Method and device for processing user task, equipment and medium

Abstract

The invention provides a user task processing method and device, equipment and a medium. In one method, user input represented in a natural language from a user is received, the user input indicating a task to be performed. A task type of the task is determined based on the user input. And in response to determining that the task type indicates that the execution result of the task cannot be obtained within the predetermined time period, creating a subsequent task of the task so as to provide the execution result of the task by utilizing a machine learning model when the trigger condition of the task is met. By utilizing the exemplary implementation mode of the invention, the user task which cannot obtain the execution result of the task within the preset time period can be processed in a more flexible and effective mode.

Description

Technical Field

[0001] Exemplary implementations of the present disclosure generally relate to processing user tasks, and more particularly to methods, apparatuses, devices, and computer-readable storage media for utilizing machine learning models to process user tasks that do not immediately provide execution results. Background Art

[0002] Machine learning technology has been widely used in a variety of technical fields. In the prior art, machine learning models can understand the user's language input, interact with the user in real time and generate appropriate responses. Machine learning models can usually call various built-in or third-party tools to perform tasks. However, for some tasks, the execution results cannot be obtained within the predetermined time period. At this time, the user needs to maintain a dialogue with the machine learning model, which may prevent the user from performing other tasks, thereby causing unnecessary waste of resources and other problems. At this time, it is expected that such tasks can be handled in a more efficient way. Summary of the invention

[0003] In a first aspect of the present disclosure, a method for processing a user task is provided. In the method, a user input in natural language is received from a user, the user input indicating a task to be performed. A task type of the task is determined based on the user input. In response to determining that the task type indicates that an execution result of the task cannot be obtained within a predetermined time period, a subsequent task of the task is created so as to provide an execution result of the task using a machine learning model when a trigger condition of the task is met.

[0004] In a second aspect of the present disclosure, a device for processing a user task is provided. The device includes: a receiving module configured to receive a user input expressed in a natural language from a user, the user input indicating a task to be performed; a determining module configured to determine a task type of the task based on the user input; and a creating module configured to create a subsequent task of the task in response to determining that the task type indicates that the execution result of the task cannot be obtained within a predetermined time period so as to provide the execution result of the task using a machine learning model when a trigger condition of the task is met.

[0005] In a third aspect of the present disclosure, an electronic device is provided. The electronic device includes: at least one processing unit; and at least one memory, the at least one memory is coupled to the at least one processing unit and stores instructions for execution by the at least one processing unit, and when the instructions are executed by the at least one processing unit, the electronic device executes the method according to the first aspect of the present disclosure.

[0006] In a fourth aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the processor implements the method according to the first aspect of the present disclosure.

[0007] It should be understood that the content described in this content section is not intended to limit the key features or important features of the implementation of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In the following, in conjunction with the accompanying drawings and with reference to the following detailed description, the above and other features, advantages and aspects of various implementations of the present disclosure will become more apparent. In the accompanying drawings, the same or similar figure labels represent the same or similar elements, wherein:

[0009] Figure 1 A block diagram showing an application environment in which an exemplary implementation according to the present disclosure may be implemented;

[0010] Figure 2 A block diagram illustrating a process for processing user tasks according to some implementations of the present disclosure;

[0011] Figure 3 A block diagram showing a process for determining a task type, a trigger condition, and a task content of a task according to some implementations of the present disclosure;

[0012] Figure 4 A block diagram showing creation of a subsequent task according to some implementations of the present disclosure is shown;

[0013] Figure 5 A block diagram showing creation of a subsequent task according to some implementations of the present disclosure is shown;

[0014] Figure 6 A block diagram showing a process of adding preset tasks in the process of creating a digital assistant according to some implementations of the present disclosure;

[0015] Figure 7 A flowchart of a method for processing user tasks according to some implementations of the present disclosure is shown;

[0016] Figure 8 A block diagram showing an apparatus for processing user tasks according to some implementations of the present disclosure; and

[0017] Fig. 9 A block diagram of a device capable of implementing multiple implementations of the present disclosure is shown. DETAILED DESCRIPTION

[0018] The implementation of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain implementations of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be interpreted as being limited to the implementations described herein. On the contrary, these implementations are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and implementations of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0019] In the description of the implementation methods of the present disclosure, the term "including" and similar terms should be understood as open inclusion, that is, "including but not limited to". The term "based on" should be understood as "based at least in part on". The term "an implementation" or "the implementation" should be understood as "at least one implementation". The term "some implementations" should be understood as "at least some implementations". The following may also include other explicit and implicit definitions. As used herein, the term "model" can represent the association relationship between various data. For example, the above-mentioned association relationship can be obtained based on a variety of technical solutions currently known and / or to be developed in the future.

