Abstract: A method may include finetuning a generative artificial intelligence (AI) model based on training data including finite state machine (FSM) conversation flows and corresponding intents. The method may include finetuning the generative AI model by tuning a set of base weights with a single set of weights that are based on the FSM conversation flows and the intents. The method may include receiving a query corresponding to an FSM conversation flow. The method may include determining that the first query contains a first intent associated with a first state of the FSM conversation flow. The method may include generating a first response to the first query that corresponds with the first state of the first FSM conversation flow. The method may include communicating commands to a service associated with the first FSM conversation flow, the commands corresponding with actions associated with the first intent and the first state.

Abstract: A cloud platform trains a machine-learned entity matching model that generates predictions on whether a pair of electronic records refer to a same entity. In one embodiment, the entity matching model is configured as a transformer architecture. In one instance, the entity matching model is trained using a combination of a first loss and a second loss. The first loss indicates a difference between an entity matching prediction for a training instance and a respective match label for the training instance. The second loss indicates a difference between a set of named-entity recognition (NER) predictions for the training instance and the set of NER labels for the tokens of the training instance.

Abstract: A hierarchical neural network for predicting out of stock products comprises an input layer that receives data from data sources that store disparate datasets having different levels of attribute detail pertaining to products for sale in stores of a retailer. A first level of neural networks processes the data from the data sources into respective learned intermediate vector representations. A second level comprises a concatenate layer that concatenates the learned intermediate vector representations from the second level into a combined vector representation. A third level comprises a feed forward network that receives the combined vector representation and outputs to the retailer an out of stock probability indicating which store and product combinations are likely to have out of stock products over a predetermined timeframe.

Abstract: Database systems and methods are provided for managing usage of large language models (LLMs). One method involves dividing text data into primary chunks using input criteria associated with an LLM service, generating secondary chunks by merging respective pairs of adjacent primary chunks, and inputting a respective secondary chunk to the LLM service when a semantic similarity between a conversational input to a user interface and the respective secondary chunk of the one or more secondary chunks is greater than a threshold. The LLM service generates response data responsive to the conversational input based at least in part on a subset of the text data associated with the respective secondary chunk, and a response is provided to the conversational input at the user interface based at least in part on the response data generated by the LLM service.

Abstract: A method and system for synthetic data generation are provided that receive a schema configuration file in a synthetic data set request from a client application, create a set of worker processes to generate the synthetic data set based on the schema configuration file, upload the generated synthetic data to an analytics platform, and enable the client application to utilize the generated synthetic data in prediction models for the analytics platform.

Abstract: A machine learning model hosted on a cloud platform may be used to proactively predict if a maintenance procedure should be performed for a vehicle. In some examples, to support the prediction, the machine learning model may be connected to a different cloud platform that includes a customer relationship management (CRM) system and receives data from sensors of the vehicle. As such, the cloud platform with the CRM data may transmit the CRM data and the sensor data of the vehicle to the cloud platform hosting the machine learning model to aid in generating the maintenance procedure predictions. Further, the maintenance procedure predictions may also include the generation of a prediction score associated with a maintenance procedure. In some examples, the prediction score may satisfy a prediction score threshold, thus a notification may be transmitted to a computing device that indicates the maintenance procedure to be performed for the vehicle.

Abstract: Methods, systems, apparatuses, devices, and computer program products are described. A system may support a machine learning model for legal clause extraction. The machine learning model may receive, as an input, at least a portion of a document and may output an indication of one or more legal clauses included in the document. To train the model, the system may receive a document and an indication of ground truths (e.g., legal clauses) for the document. The system may determine one-to-one mappings between the legal clauses indicated by the ground truths and the legal clauses indicated by the output of the machine learning model. The system may perform a longest common substring analysis on the one-to-one mappings to determine an accuracy of the machine learning model and may iteratively update the model based on the analysis.

Abstract: Methods, apparatuses, systems, and computer-program products are disclosed. For example, a system may receive, via a cloud-based platform, first user input including a request for generation of the output data object. The system may generate a prompt based on the first user input and a prompt appendix that defines a response format for a plurality of responses to the prompt that are to be generated by a large language model (LLM). The system may transmit the prompt to the LLM and may receive, from the LLM, the plurality of responses formatted in the response format. The system may generate the output data object that comprises the plurality of responses.

Abstract: A method or a system for predicting a likelihood of an occurrence of a transaction. The system accesses a graph including multiple nodes and multiple edges linking the nodes. The multiple nodes include a first type of nodes representing a first type of entities an a second type of nodes representing a second type of entities. The system extract a set of node features for each node, and a set of edge features for each edge. For an edge connecting a first node of the first type and a second node of the second type, the system generates a set of edge embeddings based in part on the node features and edge features, and computes a score based in part on the set of edge embeddings. The score indicates a likelihood of an occurrence of a transaction between the first node and the second node.

