Abstract: A method implements brand engine for extracting and presenting brand data with user interfaces. The method includes receiving a blueprint with a set of structure blocks extracted from a selected content. A structure block of the set of structure blocks includes a set of style parameter requests for a section of the selected content. The method further includes processing the set of structure blocks with a first set of smart blocks to generate a set of scores. A smart block of the first set of smart blocks includes brand data with style parameter selections. The method further includes selecting a second set of smart blocks, for the set of structure blocks, from the first set of smart blocks, using the set of scores. The method further includes presenting the second set of smart blocks with the brand data.

Abstract: A method includes receiving an untransformed transaction including unstructured data. An embedding model generates a vector from the unstructured data. A cluster model matches the vector to a vector cluster. A cluster ID is assigned to the vector. The unstructured data in the untransformed transaction is replaced with the cluster ID to obtain a transformed transaction. A query including the cluster ID and based on the transformed transaction is generated. The query is processed to generate a query result from features of prior transformed transactions. A fraud determination model processes the query result to generate a fraud score for the transformed transaction. The fraud score is presented to a user of a software application. The cluster model is updated to add or delete or modify vector clusters to generate cluster IDs, whereby generating the set of cluster IDs does not affect an input or output of the fraud determination model.

Abstract: Systems and methods for training a machine learning model are disclosed. A system may be configured to obtain a plurality of training samples. The system includes a machine learning model to generate predictions and generate a confidence score for each generated prediction. In this manner, the system is configured to, for each training sample of the plurality of training samples, generate a prediction by a machine learning model based on the training sample and generating a confidence score associated with the prediction by the machine learning model. The system is also configured to train the machine learning model based on the plurality of predictions and associated confidence scores. For example, one or more training samples may be excluded from use in training the machine learning model based on the associated one or more confidence scores (such as the confidence score being less than a threshold).

Abstract: Embodiments disclosed herein generate a strategy insight report for a user's business, leveraging generative artificial intelligence—particularly large language models—and pre-stored data associated with the user. The large language models are used to capture subjective information associated with different insight areas, e.g., strength, weakness, opportunity, and threat (SWOT) of a SWOT model. The captured subjective information is augmented, supplemented, and/or modified by the pre-stored data to generate the strategy insight report. In contrast to conventional results and reports, the disclosed strategy insight report provides a current state of the user's business as well as next steps and recommendations.

Abstract: A method including receiving a digital image including text arranged in a layout. The method also includes generating, by an optical character recognition model, a layout text vector that encodes at least one word in the text of the digital image and a position of the at least one word in the layout of the digital image. The method also includes generating, by a visual encoder model, a visual representation vector embedding a content of the digital image. The method also includes converting both the layout text vector and the visual representation vector into a projected text vector including a digital format suitable for input to a large language model. The method also includes combining, into a prompt, the projected text vector, a system message, and a task instruction. The method also includes generating an output including a key-value pair.

Abstract: Embodiments disclosed herein provide automated account recommendations for a hierarchical account structure. For an incoming transaction data record, a first language model is used to generate a recommended account name that is agnostic to the existing list of accounts. The recommended account name is appended to the existing list of accounts, which is consolidated to remove synonymous accounts. Additionally, a hierarchical relationship between the different accounts in the consolidated list of accounts is determined. A second language model is used to select an account name from the list of accounts. The selected account name along with any hierarchically related account name may be displayed to the user for selection.

Abstract: Systems and methods are provided for solving mathematical word problems using large language models.

Abstract: A contrastive in-context learning protocol for large language models. The protocol includes inputting positive and negative examples to a large language model. Additionally, the large language model may be instructed to analyze the reasons behind the positive examples being positive and the negative examples being negative. The large language model with such contrastive in-context learning can generate specific responses/answers based on user preferences, generally not possible using conventional models.

Abstract: Output sentences of a primary large language model is provided to a criteria model including a second large language model. The criteria model compares the output to a reference source. As a result of comparing, the criteria model generates a first data structure including a first vector. The first vector stores, an evaluation of the output as being consistent or inconsistent with the reference source, and a corresponding reason for the evaluation. The criteria model identifies an inconsistent sentence, in the sentences, that is inconsistent with the reference source. The method also includes rewriting, by a reason improver model including a third large language model, the inconsistent sentence into a consistent sentence. The consistent sentence is consistent with the reference source. The output is modified by replacing the inconsistent sentence in the sentences with the consistent sentence. Modifying generates a modified output. The method also includes returning the modified output.

Abstract: Providing an output of a primary large language model to a criteria model including a second large language model. The criteria model compares each of the sentences to a reference source and generates a first data structure including a first vector. The first vector stores, for each of the sentences, a corresponding evaluation of a given sentence as being consistent or inconsistent with the reference source, and a corresponding reason for the corresponding evaluation of the given sentence. The first data structure is provided to a converter model including a third large language model. The converter model converts the first data structure to a second data structure. The second data structure includes a second vector storing scores indicating a corresponding consistency value for each of the sentences. A metric, indicating an overall consistency of the output with respect to the reference source, is generated from the second data structure.