[0020] It is understandable that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws, regulations and relevant provisions.

[0021] It is understandable that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, scope of use, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0022] For example, in response to receiving an active request from a user, a prompt message is sent to the user to clearly prompt the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, application, server, or storage medium that performs the operation of the technical solution of the present disclosure according to the prompt message.

[0023] As an optional but non-limiting implementation, in response to receiving an active request from the user, the prompt information is sent to the user in a manner such as a pop-up window, in which the prompt information can be presented in text form. In addition, the pop-up window can also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0024] It is understandable that the above notification and the process of obtaining user authorization are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that meet relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0025] The term "in response to" as used herein refers to a state in which a corresponding event occurs or a condition is satisfied. It will be understood that the timing of executing a subsequent action executed in response to the event or condition is not necessarily strongly related to the time when the event occurs or the condition is satisfied. For example, in some cases, the subsequent action may be executed immediately when the event occurs or the condition is satisfied; while in other cases, the subsequent action may be executed some time after the event occurs or the condition is satisfied.

[0026] Example Environment

[0027] The machine learning model can understand the user's language input, interact with the user in real time and generate appropriate responses. Figure 1 A block diagram 100 of an application environment in which an exemplary implementation according to the present disclosure may be implemented is shown. Figure 1 As shown, the user 110 can access the machine learning model 140 via the client device 120. Specifically, the user 110 can call the machine learning model 140 via the user interface of the machine learning model 140; alternatively and / or additionally, the machine learning model 140 can be called via a digital assistant 130 (e.g., a client application, etc.) installed at the client device 120.

[0028] Further, the machine learning model 140 can call various built-in or third-party tools to perform tasks. These tools can be functions, APIs, microservices, or other programs that can complete specific tasks. By calling these tools, the machine learning model 140 can perform various tasks, such as obtaining weather forecasts, searching for information, calculating data, etc. When the machine learning model 140 receives a user's request, it can first parse the user's input, determine the user's needs, and then select the appropriate tool to perform the task based on the needs. For example, the machine learning model can call a weather forecast tool to provide weather information, call a search tool to return the expected search results, and so on. The execution process of the task may require the machine learning model to interact with the tool multiple times, such as sending requests, receiving responses, processing results, etc.

[0029] However, current machine learning models mainly rely on real-time conversational interactions, which means that they cannot handle tasks that need to be performed at specific times or under specific conditions. For example, a user may want to receive a weather forecast at 8 a.m. every day, or receive a reminder when the battery level is less than 20%, and so on. The execution results cannot be obtained within the scheduled time period, and the user needs to maintain a dialogue with the machine learning model at all times. In other words, existing machine learning models cannot meet these requirements because they cannot perform tasks without user interaction. This greatly limits the functions and application areas of machine learning models. At this time, it is expected that user tasks that cannot obtain the execution results of the task within the scheduled time period can be handled in a more flexible and effective way.

[0030] Summary of handling user tasks

[0031] In order to at least partially solve the deficiencies in the prior art, according to an exemplary implementation of the present disclosure, a method for processing user tasks is proposed. Figure 2 Describes an overview of an exemplary implementation according to the present disclosure, Figure 2 A block diagram 200 of a process for processing user tasks according to some implementations of the present disclosure is shown. Figure 2 As shown, a user input 210 in natural language may be received from the user 110, and the user input 210 may indicate a task to be performed. For example, an example of the user input 210 may include "please provide the weather forecast for Beijing at 8 o'clock tomorrow morning", "please provide the weather forecast for 100 cities across the country", and so on.

[0032] The task type 220 of the task may be determined based on the user input 210. Here, the task type 220 may include an immediate task and a non-immediate task. For example, an immediate task may represent a task for which the execution result of the task can be obtained within a predetermined time period (e.g., 1 second, 5 seconds, or other lengths), and a non-immediate task may represent a task for which the execution result of the task cannot be obtained within a predetermined time period. If the judgment result at block 230 is "no", a subsequent task 240 of the task may be created.

[0033] It should be understood that the subsequent task 240 may mean that the task is not executed immediately at present, but the execution result of the task is provided by using the machine learning model 140 when the trigger condition of the task is met. As shown in box 250, if the trigger condition of the task is met, the execution result 260 of the task is provided to the user 110. Specifically, assuming that the user input 210 is "please provide the weather forecast for Beijing at 8 o'clock tomorrow morning", the trigger condition can be "8 o'clock tomorrow morning", and when the current time reaches 8 o'clock the next morning, the weather forecast for Beijing is provided to the user 110. For another example, assuming that the user input 210 is "please provide the weather forecast for 100 cities across the country", it takes some time to execute the task at this time, so the trigger condition can be "when the task is completed", and the weather forecast for 100 cities across the country is provided to the user 110.

