Abstract: A method in an illustrative embodiment comprises collecting a plurality of audio signals from a plurality of sensors, and combining the plurality of audio signals to generate synthetic audio, wherein the plurality of sensors are located at different locations in the same space, and the plurality of audio signals each have a location identifier. The method further comprises analyzing the synthetic audio to find an audio subset in the plurality of audio signals, and extracting first image features of a first image of the space. The method further comprises modifying the first image features based on the location identifiers of the audio signals in the audio subset to generate second image features, and generating a second image of the space based on the second image features.

Abstract: The present disclosure relates to a method, a device, and a computer program product for threat notification. The method includes receiving an image indicating a first object, wherein the first object includes at least one of a vehicle and a vulnerable road user (VRU). The method further includes detecting a first threat associated with the first object based on processing of the image, and sending, in response to detection of the first threat, a notification about the first threat to at least one of the first object and a second object associated with the first object. In this way, once a threat is detected, a notification can be sent only to a specific road participant such as the first object and the second object, so that the notification can be sent and processed more accurately, and less bandwidth is used, thereby improving the communication efficiency.

Abstract: Embodiments of the present disclosure relate to a method, a device, and a computer program product for determining a node of a decision tree. The method includes determining multiple features of multiple modals corresponding to input information. The method further includes generating a multi-modal feature representation by combining the multiple features of the multiple modals. The method further includes determining a target path in a decision tree that is associated with the multi-modal feature representation, the decision tree comprising multiple nodes. The method further includes determining, in the target path based on the multi-modal feature representation, a target node associated with the input information and used to indicate a question or an answer. This method enables the fusion of feature representations corresponding to input information of different modals to determine a multi-modal feature representation. In this way, it is possible to determine richer and more accurate user intentions.

Abstract: Embodiments of the present disclosure relate to a method, a device, and a computer program product for data augmentation. The method includes generating an image embedding based on an image in an unstructured document, and generating a text embedding based on text in the unstructured document and associated with the image. The method further includes acquiring descriptive information from a storage library based on the generated image embedding and text embedding. The method further includes adding the acquired descriptive information into the unstructured document. In this way, it can be possible not only to understand and analyze the unstructured document across modalities, but also to enrich it with a characterization of multimodal data in the unstructured document, thus increasing the amount and diversity of data.

Abstract: Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for constructing training data. The method includes determining multiple clusters by clustering prompts in a training dataset; and determining, based on multiple cohesion levels of the multiple clusters, multiple sampling probabilities corresponding to the multiple clusters, where the cohesion levels indicate intra-cluster distances in the clusters. The method further includes determining, according to the multiple sampling probabilities, a target cluster for sampling. The method further includes constructing target training data by sampling target prompts from the target cluster.

Abstract: In a method for managing training data in an illustrative embodiment, in response to a determination that new training data is collected by a sensor, the new training data is stored into a collected data stream of a storage pool; in response to a determination that the new training data and historical data stored in a full data stream of the storage pool are refined into refined training data, the refined training data is stored into a refined data stream of the storage pool; and the new training data is stored into the full data stream. In this way, data streams become clear and storage costs are reduced. This greatly reduces the difficulty of developing complex scenarios such as autonomous driving.

Abstract: Techniques for constructing and otherwise managing knowledge graphs in information processing system environments are disclosed. For example, a method comprises the following steps. The method collects data from a plurality of data sources. The method extracts structured data and unstructured data from the collected data, wherein unstructured data is extracted using an unsupervised machine learning process. The method forms a plurality of sub-graph structures comprising a sub-graph structure for each of the data sources based on at least a portion of the extracted structured data and unstructured data. The method combines the plurality of sub-graph structures to form a combined graph structure representing the collected data from the plurality of data sources. The resulting combined graph structure is a comprehensive knowledge graph.

Abstract: The present disclosure relates to a method, a device, and a product for searching data. The method includes: encoding a search input into a first dense vector based on a first multi-modal search model; determining, based on the first dense vector, a distilled data item corresponding to the search input from a distilled dataset corresponding to the first multi-modal search model; encoding, based on the first multi-modal search model, an original data item in an original data subset corresponding to the distilled data item in an original dataset corresponding to the distilled dataset into a second dense vector; and determining, based on the second dense vector, an original data item from the original data subset as a search result corresponding to the search input. The method for searching data according to the present disclosure can improve the efficiency and security of data storage, model reproduction, and multi-modal data management.

Abstract: Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for content recommendation. A method for content recommendation includes determining a similarity between a first recommendation result and a second recommendation result for a content set. The first recommendation result and the second recommendation result are determined based on different recommendation techniques and respectively indicative of a recommendation degree for each content in the content set. The method further includes adjusting the second recommendation result using the similarity. In addition, the method further includes determining a target recommendation result for the content set based on the first recommendation result and the adjusted second recommendation result. In this manner, the accuracy and stability of the finally obtained recommendation results can be improved.

Abstract: Embodiments of the present disclosure relate to a method, a device, and a computer program product for user behavior prediction. In some embodiments, at a client, a first user behavior embedding engine in the client generates behavior prediction information of a target user based on feature information of the target user. The client sends the behavior prediction information of the target user to a server, and receives information about a target item recommended for the target user from the server. Such method enables user privacy-related information to be processed only locally, thereby not only ensuring user privacy and security, but also significantly reducing overall resource overhead.

