Query response generation

1 . At least one processor comprising:

one or more circuits to:

obtain input data from a device, the input data representative of a query to a conversational artificial intelligence (AI) application;

determine, using one or more classifier models, one or more classes of one or more slots associated with the query and an intent expressed in the query, the intent representing a user requested instruction to retrieve one or more slot values that a user previously requested be stored for the one or more slots;

based at least on the intent and the one or more classes indicating that data representing the one or more slot values is included in at least one of one or more transcripts or one or more logs of user interactions stored in at least one unstructured data structure, cause a search for the one or more slot values in a structured data structure of slot values derived from the user interactions stored in the at least one unstructured data structure; and

generate a response to the query that includes the one or more slot values.

2 . The at least one processor of claim 1 , wherein the at least one unstructured data structure comprises a document associated with the user, the document being populated based at least on one or more prior interactions between the user and the conversational AI application including the user requesting that the one or more slot values be stored for the one or more slots.

3 . The at least one processor of claim 1 , wherein the slot values in the structured data structure include key-value pairs extracted from the at least one unstructured data structure, a key of the key-value pairs represents a class of the one or more classes and a value for the key represents a slot value of the one or more slot values, and the search includes retrieving, using the class, the slot value using the class as the key.

4 . The at least one processor of claim 1 , wherein the one or more circuits are to determine, using the intent and the one or more classes, whether the query is a personalization query or a non-personalization query, wherein the search is caused based at least on the determination being that the query is the personalization query.

5 . The at least one processor of claim 1 , the structured data structure includes short term memory of the conversational AI application, the at least one unstructured data structure includes long term memory of the conversational AI application, and the one or more circuits are to, based at least on the intent, store the one or more slot values in the short term memory.

6 . The at least one processor of claim 1 , wherein the data representing the one or more slot values were stored in the at least one unstructured data structure based at least on the user previously requesting the one or more slot values to be stored for the one or more slots.

7 . The at least one processor of claim 1 , wherein the one or more classifier models include a classifier that is trained to jointly classify the intent of the query and the one or more classes of the one or more slots.

8 . The at least one processor of claim 1 , wherein the conversational AI application is at least one of a conversation AI platform, an in-cabin application of a vehicle, a food or beverage ordering application, a computer aided design (CAD) application, a customer service application, a web service application, a smart speaker or smart display application, a retail application, a financial application, or a food service application.

9 . The at least one processor of claim 1 , wherein the at least one processor is comprised in at least one of:

a control system for an autonomous or semi-autonomous machine;

a perception system for an autonomous or semi-autonomous machine;

a system for performing simulation operations;

a system for performing deep learning operations;

a system implemented using an edge device;

a system implemented using a robot;

a system incorporating one or more virtual machines (VMs);

a system implemented at least partially in a data center;

a system including a collaborative creation platform for three-dimensional (3D) content; or

a system implemented at least partially using cloud computing resources.

10 . A system comprising:

one or more processing units comprising processing circuitry to:

receive a query directed to a conversational artificial intelligence (AI) platform;

determine, using one or more classifier models, a tag associated with one or more slots of the query, and an intent expressed in the query, the intent representing a user requested instruction to retrieve one or more slot values that a user previously requested be stored for the one or more slots;

determine, in response to the query and based at least on the intent and the tag indicating that data representing the one or more slot values is included in at least one of one or more transcripts or one or more logs of user interactions stored in at least one unstructured data structure, that the one or more slot values are stored in a structured data structure of slot values derived from the user interactions stored in the at least one unstructured data structure;

obtain, based at least on the determination that the one or more slot values are stored in the structured data structure, the one or more slot values from the structured data structure; and

generate a response to the query that includes the one or more slot values.

11 . The system of claim 10 , wherein the processing circuitry is further to determine whether to store data representing the query and the response in the at least one unstructured data structure based at least on the intent.

12 . The system of claim 10 , wherein the one or more slot values are derived from the at least one unstructured data structure based at least on one or more confidence values associated with the one or more slot values.

13 . The system of claim 10 , wherein the tag is determined by a classifier of the one or more classifier models that is trained to classify a slot of the one or more slots with a class that represents a type of relationship between the slot and the user.

14 . The system of claim 10 , wherein the system is comprised in at least one of:

a control system for an autonomous or semi-autonomous machine;

a perception system for an autonomous or semi-autonomous machine;

a system for performing simulation operations;

a system for performing deep learning operations;

a system implemented using an edge device;

a system implemented using a robot;

a system incorporating one or more virtual machines (VMs);

a system implemented at least partially in a data center;

a system including a collaborative creation platform for three-dimensional (3D) content; or

a system implemented at least partially using cloud computing resources.

15 . A method comprising:

storing data representing slot values included in queries between a user and a conversational artificial intelligence (AI) agent in at least one unstructured data structure based at least on determining the queries express user requested instructions to store the slot values for slots;

determining, in response to a query directed to the conversational AI agent and using one or more classifiers, one or more classes of one or more slots associated with the query and an intent expressed in the query, the intent representing a user requested instruction to retrieve one or more slot values for the one or more slots from the data;

based at least on the intent and the one or more classes indicating that the data representing the one or more slot values is included in at least one of one or more transcripts or one or more logs of user interactions stored in the at least one unstructured data structure, causing a search for the one or more slot values in a structured data structure of the slot values derived from the user interactions stored in the at least one unstructured data structure; and

generating a response to the query that includes the one or more slot values.

16 . The method of claim 15 , wherein the method further comprises determining domain information associated with the query based at least on the at least one unstructured data structure.

17 . The method of claim 15 , wherein the user interactions correspond to at least one of video, audio, or text generated by at least one of a camera, a microphone, or an input device of a user device.

18 . The method of claim 15 , wherein the structured data structure comprises a JavaScript Object Notation (JSON) data object.

19 . The method of claim 15 , wherein the at least one unstructured data structure comprises a document including a set of interactions between the user and the conversational AI agent.

20 . The method of claim 15 , wherein the query corresponds to a plurality of domains.