[0034] Using the exemplary implementation of the present disclosure, the subsequent task 240 can be used to provide the execution result of the task when the trigger condition of the task is met. In this way, the user does not need to always maintain a session with the machine learning model, and the execution result of the task can be provided without the user interacting with the machine learning model.

[0035] Detailed process for handling user tasks

[0036] An overview of an example implementation of the present disclosure has been described, and more details about processing user tasks are described below. According to an example implementation of the present disclosure, a method according to an example implementation of the present disclosure can be performed in a digital assistant 130. Here, the digital assistant 130 can be, for example, a pre-created application that can call the machine learning model 140, and the digital assistant 130 can act as an intermediary between the user and the machine learning model 140, thereby calling the comprehensive processing capabilities of the machine learning model 140 to serve the user 110.

[0037] According to an example implementation of the present disclosure, the digital assistant 130 may receive a user input 210 from a user in a natural language. The task type 110 of determining the task represented by the user input 210 may be performed locally at the digital assistant 130. For example, the task type 110 may be determined by an algorithm such as language analysis. Alternatively and / or additionally, the digital assistant 130 may call upon the capabilities of the machine learning model 140 to determine the task type 220. Here, the machine learning model 140 may be a large language model (LLM).

[0038] Figure 3 A block diagram 300 is shown of a process for determining a task type, triggering conditions, and task content of a task according to some implementations of the present disclosure. Figure 3As shown, in the process of determining the task type 220, a prompt word (e.g., referred to as a first prompt word) 310 can be generated based on the user input 210. Here, the prompt word 310 can indicate that the task type 220 is determined by the machine learning model. The prompt word can be determined based on a variety of methods currently known and / or to be developed in the future. Further, the prompt word 310 can be provided to the machine learning model 140, and a response 320 can be received from the machine learning model 140. Further, the task type 220 can be determined based on the response 320 of the machine learning model to the prompt word 310. In this way, the complexity of the digital assistant 130 can be simplified, and the powerful processing power of the machine learning model 140 can be called to determine the task type in a more accurate manner.

[0039] According to an example implementation of the present disclosure, the powerful capabilities of the machine learning model 140 can be further utilized to determine the triggering conditions and / or task content of the task from the user input 210. Here, the triggering condition can represent the conditions for triggering the provision of the execution result of the task, for example, "tomorrow morning at 8 o'clock", "when the task is completed", etc. in the above example. Further, the task content can define the specific content of the task to be executed, for example, "provide weather forecast for Beijing" and "provide weather forecast for 100 cities across the country" in the above example.

[0040] At this point, a corresponding prompt word 310 can be constructed, and the prompt word 310 can further indicate that the triggering condition and task content of the task are determined by the machine learning model 140. At this point, the response 320 from the machine learning model 140 will include relevant information about the triggering condition and task content, and the triggering condition 330 and task content 332 of the task can be determined based on the response 320. In this way, the complexity of the digital assistant 130 can be simplified, and the powerful processing power of the machine learning model 140 can be called to provide more accurate triggering conditions and task content.

[0041] According to an example implementation of the present disclosure, a subsequent task of a task can be created based on a trigger condition 330 and a task content 332, so that the subsequent task 240 can use the machine learning model 140 to execute the specified task content 332 when the trigger condition 330 of the task is met, thereby providing the execution result of the task. Specifically, in the process of creating a subsequent task of a task based on the trigger condition 330 and the task content 332, another prompt word (for example, referred to as a second prompt word) can be generated for the task content 332. Here, the prompt word can indicate that the task content is executed by the machine learning model 140. Further, the subsequent task can be configured to provide the second prompt word to the machine learning model based on the trigger condition.

[0042] See also Figure 4Describes more details about creating a subsequent task. Figure 4 A block diagram 400 of creating a subsequent task according to some implementations of the present disclosure is shown. Figure 4 As shown, a conversation 410 may be provided between the user and the digital assistant, with the user represented on the left side of the conversation 410 and the digital assistant represented on the right side. For example, user input 420 may represent: "Please provide Beijing's weather forecast at 8 o'clock tomorrow morning." At this time, the digital assistant 130 may call upon the capabilities of the machine learning model 140 to determine: the trigger condition 422 is "8 o'clock tomorrow morning," and the task content 424 is "provide Beijing's weather forecast." Further, a subsequent task 450 may be created locally at the digital assistant 130. In this way, there is no need to modify the machine learning model, but a new processing interface may be added locally at the digital assistant 130 to complete the function of processing non-immediate tasks.