Abstract: Methods, system, and non-transitory processor-readable storage medium for bug fix system are provided herein. An example method includes executing at least one test on a system using an automated test tool. The bug fix system predicts at least one characteristic of a test related failure that occurred during the execution using test data produced by the automated test tool. The bug fix system predicts at least one source of the test related failure using the test data produced by the automated test tool.

Abstract: Embodiments of the present disclosure provide a method, an electronic device, and a computer program product for data processing. In a method for data processing, a first electronic device processes data based on a first data processing model to generate an initial result. A data size of the initial result is smaller than a data size of the data. The first electronic device sends the initial result to a second electronic device. The initial result is adjusted at the second electronic device and based on a second data processing model to generate an adjusted result. The second electronic device has more computing resources than the first electronic device, the second data processing model occupies more computing resources than the first data processing model, and an accuracy of the adjusted result is higher than that of the initial result.

Abstract: Image processing is described. An example method includes generating a feature extraction layer portion of an image classification model based on a deep neural network (DNN) model, and extracting features of a group of images by using the feature extraction layer portion. The method further includes training an output layer portion of the image classification model according to the features of training images in the group of images and classification labels of the training images. The method further includes generating the image classification model by combining the feature extraction layer portion and the output layer portion. Embodiments of the present disclosure implement a lightweight artificial intelligence (AI) solution on a storage system. An expanded storage system can facilitate the generation of AI applications and assist in training an image classification model more quickly, and the obtained image classification model can also produce high accuracy on a small training set.

Abstract: The present disclosure relates to a method, a device, and a program product for processing sample data in an Internet of Things environment. A method in one embodiment includes: receiving features of the sample data from an encoder deployed in a remote device in the Internet of Things environment; acquiring a category probability corresponding to the sample data based on a classifier deployed in a local device in the Internet of Things environment and the features; and classifying the sample data to a predetermined category in response to determining that the category probability satisfies a first threshold condition. Further, a corresponding device and a corresponding program product are provided. With example implementations of the present disclosure, computing resources of devices in an Internet of Things environment can be fully utilized to process sample data.

Abstract: Embodiments of the present disclosure provide a method, an electronic device, and a computer program product for generating logs. The method comprises: generating, by a generator, a pseudo-log based on a generating condition and random noise; determining, by a discriminator, event types of log messages in a real log and the pseudo-log; determining, by the discriminator, a feature matrix of log sequences in the real log and the pseudo-log based on the event types; determining, by the discriminator, authenticity of the real log and the pseudo-log based on the feature matrix; and training the generator and the discriminator based on the authenticity. The method according to embodiments of the present disclosure may improve the accuracy of a training model for generating pseudo-logs that are sufficiently real.

Abstract: Embodiments of the present disclosure relate to a computer-implemented method, a device, and a computer program product. The method includes extracting respective themes of a set of documents with release time within a first period; determining respective semantic information of the themes and frequencies of the themes appearing in the set of documents; and determining the number of documents associated with the themes within a second period according to a prediction model and based on the semantic information and frequencies of the themes. The second period is after the first period. Embodiments of the present disclosure can better predict the tendency of the themes appearing in the future based on the semantic information and frequencies of the themes.

Abstract: Embodiments of the present disclosure provide a method, an apparatus, an electronic device, and a medium for detecting an abnormality in a network. The method for detecting an abnormality in a network includes acquiring a reference tensor and a target tensor representing traffic in the network, the reference tensor and the target tensor having at least dimensions of source, destination, and time of the traffic. The method further includes determining a target core tensor of the target tensor based on a reference decomposition factor of the reference tensor related to the dimensions of source and destination of the traffic. The method further includes determining that there is an abnormality in the network if a difference between the target core tensor of the target tensor and a reference core tensor of the reference tensor is greater than a preset value.

Abstract: Embodiments of the present disclosure relate to a method for generating a cross-modality encoder. The method includes: extracting a sample video feature of a sample video, a sample audio feature of a sample audio, and a sample text feature of a sample text, wherein contents of the sample video, the sample audio, and the sample text are associated with each other. The method further includes: generating, by the cross-modality encoder, a first cross-modality sample feature by combining the sample video feature, the sample audio feature, and the sample text feature; generating a scrambled cross-modality sample feature by performing a scrambling operation on the cross-modality feature; generating by the cross-modality encoder a second cross-modality sample feature by encoding the scrambled cross-modality sample feature; and tuning the cross-modality encoder based on the first cross-modality sample feature and the second cross-modality sample feature.

Abstract: Embodiments relate to a method, an electronic device, and a program product for training an encoder and processing data. The method includes inputting sample point cloud data for an object to an encoder to obtain encoded data for the object, and determining, by transforming the encoded data, a plurality of invariant portions for the object and a plurality of variable portions for the object, an invariant portion in the plurality of invariant portions indicating an invariant feature of the object and a variable portion in the plurality of variable portions indicating a variable feature of the object. The method further includes determining, based on the plurality of invariant portions and the plurality of variable portions, a similarity loss and a spatial loss for the sample point cloud data, and adjusting, based on the similarity loss and the spatial loss, a parameter of the encoder to obtain a trained encoder.

Abstract: Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for generating a virtual image. The method includes extracting an audio feature of an audio input of a target object; and acquiring an expression parameter and a pose parameter associated with the target object based on the audio feature. The method further includes generating, based on the audio feature, auxiliary information related to a texture for at least a portion of the target object and a geometric shape of at least a portion of the target object. The method further includes generating a virtual image of the target object based on the expression parameter, the pose parameter, and the auxiliary information.