Abstract: Systems and methods are described for batch materialization of an incremental change data capture (CDC) changeset with full row changes. The primary keys are extracted from the incremental CDC changeset and an indication of the extracted primary keys are broadcast to a plurality of executors. The primary keys may be added to Bloom filter or a plurality of Bloom filters that are broadcast to the executors. Each executor filters a baseline data table based on the extracted primary keys to generate a baseline match dataframe with all primary keys matching the extracted primary keys, and a baseline unmatched dataframe with all primary keys not matching the extracted primary keys. Each executor receives full row changes from a partitioned incremental CDC changeset and combines the changes with the baseline unmatched dataframe to produce a final changed baseline data table.

Abstract: Certain aspects of the present disclosure provide techniques for detecting data entry errors. A method generally includes receiving a string value as user input for a data field, selecting a plurality of reference values previously entered into the data field within a time period, processing, with an embedding model configured to classify an input string value as a valid or invalid entry, the string value and the reference values and thereby generating a first vector as output, computing one or more statistics for the reference values and the string value, creating a second vector based on the one or more statistics, generating a concatenated vector by concatenating the first vector and the second vector, processing, with a classifier model configured to classify the string value as valid or invalid, the concatenated vector and thereby generating a classification output, and taking action based on the classification output.

Abstract: Systems and methods for adapting an onboarding session to a user are disclosed. An example method is performed by one or more processors of a system and includes receiving a transmission over a communications network from a computing device associated with a user of the onboarding system, the transmission including one or more files, extracting data from each of the one or more files using one or more parser plugins, transforming the extracted data into a set of arrays, feeding a prompt including the set of arrays to a large language model (LLM), inferring characteristics of the user based on a response to the prompt from the LLM, mapping the inferred characteristics to a predefined list of system features, and optimizing components of an onboarding session for the user based on the mapping.

Abstract: A method for performing a parity check of a table by a software application may include obtaining, from a data lake, data lake records stored in the table during a time interval, obtaining partitioning information used to partition the table in a database during the time interval, extracting, from the data lake records and for the partitioning information, partition identifiers stored in the table during the time interval, generating a partition-specific database query including a partition identifier, executing the partition-specific database query to obtain database records stored in the table in a partition of the database during the time interval, extracting a subset of the data lake records that include the partition identifier, and performing a parity comparison on the subset of the data lake records and the database records to generate a parity result.

Abstract: Systems and methods for intelligently repairing data are disclosed. An example method is performed by one or more processors of a data quality management (DQM) system and includes receiving a transmission over a communications network from a computing device associated with the DQM system, the transmission including an indication that source data stored in a source database was ingested and stored as target data in a target database at a time of ingestion, comparing, using an advanced DQM algorithm, the target data with the source data, the advanced DQM algorithm including generating a first set of parity results based on changes occurring before the time of ingestion, generating a second set of parity results based on changes occurring after the time of ingestion, and generating differential results based on the first and the second set of parity results, and selectively repairing ones of the changes based on the differential results.

Abstract: Systems and methods for detecting errors in a data transfer uses a machine learning model to identify potential anomalies in the data transfer based on metadata. Mismatches between input data from the data transfer and output data after importing the data transfer may additionally be identified. User review and correction of data errors and potential anomalies identified using the machine learning model may be proactively prompted to ensure any errors or discrepancies are addressed before finalizing the import of the data transfer. User corrections are further used to retrain the machine learning model to enable continuous improvement and learning from the data transfer process.

Abstract: Aspects of the present disclosure provide techniques for providing a graphical user interface for customizable application navigation. Embodiments include displaying a list of pages associated with a software application in a navigation customization screen and receiving selections of two or more pages of the pages as bookmarks. Embodiments include receiving drag and drop input via the navigation customization screen that changes an ordering of the two or more pages within the list of the plurality of pages and receiving a search query comprising a text string. Embodiments include moving one or more pages matching the search query to a top of the list of the pages within the navigation customization screen and displaying an indication in the navigation customization screen that one of the two or more pages also matches the search query without changing the ordering of the two or more pages within the list of the pages.

Abstract: A method implements efficient real time serving of ensemble models. The method includes receiving an input and processing the input with an abridged model to generate a set of component scores and an abridged score. The method further includes processing the set of component scores with a deviation threshold to select one of the abridged score and an ensemble score as an output and presenting the output.

Abstract: Certain aspects of the present disclosure provide techniques for providing smart content to a user of an application. Embodiments include receiving a request from a client for content. The request may include context data. Embodiments include identifying a content template for the content based on the request. Embodiments include identifying a rule associated with the content template. Embodiments include evaluating the rule based on the context data in order to determine a value of a variable. Embodiments include generating personalized content based on the content template and the value of the variable. Embodiments include providing the personalized content to the client.

Abstract: A method including extracting data from disparate data sources. The data includes data pairs including a corresponding data point and a corresponding time associated with the corresponding data point. The method also includes extracting insights from the data at least by identifying a trend in the data pairs. The method also includes forming a model vector including the insights and an additional attribute to the insights. The additional attribute characterizes the insights. The additional attribute includes at least user feedback including a user ranking of a ranked subset of the insights from a user. The method also includes inputting the model vector into a trained insight machine learning model to obtain a predicted ranking of the insights. The method also includes selecting, based on the predicted user ranking, a pre-determined number of insights to form predicted relevant insights. The method also includes reporting the predicted relevant insights.