[0043] According to an example implementation of the present disclosure, in the process of configuring a subsequent task, if it is determined that the trigger condition indicates that the execution result of the task is provided at a predetermined time point in the future, the second prompt word can be provided to the machine learning model at the predetermined time point in the future. In this example, the trigger condition explicitly specifies a future time point, and a subsequent task 450 of a delayed type can be created at this time. According to an example implementation of the present disclosure, the delayed type can indicate that the task is not executed immediately and the execution result is not provided, but the execution of the task is delayed and the task is executed when the current time reaches a specified future time point. In this way, the task can be executed at a suitable time point to meet the user's needs, and then the execution result can be provided.

[0044] According to an example implementation of the present disclosure, the digital assistant 130 can provide a response 430 to indicate to the user that the user request has been received and the execution result of the task will be provided later. Specifically, a prompt word can be created based on the task content 424, for example, "Please provide the weather forecast for Beijing".

[0045] According to an example implementation of the present disclosure, subsequent tasks can be executed to provide execution results when the trigger condition is met. Specifically, the prompt word can be sent to the machine learning model 140 when the current time point reaches 8 a.m. the next day. Further, upon receiving the prompt word, the machine learning model 140 can quickly call the weather forecast tool to query the weather forecast for Beijing and provide a response to the digital assistant 130. In this way, the user can be ensured to obtain the execution result at the future time point specified by the user by means of delayed tasks.

[0046] Then, the execution result 440 can be returned to the user based on the response of the machine learning model 140. For example, the current time can be presented in the header of the execution result 440, and then the weather forecast for Beijing can be provided. It should be understood that the response of the machine learning model 140 can be considered to be rapid, and the user can be ensured to obtain the execution result at the specified time point. Alternatively and / or additionally, the trigger condition 422 can be adjusted, for example, the predetermined time point can be adjusted forward by 1 second (or other time length). At this time, a prompt word can be provided to the machine learning model 140 at 07:59:59 the next morning to ensure that the user receives the weather forecast at 08:00 in the morning.

[0047] It should be understood that although the user input described above only involves creating a subsequent task, alternatively and / or additionally, the user input may involve multiple subsequent tasks. For example, the user input may indicate "provide Beijing's weather forecast at 8 am every day." In this case, a recurring subsequent task may be created at the digital assistant 130, that is, sending a prompt word to the machine learning model at 8 am every day to obtain the weather forecast for Beijing.

[0048] Already seen Figure 4 The trigger condition is described as involving a predetermined future time point. Alternatively and / or additionally, the trigger condition may also involve "task completion". Figure 5 Describe in more detail, Figure 5 A block diagram 500 of creating a subsequent task according to some implementations of the present disclosure is shown. Figure 5 As shown, a conversation 510 may be provided between the user and the digital assistant. For example, the user input 520 may indicate: "Please calculate the weather forecast for 100 cities across the country". At this time, the digital assistant 130 may call the capabilities of the machine learning model 140 to determine: the trigger condition 522 is "when the statistics are completed", and the task content 524 is "provide the weather forecast for 100 cities across the country". Further, a subsequent task 550 may be created locally at the digital assistant 130.

[0049] According to an example implementation of the present disclosure, in the process of configuring subsequent tasks, if it is determined that the trigger condition indicates that the execution result of the task is provided when the task is completed, it means that the execution of the task will take a long time, so the prompt word can be immediately provided to the machine learning model. In order to avoid the user waiting for a long time, a response 530 can be provided to provide the user with feedback on the user input 520 to indicate that the user input is being processed.

[0050] In this example, the trigger condition does not explicitly specify a future time point, and an asynchronous type subsequent task 550 can be created at this time. According to an example implementation of the present disclosure, the asynchronous type can indicate that the execution result cannot be provided immediately, but the execution result is provided based on a callback message from the machine learning model when the task is completed. In this way, the execution result can be provided immediately when the task is completed. The digital assistant 130 can provide the user with weather forecasts for 100 cities across the country based on the response from the machine learning model. In this way, the user does not need to maintain the session 510 and wait for a long time, but can obtain the execution result immediately when the task is completed through an asynchronous task.

[0051] According to an exemplary implementation of the present disclosure, the digital assistant 130 herein may be a client tool created by the user 110 and / or other users to communicate with the machine learning model. In this case, the digital assistant 130 may include a preset task, that is, the preset task is added to the digital assistant 130 during the creation of the digital assistant 130. Figure 6 Describes more details of creating a digital assistant 130, which Figure 6 Block diagram 600 is shown of a process for adding preset tasks during creation of a digital assistant according to some implementations of the present disclosure.

[0052] like Figure 6 As shown, creator 610 can use digital assistant creation tool 620 to create digital assistant 130. In digital assistant creation tool 620, control 660 can indicate whether to allow the user of the generated digital assistant 130 to create subsequent tasks during the session. Specifically, if control 660 is not selected, digital assistant 130 will operate in a conventional manner.

