Abstract: A computerized method trains and uses a multimodal fusion transformer (MFT) model for content moderation. Language modality data and vision modality data associated with a multimodal media source is received. Language embeddings are generated from the language modality data and vision embeddings are generated from the vision modality data. Both kinds of embeddings are generated using operations and/or processes that are specific to the associated modalities. The language embeddings and vision embeddings are combined into combined embeddings and the MFT model is used with those combined embeddings to generate a language semantic output token, a vision semantic output token, and a combined semantic output token. Contrastive loss data is generated using the three semantic output tokens and the MFT model is adjusted using that contrastive loss data. After the MFT model is trained sufficiently, it is configured to perform content moderation operations using semantic output tokens.

Abstract: To avoid inconsistencies caused by clock skew across data storage tiers, a point-in-time restore service creates a restore account, restores a partition of the database to its state at the designated point-in-time, including data from that time. A restored value is read from the data and a list is retrieved of segments written to a disaggregated storage tier by the existing partition. The disaggregated storage tier includes a first segment with a first value and a second segment with a second value, the values indicating their creation order. A first comparison is made between the restored value and the first value, and a second comparison is made between the restored value and the second value. Based on these comparisons, the first segment but not the second segment is copied to the restore account.

Abstract: Example solutions for providing route aggregation as a service (RaaS) are disclosed. A RaaS server receives original routing tables from routers in a wide area network (WAN), aggregates the routes by identifying common prefixes, and returns more compact aggregated routing tables to the routers. The routers use the aggregated routing tables to efficiently route data traffic in the WAN. Multiple resiliency approaches are disclosed, in order to prevent the routers from flushing the aggregated routes in the event that the RaaS server experiences an outage and the border gateway protocol (BGP session between each router and the RaaS server is lost. Example resiliency measures includes the use of a long lived stale time (LLST), reaching a length of perhaps weeks, during which the routers continue to use the aggregated routing tables, and the use of mirroring the aggregated routing tables in other RaaS.

Abstract: Natural language to database query translation optimization is disclosed. Due to the flexibility of natural language and database query languages, translations may differ in form, yet still be valid. A natural language query is translated to a first database query in a database query language using a query translator. A first and second database query result are generated by performing a first and second database query on a database. The second database query differs from the first database query in form (e.g., order of the query parameters), but is functionally similar to the first database query, such that an ideal query translation would produce equivalent database query results (e.g., same retrieved information, even if in a different order). If the first database query result matches the second database query result, training reinforcement is provided, but if not, a training adjustment is provided for the query translator.

Abstract: Aspects of the disclosure identify a pattern of features associated with a computing component indicating an increased probability that the component incurs an uncorrectable error. Historical component data, component features, server data, and server services are utilized to identify the patterns that correlate to an increased probability that the component incurs an uncorrectable error. For example, this data is used as input into a machine learning platform. Proactive and/or mitigating actions that reduce the probability of an uncorrectable error or its negative effects are presented and/or implemented to minimize or eliminate disruption in cloud computing services.

Abstract: In various examples there is a computer-implemented method performed by a rendering device. A depth image is rendered from a 3D model according to a predicted pose. The depth image is divided into a plurality of tiles. An error map is computed comprising, for each tile of the plurality of tiles, a planar deviation error value that represents a geometric complexity of the tile and comprises an extent to which a geometry of the tile deviates from a plane. The error map, or information derived from the error map, is sent to a display device.

Abstract: Systems and methods for inferring the favorite authors and collaborators of a user are provided. The method includes storing an enterprise authorship database, identifies a document accessed by a user identifier (UID), identifying contributors to the identified document, generating a contribution score for each contributor of the identified contributors, based on each generated contribution score, determining one or more authors of the identified document, creating an authorship record by associating UIDs of the determined one or more authors with the identified document, and updating the enterprise authorship database with the created authorship record, wherein the authorship record is used to identify a relationship between the UID and the determined one or more authors and generate a recommendation for the UID based on the identified relationship.

Abstract: The disclosure herein describes enabling menuless operations using spatially aware tags with virtual geo-fenced boundaries. A tag manager implemented on a processor obtains movement data associated with an ultrawide band (UWB) enabled initiator device within a selected geo-fenced zone. The movement data is generated by one or more UWB enabled responder(s) within the selected geo-fenced zone. The movement data includes data describing 3D movements of the initiator device within the selected geo-fenced zone. A tag manager identifies a sequence of 3D movements of the initiator device using the movement data. The tag manager maps the sequence of 3D movements to zone-specific and/or user-specific action(s) using a menuless operations mapping table. The tag manager triggers a target computing device within the selected geo-fenced zone to perform the matched action(s). The target computing device can include an internet of things (IoT) device, a computing device, or the initiator device itself.

Abstract: A method, computer program product, and computing system for reducing communication session downtime includes transmitting first signals at a predetermined time interval of X seconds; upon initiation of a reboot operation of a first device associated with the processor, decreasing the predetermined time interval from X seconds to Y seconds; transmitting the first signals at the time interval of Y seconds; and once the reboot of the driver is complete, increasing the predetermined time interval from Y seconds to X seconds.