Abstract: A method for using a neural network to generate node embeddings and edge embeddings for graphs. The neural network has K layers. The graph includes multiple nodes and edges linking the multiple nodes. The method includes determining a set of node features for the multiple nodes, and determining a set of edge features for the multiple edges. A first layer of the neural network is applied to the node features and the edge features to output a first set of node embeddings and a first set of edge embeddings. A k-th layer of the neural network is applied to (k?1)th set of node embeddings and (k?1)th set of edge embeddings to output a k-th set of node embeddings and a k-th set of edge embeddings, where the (k?1)th set of node embeddings and (k?1)th set of edge embeddings are output from (k?1)th layer of neural network.

Abstract: A cloud platform trains a machine-learned entity matching model that generates predictions on whether a pair of electronic records refer to a same entity. In one embodiment, the entity matching model is configured as a transformer architecture. In one instance, the entity matching model is trained using a combination of a first loss and a second loss. The first loss indicates a difference between an entity matching prediction for a training instance and a respective match label for the training instance. The second loss indicates a difference between a set of named-entity recognition (NER) predictions for the training instance and the set of NER labels for the tokens of the training instance.

Abstract: A hierarchical neural network for predicting out of stock products comprises an input layer that receives data from data sources that store disparate datasets having different levels of attribute detail pertaining to products for sale in stores of a retailer. A first level of neural networks processes the data from the data sources into respective learned intermediate vector representations. A second level comprises a concatenate layer that concatenates the learned intermediate vector representations from the second level into a combined vector representation. A third level comprises a feed forward network that receives the combined vector representation and outputs to the retailer an out of stock probability indicating which store and product combinations are likely to have out of stock products over a predetermined timeframe.

Abstract: Machine learning based models recognize objects in images. Specific features of the object are extracted from the image using machine learning based models. The specific features extracted from the image assist deep learning based models in identifying subtypes of a type of object. The system recognizes the objects and collections of objects and determines whether the arrangement of objects violates any predetermined policies. For example, a policy may specify relative positions of different types of objects, height above ground at which certain types of objects are placed, or an expected number of certain types of objects in a collection.

Abstract: A method and system for synthetic data generation are provided that receive a schema configuration file in a synthetic data set request from a client application, create a set of worker processes to generate the synthetic data set based on the schema configuration file, upload the generated synthetic data to an analytics platform, and enable the client application to utilize the generated synthetic data in prediction models for the analytics platform.

Abstract: Disclosed herein are system, method, and computer program product embodiments for compliance auditing using cloud based computer vision. In one aspect, a system is configured to receive, from a mobile device, a compliance audit request to at least recognize one or more products within the audit image. The system is further configured to select a first object recognition model having a first associated object recognition model identifier from a model selection list based at least on a required object recognition list, wherein the first object recognition model is configured to recognize a first set of object names within the required object recognition list. The system is further configured to request the computer vision system to perform object recognition using the first object recognition model to recognize the first set of object names within the audit image, and transmit audit result information to the mobile device.

Abstract: Machine learning based models recognize objects in images. Specific features of the object are extracted from the image using machine learning based models. The specific features extracted from the image assist deep learning based models in identifying subtypes of a type of object. The system recognizes the objects and collections of objects and determines whether the arrangement of objects violates any predetermined policies. For example, a policy may specify relative positions of different types of objects, height above ground at which certain types of objects are placed, or an expected number of certain types of objects in a collection.

Abstract: Machine learning based models recognize objects in images. Specific features of the object are extracted from the image using machine learning based models. The specific features extracted from the image assist deep learning based models in identifying subtypes of a type of object. The system recognizes the objects and collections of objects and determines whether the arrangement of objects violates any predetermined policies. For example, a policy may specify relative positions of different types of objects, height above ground at which certain types of objects are placed, or an expected number of certain types of objects in a collection.

Abstract: Disclosed herein are system, method, and computer program product embodiments for compliance auditing using cloud based computer vision. In one aspect, a system is configured to receive, from a mobile device, a compliance audit request to at least recognize one or more products within the audit image. The system is further configured to select a first object recognition model having a first associated object recognition model identifier from a model selection list based at least on a required object recognition list, wherein the first object recognition model is configured to recognize a first set of object names within the required object recognition list. The system is further configured to request the computer vision system to perform object recognition using the first object recognition model to recognize the first set of object names within the audit image, and transmit audit result information to the mobile device.

Abstract: Machine learning based models recognize objects in images. Specific features of the object are extracted from the image using machine learning based models. The specific features extracted from the image assist deep learning based models in identifying subtypes of a type of object. The system recognizes the objects and collections of objects and determines whether the arrangement of objects violates any predetermined policies. For example, a policy may specify relative positions of different types of objects, height above ground at which certain types of objects are placed, or an expected number of certain types of objects in a collection.

Abstract: Various embodiments for providing a system for the automatic segmentation and ranking of leads and referrals are described herein. An embodiment operates by receiving historical data including information about prospective customers who purchased one or more products. A set of segments of the prospective customers are identified, the historical data is grouped into the set of segments, and a predictive model for a conversion is generated for each segment based on the grouped historical data. A processor generates two or more predictive scores a new prospective customer, wherein each predictive score is based on the generated predictive model for two or more of the segments to which the new prospective customer belongs. The predictive score for the at least one new prospective customer is ranked along with predictive scores of a plurality of other prospective customers for display for at least one of the two or more segments.