[0053] If control 660 is selected, the user is allowed to create a scheduled task in natural language during a conversation with the digital assistant 650. At this time, the digital assistant 130 can operate according to the process described above when receiving the user input. For example, a corresponding prompt word is generated based on the user input, and it is determined whether a subsequent task needs to be created based on the answer of the machine learning model to the prompt word. If it is determined that the task specified by the user input belongs to a non-immediate task, a subsequent task is created. At this time, assuming that the digital assistant 130 receives "Please provide the weather forecast for Beijing at 8 o'clock tomorrow morning", the digital assistant 130 will create a subsequent task in the manner described above and provide a weather forecast at 8 o'clock the next morning. Alternatively and / or additionally, if control 660 is selected, the creator can further use control 670 to set a reminder statement. For example, "Please ask a question in the dialog box" can be set. At this time, the digital assistant will remind the user to set the task through the dialog box when interacting with the user for the first time.

[0054] Alternatively and / or additionally, in the process of creating the digital assistant 130, the preset task dialog box 622 may be called to add a preset task to the digital assistant. For example, the creator 610 may perform interaction 640, may enter a trigger condition of the preset task in box 630, may enter the task content of the preset task in box 632, etc. At this point, the preset task 650 will be included in the created digital assistant 130.

[0055] According to an example implementation of the present disclosure, the user of the digital assistant 130 can choose to activate the preset task 650, or to disable the preset task 650. If it is determined that the preset task 650 is activated, the trigger condition of the preset task can be determined. At this time, the trigger condition can be determined locally at the digital assistant 130 (for example, read from local storage, etc.). Further, a subsequent task of the preset task can be created so that when the trigger condition of the preset task is met, the execution result of the preset task is provided based on the machine learning model. In this way, it is allowed to receive user tasks in a more diverse manner, thereby receiving user input in a more convenient and quicker manner.

[0056] According to an example implementation of the present disclosure, the preset task may be processed in a manner similar to that described above. Specifically, if the trigger condition of the preset task indicates that the execution result of the preset task is to be provided at a predetermined time point in the future, then the preset task may be processed in a manner similar to that described above. Figure 4 Specifically, a prompt word (for example, called a third prompt word) can be generated based on the preset task, and the prompt word can indicate the task content of calling the machine learning model to execute the preset task. Figure 6 In the example, the prompt word may mean "provide weather forecast for Shanghai".

[0057] Further, a subsequent task of the preset task can be configured so that the subsequent task provides the generated prompt word to the machine learning model at a future time point (for example, 09:00 AM every Saturday). Subsequently, the subsequent task of the preset task can be executed to provide an execution result based on the response of the machine learning model to the prompt word. At this time, the machine learning model will receive the prompt word "Provide weather forecast for Shanghai" at 09:00 AM every Saturday, thereby returning the weather forecast for Shanghai to the digital assistant 130. In this way, the machine learning model can be periodically called at the digital assistant 130 to complete the preset task. Alternatively and / or additionally, another time point can be determined based on a future time point, for example, a prompt word is sent to the machine learning model at 08:59:59 AM (or other time point) 1 second in advance so that the user can receive a response to the prompt word on time.

[0058] By using the example implementation of the present disclosure, a new method of invoking a machine learning model to process user tasks is provided. In this way, user tasks that cannot obtain the execution results of the tasks within a predetermined time period can be processed in a more flexible and effective manner by creating delayed tasks and asynchronous tasks.

[0059] Example Process

[0060] Figure 7 A flow chart of a method 700 for processing a user task according to some implementations of the present disclosure is shown. At box 710, a user input in natural language is received from a user, the user input indicating a task to be performed. At box 720, a task type of the task is determined based on the user input. At box 730, it is determined whether the task type indicates that the execution result of the task cannot be obtained within a predetermined time period. If the result at box 730 is "yes", the method 700 proceeds to box 740. At box 740, a subsequent task of the task is created so that when the trigger condition of the task is met, the execution result of the task is provided using a machine learning model.

[0061] According to an example implementation of the present disclosure, determining a task type includes: generating a first prompt word based on user input, the first prompt word indicating that the task type is determined by a machine learning model; providing the first prompt word to the machine learning model; and determining the task type based on a first response of the machine learning model to the first prompt word.

[0062] According to an example implementation of the present disclosure, the first prompt word further indicates that the triggering conditions and task content of the task are determined by the machine learning model, and the creation of subsequent tasks includes: determining the triggering conditions and task content of the task based on the first response; and creating subsequent tasks of the task based on the triggering conditions and task content.

[0063] According to an example implementation of the present disclosure, creating a subsequent task of a task based on a trigger condition and task content includes: generating a second prompt word based on the task content, the second prompt word indicating that the task content is to be executed by a machine learning model; and configuring the subsequent task to provide the second prompt word to the machine learning model based on the trigger condition.

