Walmart Apollo, LLC

Shankara Bhargava

Snehasish Mukherjee

Priyanka Bhatt

Phani Ram Sayapaneni

Navin Naidu

Neeraj Agrawal

Anshika Singh

Cole Warren Dutcher

Muzhou Liang

Akshay Kumar

Saurabh Kumar

Venkata Naga Vamsi Krishna Nandanavan

Sriprabha Gopalan

A method can include receiving a signal from a user device of a user. The method can further include processing, via a machine learning model, user intent labels, wherein: the machine learning model is pre-trained based on historical input data and historical output data associated with multiple users comprising the user, the historical input data comprise historical feature embedding vectors associated with the multiple users, and the historical output data comprise historical intent labels based at least in part on uttered intents of the multiple users. The method can also include processing one or more user intent candidates of the user intent labels. The method can further include processing one or more user interface components for the one or more user intent candidates. Additionally, the method can include transmitting the one or more user interface components to be presented on a user interface executed on the user device of the user. Other embodiments are described.

Personalizing user interface displays in real-time

A system can receive a first set of data. The first set of data can include information indicating a first set of user sessions and for each of the first set of user sessions having an associated summary and a corresponding agent indicated intent. The system can also, based on the first set of data, determine a set of utterances and for each of the set of utterances a corresponding set of intents. Additionally, the system can receive a second set of data. The second set of data including information indicating a second set of user sessions and for each of the second set of user sessions having an associated determined utterance and corresponding interaction of a user. Moreover, the system can validate a corresponding intent of one or more utterances of the set of utterances, based on the second set of data.

Systems and methods for utilizing feedback data

A system for generating compressed product titles that can be used in conversational transactions includes a computing device configured to obtain product title data characterizing descriptive product titles of products available on an ecommerce marketplace and to determine compressed product titles based on the product title data using a machine learning model that is pre-trained using a replaced-token detection task. The computing device also stores the compressed product titles for use during conversational transactions.

Methods and apparatuses for discriminative pre-training for low resource title compression

The disclosed subject matter relates to a system and method for facilitating automated conversations with users over a network. In some embodiment, a disclosed system includes a computing device operably connected to a database, and is configured to receive a plurality of incoming messages from a first user in an asynchronous manner; and identify a user request associated with the plurality of incoming messages. In these embodiments, the computing device determines that the plurality of incoming messages are all incoming messages that are associated with the user request; processes the plurality of incoming messages together; and generates at least one outgoing message as a response to the user request.

Methods and apparatus for exchanging asynchronous messages

Growing community of inventors

Santa Clara, CA, United States of America

Shankara Bhargava