Abstract: A system and method for generating personalized explanations for AI model predictions. Operations may involve using an AI model to make inferences, determining the importance of different factors used in those inferences, and creating customized explanation prompts. Explainable AI techniques may be used to generate these prompts, taking into account the specific features involved in the inference as well as relevant information about the intended audience. A natural language processing component may use these prompts to produce explanations tailored to the target audience, making the AI's decision-making process more understandable. Feedback from users about these explanations may be obtained and used to refine and improve its explanation generation process over time.

Abstract: A method including receiving a processor command to execute a constraint-based one-to-many matching problem including matching a source entity to target entities selected from test entities. A matching machine learning model is executed on the source entity and the test entities to generate scores representing degrees of match between the source entity and corresponding ones of the test entities. The test entities are arranged in an ordered list according to the scores. A greedy optimization algorithm, that is constrained by the constraint to generate combinations of the test entities, is executed. Each of the combinations satisfies the constraint. The method also includes selecting, from among the combinations of the test entities, an optimized combination of test entities from among the combinations of test entities. The optimized combination of test entities is optimized relative to the constraint. The optimized combination of test entities is stored as the target entities.

Abstract: At least one processor can receive at least one preliminary response generated by a machine learning (ML) model having a predetermined level of randomness. The at least one processor can determine at least one transformation applying a new level of randomness, different from the predetermined level of randomness, to the at least one preliminary response. The at least one processor can generate at least one modified preliminary response, the generating comprising applying the at least one transformation to the at least one preliminary response. The at least one processor can replace the at least one preliminary response with the at least one modified preliminary response within the ML model, wherein the ML model generates a final response using the at least one modified preliminary response.

Abstract: A system and method are provided for refining lingual matching tuning.

Abstract: An address encoder and a phone number encoder can be trained on training data including an address dataset, a phone number dataset, and information associating respective addresses in the address dataset with respective phone numbers in the phone number dataset as associated pairs. The training can be by a constrastive learning process such that respective distances between respective pairs of address vectors from the trained address encoder and phone number vectors from the trained phone number encoder are minimized for respective associated pairs. In production, the trained encoders can determine a production distance between a production address and a production phone number, and these results can be used to modify a production computing process.

Abstract: A method for training an ensemble machine learning model. The method includes applying a first language model to a training data set, having emails stored in a non-transitory computer readable storage medium, to split email addresses in the emails into gibberish email addresses and non-gibberish email addresses. The gibberish email addresses include a first text string that the first language model classifies as gibberish. The non-gibberish email addresses include a second text string that the first language model classifies as non-gibberish. The method also includes training a second language model on the gibberish email addresses. The second language model is trained to determine whether the gibberish email addresses are valid or invalid. The method also includes training a third language model on the non-gibberish email addresses. The third language model is trained to determine whether the non-gibberish email addresses are valid or invalid.

Abstract: Certain aspects of the disclosure provide a method for enhanced object detection.

Abstract: A system for vendor deduplication. The system creates embeddings of entity names and creates an initial entity graph comprising entities whose embeddings are related. The initial entity graph includes nodes representing the entity names linked together by edge weights indicating their similarity. The system adjusts the edge weights between the entity names according to transactional data related to the entities to create a final entity graph and merges the entity names as a common entity based on the adjusted edge weights in the final entity graph.

Abstract: A system for providing content-based (e.g., textual-based) recommendations. The system constructs a knowledge graph from campaign content data including nodes representing individual campaigns and edge weights representing text similarity and collaborative consumption between campaigns. The system adjusts the edge weights within the knowledge graph based on the collaborative consumption of customer groups to emphasize common keywords belonging to common customer groups and deemphasize keywords belonging to different customer groups. The system processes a new campaign content to align with interests of a closest customer group of the customer groups identified in the knowledge graph, thereby enabling targeted delivery of the new campaign to customers associated with that group.

Abstract: A system for vendor deduplication. The system creates embeddings of entity names and creates an initial entity graph comprising entities whose embeddings are related. The initial entity graph includes nodes representing the entity names linked together by edge weights indicating their similarity. The system adjusts the edge weights between the entity names according to transactional data related to the entities to create a final entity graph and merges the entity names as a common entity based on the adjusted edge weights in the final entity graph.

Abstract: A method including receiving a command to perform a one-to-many matching task between first and second datasets. The first and second datasets are vectorized into first and second embedded datasets. A first self-attention model is executed on the first embedded dataset to generate a first attention dataset in which each value of a first number of first features of the first dataset is weighted based on each other value of the first number of first features. A second self-attention model is executed on the second embedded dataset to generate a second attention dataset in which each value of a second number of second features of the second dataset is weighted based on each other value of the second number of second features. The first and second attention datasets are combined into a relationship matrix expressing relationships between the first and second features. The method also includes returning the relationship matrix.

Abstract: A method of executing a matching language model to perform a many-to-one matching task. The matching task uses a number of tokens that exceeds a token limit of the matching language model. The number of tokens is reduced to a reduced number of tokens by executing a rule-based application on a target entry and a dataset of entries to output a data subset including fewer entries. The reduced number of tokens are reduced within the token limit by executing a sorting language model on the data subset and the target entry to output candidate matching sets. The candidate matching sets are evaluated as to whether a potential match exists with the target entry. The matching language model is executed on the number of candidate matching sets and the target entry to output a selected matching set matching the target entry. The selected matching set is returned.