[0064] According to an example implementation of the present disclosure, configuring a subsequent task to provide a second prompt word to a machine learning model based on a trigger condition includes: in response to determining that the trigger condition indicates providing an execution result of the task at a predetermined future time point, providing the second prompt word to the machine learning model at a predetermined future time point.

[0065] According to an example implementation of the present disclosure, configuring a subsequent task to provide a second prompt word to a machine learning model based on a trigger condition includes: providing a second prompt word to the machine learning model in response to determining that the trigger condition indicates providing an execution result of the task when the task is completed; and providing feedback for the user input, the feedback indicating that the user input is being processed.

[0066] According to an exemplary implementation of the present disclosure, the method 700 further includes: executing a subsequent task to provide an execution result when a trigger condition is met.

[0067] According to an example implementation of the present disclosure, providing an execution result when a trigger condition is met includes: providing an execution result of the task based on a second response to a second prompt word based on a machine learning model.

[0068] According to an example implementation of the present disclosure, method 700 is performed at a digital assistant for accessing a machine learning model, and creating a subsequent task includes: creating a subsequent task at the digital assistant.

[0069] According to an example implementation of the present disclosure, the digital assistant further includes a preset task, which is added to the digital assistant during creation of the digital assistant, and the method further includes: in response to determining that the preset task is activated, determining a trigger condition for the preset task; and creating a subsequent task of the preset task so as to provide an execution result of the preset task based on a machine learning model when the trigger condition of the preset task is met.

[0070] According to an example implementation of the present disclosure, a trigger condition of a preset task indicates providing the execution result of the preset task at a predetermined time point in the future, and creating a subsequent task of the preset task includes: generating a third prompt word based on the preset task, the third prompt word indicating calling a machine learning model to execute the task content of the preset task; and configuring the subsequent task of the preset task to provide the third prompt word to the machine learning model at a future time point.

[0071] According to an example implementation of the present disclosure, the method 700 further includes: executing a subsequent task of the preset task to provide an execution result based on a third response to a third prompt word by the machine learning model.

[0072] Example devices and equipment

[0073] Figure 8A block diagram of an apparatus 800 for processing a user task according to some implementations of the present disclosure is shown. The apparatus 800 includes: a receiving module 810 configured to receive a user input represented in a natural language from a user, the user input indicating a task to be performed; a determining module 820 configured to determine a task type of the task based on the user input; and a creating module 830 configured to create a subsequent task of the task in response to determining that the task type indicates that the execution result of the task cannot be obtained within a predetermined time period so as to provide the execution result of the task using a machine learning model when the triggering condition of the task is met.

[0074] According to an example implementation of the present disclosure, the determination module 820 includes: a first generation module, configured to generate a first prompt word based on user input, the first prompt word indicating that the task type is determined by the machine learning model; a providing module, configured to provide the first prompt word to the machine learning model; and a type determination module, configured to determine the task type based on a first response of the machine learning model to the first prompt word.

[0075] According to an example implementation of the present disclosure, the first prompt word further indicates that the triggering condition and task content of the task are determined by a machine learning model, and the creation module 830 includes: an information determination module, configured to determine the triggering condition and task content of the task based on the first response; and a task creation module, configured to create subsequent tasks of the task based on the triggering condition and task content.

[0076] According to an example implementation of the present disclosure, a task creation module includes: a second generation module, configured to generate a second prompt word based on the task content, the second prompt word indicating that the task content is to be executed by a machine learning model; and a configuration module, configured to configure a subsequent task to provide the second prompt word to the machine learning model based on a trigger condition.

[0077] According to an example implementation of the present disclosure, the configuration module includes: a first providing module, configured to provide a second prompt word to the machine learning model at a predetermined future time point in response to determining that a trigger condition indicates providing an execution result of a task at a predetermined future time point.

[0078] According to an example implementation of the present disclosure, the configuration module includes: a second providing module, configured to provide a second prompt word to the machine learning model in response to determining that a trigger condition indicates providing an execution result of the task when the task is completed; and a feedback module, configured to provide feedback for the user input, the feedback indicating that the user input is being processed.

[0079] According to an exemplary implementation of the present disclosure, the apparatus 800 further includes: an execution module configured to execute a subsequent task so as to provide an execution result when a trigger condition is met.

[0080] According to an example implementation of the present disclosure, the execution module includes: a response-based providing module, configured to provide an execution result of the task based on a second response to a second prompt word of the machine learning model.

[0081] According to an example implementation of the present disclosure, the apparatus 800 is executed at a digital assistant for accessing a machine learning model, and the creation module is further configured to include: creating a subsequent task at the digital assistant.