Abstract: Database management techniques can gauge real-time caching states of no-schema-based (noSQL) databases (DBs) during runtime operation by maintaining a log table to reflect modifications to the noSQL DB. The log table is maintained by adding row entries to reflect modifications to the noSQL DB and therefore does not requiring querying the noSQL DB. The log table may then be deduplicated to remove stale entries and queried to generate a cost table reflecting a real-time caching state of the noSQL DB.

Abstract: The present disclosure provides methods, systems and storage media for conducting a security review of a system. Certain examples relate to the use of trained generative AI to generating a root security query using a machine learning (ML) generator, based on a system description. A security requirement associated with the root security query is extracted, and an indication of the root security query is output at a user interface. A user input is received in response, and the ML generator generates a follow-up request that is output via the user interface. A second user input is received in response to the follow-up request, and the ML generator then determines that the security requirement is not satisfied by the target system.

Abstract: Disclosed solutions for improved machine learning (ML) employ knowledge balancing self-distillation with adaptive mutual information (AMI). Examples include: for a neural network (NN) having a plurality of modules, determining a task objective for at least a final module of the plurality of modules; for the NN, determining a balancing objective using at least an output of the final module and an output of a first intermediate module of the plurality of modules; determining an overall objective, wherein determining the overall objective comprises combining the task objective with the balancing objective; and adjusting weights of the NN to minimize the overall objective. Balancing information may combine mutual information (between an intermediate module output and the output of the final module) with self-information (for the intermediate module output) to produce AMI. Adjusting weights of the NN during training, using the AMI, results in knowledge balancing self-distillation.

Abstract: A computer-implemented method, computer program product, and computing system for synchronizing machine vision and audio is executed on a computing device and includes obtaining encounter information of a user encounter, wherein the encounter information includes machine vision encounter information and audio encounter information. The machine vision encounter information and the audio encounter information are temporally-aligned to produce a temporarily-aligned encounter recording.

Abstract: A computerized method for summarizing digital content based on a query from a user is described. An inference of the query is used to identify a website that includes non-structured content. The most relevant media within the website is identified based on the inference and content from the most relevant media is extracted. Using the inference, semantic summaries are generated from the extracted content, and an aggregation of the semantic summaries are presented to the user.

Abstract: A method, computer program product, and computing system for securely transmitting voice signals. A speech signal including a content component and a speaker component of a first voice is received at an encoder. The speaker component of the speech signal is processed, using machine learning, to generate a speaker embedding. The content component of the voice signal is processed, using machine learning and based at least on the speaker embedding, to generate a content embedding having minimized speaker information. The content embedding is transmitted to a decoder for restoring the received speech signal.

Abstract: Determining a location of an entity comprises: receiving a query comprising a 2D image depicting an environment of the entity; searching for a match between the query and a 3D map of the environment. The 3D map comprising a 3D point cloud, the match indicating the location of the entity in the environment. Searching for the match comprises: extracting descriptors from the 2D image referred to as image descriptors; extracting descriptors from the 3D point cloud referred to as point cloud descriptors; correlating the image descriptors with the point cloud descriptors to produce correspondences, wherein a correspondence is an image descriptor corresponding to a point cloud descriptor; estimating, using the correspondences, the location of the entity.

Abstract: Example solutions for image paragraph captioning use a first vision language model to generate visual information (comprising text) for an image. The visual information may include tags, an initial image caption, and information on objects within the image (e.g., further tags and captions, and object attributes and locations within the image). In some examples, the visual information further includes visual clues. A generative language model generates a plurality of image story caption candidates (e.g., descriptive paragraphs) from the visual information. A second vision language model evaluates the plurality of image story caption candidates and selects a caption as the final output caption.

Abstract: This disclosure provides a filtering mechanism to manage anomalous security data items. An anomalous security data item is provided to an analysis engine (such as a Large Language Model (LLM) or another form of generative language model) for interpretation. By curating a selection of one or more relevant non-anomalous security data items to provide with the anomalous data item, the filtering mechanism enables the analysis engine to perform with increased accuracy, without requiring the analyst engine to process large numbers of data items to ascertain their relevance to the anomalous security data item.

Abstract: A computerized method for processing packets in parallel using a reorder queue is described. As packets are received, it is determined which packet is the first packet in the flow. Once identified, the first packet in the flow, along with all other packets in the flow, is sent to the reorder queue while only a copy of the first packet in the flow is sent to a packet scheduler for processing on one of a plurality of processing cores. After the copy of the first packet in the flow is processed, the reorder queue releases each of the packets for the flow from the reorder queue in an order in which they were received. Thereafter, each of the packets released from the reorder queue are processed based on the processing of the copy of the first packet.

Abstract: A computerized method performs fusion of segmentation masks to generate a refined segmentation mask. A plurality of segmentation masks, each including a group of segments, is obtained for a geographic area. For each segmentation mask, a subgroup of segments is filtered from the group of segments of the segmentation mask based on areas of the subgroup of segments. Pairs of segments of the filtered subgroup of segments are matched to form matched segment groups in the geographic area. Representative segments of the geographic area are selected from the matched segment groups to generate a refined segmentation mask including the selected representative segments. A segment-specific action is performed on at least one of the selected representative segments of the generated refined segmentation mask of the geographic area. In some examples, the segment-specific action is for precision farming in agriculture domain.