embodiments may provide visual recognition techniques that provide improved recognition accuracy and reduced use of computing resources in cases where only a small set of examples is used to train an unlimited number of recognized categories. For example, in an embodiment, a computer-implemented method of visual recognition may comprise generating a plurality of personal embedding models, each personal embedding model including categories relating to a person, and object, or a subject, wherein at least some of the personal embedding models include at least some different categories, training the plurality of personal embedding models using image training data having a limited number of examples of each category, wherein the examples of each category are used to train more than one category in more than one of the personal embedding models, recognizing images from image data using the plurality of personal embedding models, and outputting information relating to the recognized images.

Abstract: There is provided a method of identifying objects in an image, comprising: extracting query descriptors from the image, comparing each query descriptor with training descriptors for identifying matching training descriptors, each training descriptor is associated with a reference object identifier and with relative location data (distance and direction from a center point of a reference object indicated by the reference object identifier), computing object-regions of the digital image by clustering the query descriptors having common center points defined by the matching training descriptors, each object-region approximately bounding one target object and associated with a center point and a scale relative to a reference object size, wherein the object-regions are computed independently of the identifier of the reference object associated with the object-regions, wherein members of each cluster point toward a common center point, and classifying the target object of each object-region according to the referen

Abstract: There are provided system and method of classifying defects in a semiconductor specimen. The method comprises: upon obtaining by a computer a Deep Neural Network (DNN) trained to provide classification-related attributes enabling minimal defect classification error, processing a fabrication process (FP) sample using the obtained trained DNN; and, resulting from the processing, obtaining by the computer classification-related attributes characterizing the at least one defect to be classified, thereby enabling automated classification, in accordance with the obtained classification-related attributes, of the at least one defect presented in the FP image.

Abstract: There are provided system and method of segmentation a fabrication process (FP) image obtained in a fabrication of a semiconductor specimen. The method comprises: upon obtaining a Deep Neural Network (DNN) trained to provide segmentation-related data, processing a fabrication process (FP) sample using the obtained trained DNN and, resulting from the processing, obtaining by the computer segments-related data characterizing the FP image to be segmented, the obtained segments-related data usable for automated examination of the semiconductor specimen. The DNN is trained using a segmentation training set comprising a plurality of first training samples and ground truth data associated therewith, each first training sample comprises a training image; FP sample comprises the FP image to be segmented.

Abstract: There are provided system and method of examining a semiconductor specimen. The method comprises: upon obtaining a Deep Neural Network (DNN) trained for a given examination-related application within a semiconductor fabrication process, processing together one or more fabrication process (FP) images using the obtained trained DNN, wherein the DNN is trained using a training set comprising ground truth data specific for the given application; and obtaining examination-related data specific for the given application and characterizing at least one of the processed one or more FP images. The examination-related application can be, for example, classifying at least one defect presented by at least one FP image, segmenting the at least one FP image, detecting defects in the specimen presented by the at least one FP image, registering between at least two FP images, regression application enabling reconstructing the at least one FP image in correspondence with different examination modality, etc.

Abstract: An inspection system that may include a processor and a memory module; wherein the memory module is configured to store a first image of an area of an object and a second image of the area of the object; wherein the processor is configured to generate a synthetic image of the area of the object, and to compare the synthetic image to the second image to provide defect detection results.