[0082] According to an example implementation of the present disclosure, the digital assistant further includes a preset task, which is added to the digital assistant during creation of the digital assistant, and the device further includes: a trigger condition determination module, configured to determine the trigger condition of the preset task in response to determining that the preset task is activated; and a preset-based creation module, configured to create a subsequent task of the preset task so as to provide an execution result of the preset task based on a machine learning model when the trigger condition of the preset task is met.

[0083] According to an example implementation of the present disclosure, the trigger condition of a preset task indicates providing the execution result of the preset task at a predetermined time point in the future, and the preset-based creation module includes: a third generation module, configured to generate a third prompt word based on the preset task, the third prompt word indicating calling a machine learning model to execute the task content of the preset task; and a preset-based provision module, configured to configure a subsequent task of the preset task to provide the third prompt word to the machine learning model at a future time point.

[0084] According to an example implementation of the present disclosure, the device 800 further includes: a preset task execution module, configured to execute a subsequent task of the preset task so as to provide an execution result based on a third response to a third prompt word based on the machine learning model.

[0085] Fig. 9 9 is a block diagram of a device 900 capable of implementing multiple implementations of the present disclosure. It should be understood that Fig. 9 The computing device 900 shown is merely exemplary and should not constitute any limitation on the functionality and scope of the implementations described herein. Fig. 9 The computing device 900 shown may be used to implement the methods described above.

[0086] like Fig. 9As shown, computing device 900 is in the form of a general-purpose computing device. Components of computing device 900 may include, but are not limited to, one or more processors or processing units 910, memory 920, storage device 930, one or more communication units 940, one or more input devices 950, and one or more output devices 960. Processing unit 910 may be an actual or virtual processor and is capable of performing various processes according to a program stored in memory 920. In a multi-processor system, multiple processing units execute computer executable instructions in parallel to improve the parallel processing capabilities of computing device 900.

[0087] The computing device 900 typically includes a plurality of computer storage media. Such media may be any available media accessible to the computing device 900, including but not limited to volatile and non-volatile media, removable and non-removable media. The memory 920 may be a volatile memory (e.g., a register, a cache, a random access memory (RAM)), a non-volatile memory (e.g., a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), flash memory), or some combination thereof. The storage device 930 may be a removable or non-removable medium, and may include a machine-readable medium, such as a flash drive, a disk, or any other medium, which may be capable of being used to store information and / or data (e.g., training data for training) and may be accessed within the computing device 900.

[0088] The computing device 900 may further include additional removable / non-removable, volatile / non-volatile storage media. Fig. 9 As shown in , a disk drive for reading or writing from a removable, non-volatile disk (e.g., a "floppy disk") and an optical drive for reading or writing from a removable, non-volatile optical disk may be provided. In these cases, each drive may be connected to the bus (not shown) by one or more data media interfaces. Memory 920 may include a computer program product 925 having one or more program modules that are configured to perform various methods or actions of various implementations of the present disclosure.

[0089] The communication unit 940 enables communication with other computing devices via a communication medium. Additionally, the functions of the components of the computing device 900 can be implemented in a single computing cluster or multiple computing machines that can communicate via a communication connection. Therefore, the computing device 900 can operate in a networked environment using a logical connection with one or more other servers, a network personal computer (PC), or another network node.

[0090] Input device 950 may be one or more input devices, such as a mouse, keyboard, tracking ball, etc. Output device 960 may be one or more output devices, such as a display, a speaker, a printer, etc. Computing device 900 may also communicate with one or more external devices (not shown) through communication unit 940 as needed, such as storage devices, display devices, etc., communicate with one or more devices that allow a user to interact with computing device 900, or communicate with any device that allows computing device 900 to communicate with one or more other computing devices (e.g., a network card, a modem, etc.). Such communication may be performed via an input / output (I / O) interface (not shown).

[0091] According to an exemplary implementation of the present disclosure, a computer-readable storage medium is provided, on which computer-executable instructions are stored, wherein the computer-executable instructions are executed by a processor to implement the method described above. According to an exemplary implementation of the present disclosure, a computer program product is also provided, the computer program product is tangibly stored on a non-transitory computer-readable medium and includes computer-executable instructions, and the computer-executable instructions are executed by a processor to implement the method described above. According to an exemplary implementation of the present disclosure, a computer program product is provided, on which a computer program is stored, and when the program is executed by a processor, the method described above is implemented.

[0092] Various aspects of the present disclosure are described herein with reference to the flowcharts and / or block diagrams of the methods, devices, equipment, and computer program products implemented according to the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer-readable program instructions.

