Task execution method, device, storage medium and program product

Abstract

The embodiments of the present invention provide a task execution method, device, storage medium and program product, which relate to the field of computer technology. The method includes: receiving task demand information input by a user; determining an application program interface of a target function in a function library corresponding to a target application, wherein the target function corresponds to the task demand information; determining initial rendering parameters of multiple interface components corresponding to the application program interface; generating a customized interface based on the initial rendering parameters of multiple interface components; and generating and executing a target task according to task data input by the user through the customized interface. This solution can complete the rendering of a customized interface containing multiple interface components. In actual application, the user can directly input task data in the customized interface without the need to perform other long-link complex operations, and the user experience is better.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a task execution method, device, storage medium and program product. Background Art

[0002] At present, in actual use, the application has a corresponding front-end user interface, in which various functions are displayed in a hierarchical manner. When the user needs to use a certain function to perform a task, he needs to find the required interface components layer by layer according to the established operation link of the application interface, and perform corresponding operations on the interface components to generate and execute the task.

[0003] However, when executing tasks through the above approach, a long operation chain is required and it is easy to be disturbed by other irrelevant functions in the application interface, resulting in a poor user experience. Summary of the invention

[0004] Embodiments of the present invention provide a task execution method, device, storage medium, system and product, which are used to simplify the operation link of triggering tasks by users.

[0005] In a first aspect, an embodiment of the present invention provides a task execution method, the method comprising:

[0006] Receive task request information input by the user;

[0007] Determine, in a function library corresponding to a target application, an application program interface of a target function, wherein the target function corresponds to the task requirement information;

[0008] Determining initial rendering parameters of a plurality of interface components corresponding to the application program interface;

[0009] generating a customized interface based on initial rendering parameters of the plurality of interface components;

[0010] Generate and execute a target task according to the task data input by the user through the customized interface.

[0011] In a second aspect, an embodiment of the present invention provides a task execution device, the device comprising:

[0012] A receiving module, used to receive task demand information input by a user;

[0013] A determination module, configured to determine an application program interface of a target function in a function library corresponding to a target application, the target function corresponding to the task demand information, and determine initial rendering parameters of a plurality of interface components corresponding to the application program interface;

[0014] A generation module, used to generate a customized interface based on the initial rendering parameters of the multiple interface components;

[0015] The execution module is used to generate and execute the target task according to the task data input by the user through the customized interface.

[0016] In a third aspect, an embodiment of the present invention provides an electronic device, comprising: a memory, a processor, and a communication interface; wherein the memory stores executable code, and when the executable code is executed by the processor, the processor executes the task execution method described in the first aspect.

[0017] In a fourth aspect, an embodiment of the present invention provides a non-temporary machine-readable storage medium having executable code stored thereon. When the executable code is executed by a processor of an electronic device, the processor executes the task execution method as described in the first aspect.

[0018] In a fifth aspect, an embodiment of the present invention provides a computer program product, comprising: a computer program, which, when executed by a processor of an electronic device, enables the processor to execute the task execution method as described in the first aspect.

[0019] The task execution method provided by the embodiment of the present invention can determine the application program interface of the target function corresponding to the task demand information in the function library corresponding to the target application after receiving the task demand information input by the user, and then determine the initial rendering parameters of multiple interface components corresponding to the application program interface. It can be understood that since the application program interface is the application program interface of the target function corresponding to the user's task demand information, that is, the application program interface is strongly related to the user's task demand information, therefore, by determining the initial rendering parameters of multiple interface components corresponding to the application program interface, the rendering of the customized interface required by the user can be completed. That is, this solution can automatically generate a customized interface that meets the corresponding task triggering requirements based directly on the task demand information triggered by the user, thereby simplifying the user's operation link. Moreover, in actual application, the user can directly input the task data in the customized interface without the need to perform other complex operations of long links, and will not be interfered by other interface components, so the user experience is better. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 A flowchart of a task execution method provided by an embodiment of the present invention;

[0022] Figure 2 A schematic diagram of an application of a task execution method provided by an embodiment of the present invention;

[0023] Figure 3 A flowchart of generating a customized interface based on initial rendering parameters of multiple interface components provided by an embodiment of the present invention;

[0024] Figure 4 A flowchart of a task execution method provided by an embodiment of the present invention;

[0025] Figure 5 A schematic diagram of an application of a customized interface provided by an embodiment of the present invention;

[0026] Figure 6 A schematic diagram of another application of a customized interface provided by an embodiment of the present invention;

[0027] Figure 7 A schematic diagram of an application of another customized interface provided by an embodiment of the present invention;

[0028] Figure 8 A schematic diagram of the structure of a task execution device provided by an embodiment of the present invention;

[0029] Fig. 9 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in the embodiments of the present invention are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0032] Some embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the case where there is no conflict between the embodiments, the following embodiments and the features in the embodiments can be combined with each other. In addition, the step sequence in the following method embodiments is only an example and not a strict limitation.

