User controlled task execution with task persistence for assistant systems

Patent metadata

Field	Value
Publication number	US-11823289-B2
Application number	US-202218061910-A
Country	US
Kind code	B2
Filing date	Dec 5, 2022
Priority date	Oct 18, 2019
Publication date	Nov 21, 2023
Grant date	Nov 21, 2023

Abstract

In one embodiment, a method includes receiving a first user request at a client system to suspend a first task being executed by an assistant system operating on the client system, suspending the execution of the first task responsive to the first user request, receiving a second user request at the client system, determining that the second user request is a request to resume the suspended first task based on user interactions with the assistant system with respect to one or more entities associated with the first task, and presenting a prompt to resume the first task at the client system.

First claim

What is claimed is:

1. A method comprising, by a client system:
receiving, at the client system, a first user request to suspend a first task being executed by an assistant system operating on the client system;
suspending the execution of the first task responsive to the first user request;
receiving, at the client system, a second user request;
determining, based on user interactions with the assistant system with respect to one or more entities associated with the first task, that the second user request is a request to resume the suspended first task; and
presenting, at the client system, a prompt to resume the first task.

2. The method of claim 1 , further comprising:
determining, by a natural-language understanding module, the first user request is associated with a meta-intent to suspend the first task.

3. The method of claim 1 , further comprising:
determining, based on a current dialog state and contextual information associated with the first user request, that the first task is associated with the first user request.

4. The method of claim 1 , further comprising:
determining, by a natural-language understanding module, the second user request is associated with a meta-intent to resume the first task.

5. The method of claim 1 , further comprising:
determining, based on a current dialog state and contextual information associated with the second user request, that the suspended first task is associated with the second user request.

6. The method of claim 1 , wherein the first task comprises one or more sub-tasks, wherein the method further comprises:
storing a dialog state for the first task in response to the first user request, wherein the dialog state comprises results of executed sub-tasks of the first task and a status indicating the first task is suspended.

7. The method of claim 6 , wherein storing the dialog state for the first task comprises:
allocating a particular portion of the dialog state to the first task, wherein the particular portion is associated with one or more of a task identifier or a task name associated with the first task.

8. The method of claim 7 , further comprising:
accessing the dialog state for the first task.

9. The method of claim 8 , wherein accessing the dialog state for the first task comprises:
identifying the particular portion of the dialog state based on one or more of the task identifier or the task name associated with the first task; and
retrieving the particular portion of the stored dialog state, wherein the retrieved portion comprises the dialog state for the first task.

10. The method of claim 8 , wherein accessing the dialog state for the first task is based on one or more task policies.

11. The method of claim 8 , wherein the prompt comprises a summary of the first task, wherein the summary is generated based on the accessed dialog state.

12. The method of claim 11 , wherein the summary is generated by a natural-language generation (NLG) module based on one or more of a language-template or a language model.

13. The method of claim 6 , wherein the dialog state for the first task is stored in a dialog state stack, wherein the storing of the dialog state in the dialog state stack is based on a priority associated with the first task.

14. The method of claim 6 , wherein storing the dialog state for the first task is based on one or more task policies.

15. The method of claim 1 , wherein the first or second user requests are based on one or more of a textual input, an image input, a video input, an audio input, or a gesture input.

16. The method of claim 1 , wherein the first user request is from a first user, wherein the method further comprises:
determining, based on one or more of contextual information associated with the first user or a policy associated with the first task, that the first user needs to be reminded of the suspended first task; and
presenting, at the client system, a notification reminding the first user of the suspended first task.

17. The method of claim 1 , further comprising:
receiving, at the client system, a third user request to execute a second task;
executing the second task; and
resuming the first task based on execution results of the second task.

18. The method of claim 1 , further comprising:
generating a task-function based on the suspension and resuming of the first task, wherein the task-function comprises one or more default intents and one or more slot-values for one or more slots associated with the first task, respectively, and wherein the task-function is associated with a task alias;
receiving, at the client system, a user call from a user, wherein the user call comprises the task alias;
presenting, at the client system, a summary of the task-function based on the default intents and the slot-values; and
executing the first task based on the default intents and the slot-values upon receiving a confirmation from the user.

19. One or more computer-readable non-transitory storage media embodying software that is operable when executed to:
receive, at a client system, a first user request to suspend a first task being executed by an assistant system operating on the client system;
suspend the execution of the first task responsive to the first user request;
receive, at the client system, a second user request;
determine, based on user interactions with the assistant system with respect to one or more entities associated with the first task, that the second user request is a request to resume the suspended first task; and
present, at the client system, a prompt to resume the first task.

20. A system comprising: one or more processors; and a non-transitory memory coupled to the processors comprising instructions executable by the processors, the processors operable when executing the instructions to:
receive, at a client system, a first user request to suspend a first task being executed by an assistant system operating on the client system;
suspend the execution of the first task responsive to the first user request;
receive, at the client system, a second user request;
determine, based on user interactions with the assistant system with respect to one or more entities associated with the first task, that the second user request is a request to resume the suspended first task; and
present, at the client system, a prompt to resume the first task.

Assignees

• Meta Platforms Tech Llc

Inventors

• Liu Xiaohu
• Wang Zhiguang
• Subba Rajen

Classifications

• G06Q10/40 Primary

  Business processes related to social networking or social networking services · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06N3/098

  Distributed learning, e.g. federated learning · CPC title

• G06Q10/1093

  Calendar-based scheduling for persons or groups · CPC title

• G10L2015/223

  Execution procedure of a spoken command · CPC title