[0093] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processing unit of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0094] Computer-readable program instructions can be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, so that the instructions executed on the computer, other programmable data processing apparatus, or other device implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0095] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple implementations of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of a module, program segment or instruction includes one or more executable instructions for realizing the logical function of the specification. In some implementations as replacements, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be realized by a special hardware-based system that performs the function or action of the specification, or can be realized by a combination of special hardware and computer instructions.

[0096] The above descriptions of various implementations of the present disclosure are exemplary, non-exhaustive, and not limited to the disclosed implementations. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described implementations. The selection of terms used herein is intended to best explain the principles of the implementations, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the various implementations disclosed herein.

Claims

1. A method for processing a user task, comprising: receiving user input in natural language from a user, the user input indicating a task to be performed; determining a task type for the task based on the user input; as well as In response to determining that the task type indicates that the execution result of the task cannot be obtained within a predetermined time period, a subsequent task of the task is created so as to provide the execution result of the task using a machine learning model when a trigger condition of the task is met.

2. The method according to claim 1, wherein determining the task type comprises: Generate a first prompt word based on the user input, the first prompt word indicating that the task type is determined by the machine learning model; providing the first prompt word to the machine learning model; as well as The task type is determined based on a first response of the machine learning model to the first prompt word.

3. The method according to claim 2, wherein the first prompt word further indicates that the triggering condition and task content of the task are determined by the machine learning model, and creating the subsequent task comprises: Based on the first response, determining the triggering condition and the task content of the task; as well as A subsequent task of the task is created based on the trigger condition and the task content.

4. The method according to claim 3, wherein creating a subsequent task of the task based on the trigger condition and the task content comprises: Generate a second prompt word based on the task content, the second prompt word indicating that the task content is to be performed by the machine learning model; as well as The subsequent task is configured to provide the second prompt word to the machine learning model based on the trigger condition.

5. The method of claim 4, wherein configuring the subsequent task to provide the second prompt word to the machine learning model based on the trigger condition comprises: In response to determining that the trigger condition indicates providing the execution result of the task at a predetermined time point in the future, providing the second prompt word to the machine learning model according to any one of the following: At the predetermined future time point, providing the second prompt word to the machine learning model; as well as At another time point determined according to the future predetermined time point, providing the second prompt word to the machine learning model.

6. The method according to claim 4, wherein configuring the subsequent task to provide the second prompt word to the machine learning model based on the trigger condition comprises: In response to determining that the trigger condition indicates providing the execution result of the task when the task is completed, providing the second prompt word to the machine learning model; as well as Feedback is provided for the user input, the feedback indicating that the user input is being processed.

7. The method according to claim 4, further comprising: The subsequent task is executed to provide the execution result when the trigger condition is met.

8. The method according to claim 7, wherein providing the execution result when the trigger condition is met comprises: Based on a second response of the machine learning model to the second prompt word, the execution result of the task is provided.

9. The method of claim 1, wherein the method is performed at a digital assistant for accessing the machine learning model, and creating the subsequent task comprises: The follow-up task is created at the digital assistant.

10. The method of claim 9, wherein the digital assistant further comprises pre-set tasks, the pre-set tasks being added to the digital assistant during creation of the digital assistant, and the method further comprising: In response to determining that the preset task is activated, determining a triggering condition of the preset task; as well as A subsequent task of the preset task is created so as to provide an execution result of the preset task based on the machine learning model when the trigger condition of the preset task is met.

11. The method according to claim 10, wherein the triggering condition of the preset task indicates providing the execution result of the preset task at a predetermined time point in the future, and creating a subsequent task of the preset task comprises: Generate a third prompt word based on the preset task, the third prompt word indicating calling the machine learning model to execute the task content of the preset task; as well as The subsequent task of the preset task is configured so as to provide the second prompt word to the machine learning model according to any one of the following: At the predetermined future time point, providing the second prompt word to the machine learning model; as well as At another time point determined according to the future predetermined time point, providing the second prompt word to the machine learning model.

12. The method according to claim 11, further comprising: Execute a subsequent task of the preset task to provide the execution result based on a third response of the machine learning model to the third prompt word.

13. A device for processing user tasks, comprising: a receiving module configured to receive a user input expressed in a natural language from a user, the user input indicating a task to be performed; a determination module configured to determine a task type of the task based on the user input; as well as A creation module is configured to create a subsequent task of the task in response to determining that the task type indicates that the execution result of the task cannot be obtained within a predetermined time period so as to provide the execution result of the task by using a machine learning model when the triggering condition of the task is met.

14. An electronic device, comprising: at least one processing unit; as well as At least one memory, the at least one memory being coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit, the instructions causing the electronic device to perform the method according to any one of claims 1 to 12 when executed by the at least one processing unit.

15. A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, causes the processor to implement the method according to any one of claims 1 to 12.

Images