[0033] Currently, when a user needs to use a certain function in an application to perform a task, he or she needs to go through a long operation link and is easily disturbed by other irrelevant functions in the application interface, resulting in a poor user experience.

[0034] In view of this, an embodiment of the present invention provides the following ideas to solve the above problems: after receiving the task demand information input by the user, for example, by describing it in natural language, the target function corresponding to the task demand information and the application program interface of the target function are determined in the function library corresponding to the target application. Based on the initial rendering parameters of multiple interface components corresponding to the application interface, a customized interface for triggering the task demand is generated for the user, and a customized interface that meets the corresponding task triggering requirements is automatically generated directly based on the task demand information triggered by the user, thereby simplifying the user's operation link. Moreover, in actual application, the user can directly input the task data in the customized interface, without the need to perform complex operations of other links, and will not be interfered with by other interface components, so the user experience is better.

[0035] Figure 1 A flowchart of a task execution method provided in an embodiment of the present invention. The execution subject of the method may be an artificial intelligence tool (also referred to as an AI tool) or an artificial intelligence assistant (also referred to as an AI assistant) that can generate a customized interface for some application programs, which are not listed here one by one. Figure 1 As shown, the method comprises the following steps:

[0036] 101. Receive task request information input by a user.

[0037] 102. In the function library corresponding to the target application, determine the application program interface of the target function, and the target function corresponds to the task requirement information.

[0038] 103. Determine initial rendering parameters of multiple interface components corresponding to the application program interface.

[0039] 104. Generate a customized interface based on initial rendering parameters of multiple interface components.

[0040] 105. Generate and execute the target task according to the task data input by the user through the customized interface.

[0041] It should be noted that the task requirement information input by the user can be expressed in the form of natural language. Natural language refers to the language used by people in daily life, such as Chinese, English, etc. The target application can be any type of application such as financial applications, game applications, artificial intelligence applications (such as large language models), etc., and is not limited here. The function library corresponding to the target application can store a variety of functions, such as file and media processing functions (supporting file upload, download, storage, video conversion, etc.), search and indexing functions (supporting geolocation, address resolution, route planning, map rendering, etc.), data analysis and visualization (supporting data statistics, report generation, chart drawing, etc.), etc., which are not listed here one by one, as long as the function can be called by the application interface. The following is a specific explanation of this solution taking the execution subject as an AI assistant as an example:

[0042] In actual applications, users input task request information into the AI ​​assistant through natural language, such as "I want to upload several documents as a knowledge base", "Parse the documents immediately", "I have a document to do questions and answers", etc., which are not listed here one by one.

[0043] After receiving the task request information, the AI ​​assistant determines the application program interface of the target function in the function library corresponding to the target application. The specific process is as follows:

[0044] As an implementation method, the Natural Language Processing (NLP) algorithm can be used to parse the task demand information and identify the user's intention. According to the user's intention, the corresponding target function is found in the function library, and then the application interface corresponding to the target function is determined according to the mapping relationship between the set function and the application interface.

[0045] As another implementation method, a large language model (LLM) can be used to determine target function description information that matches the task requirement information in a function library corresponding to the target application, wherein the function library includes multiple functions provided by the target application, each function has corresponding function description information, and the target function description information is function description information corresponding to the target function among the multiple functions; based on the target function description information, an application program interface of the target function is determined.

[0046] In specific implementation, the user's task demand information is input into the large language model, so that the large language model can perform semantic analysis, scene analysis, context analysis and other operations, thereby improving the accuracy of identifying user intentions. In this process, the large language model can also analyze the user's preferences based on the historical interaction data between the AI ​​assistant and the user, thereby further improving the accuracy of identifying user intentions. After determining the user's intention, the corresponding target function can be found in the function library, and then the application interface corresponding to the target function can be quickly determined based on the mapping relationship between the set function and the application interface.

[0047] For the above two implementations, it should be noted that the function library stores multiple functions, each function has its corresponding function description information, and multiple functions have a one-to-one mapping relationship with multiple application program interfaces, that is, multiple functions have their own corresponding application program interfaces. In specific implementation, the function description information corresponding to the multiple functions can be preprocessed separately (such as word segmentation, part-of-speech tagging, constructing vector representation, etc.) to improve the efficiency of finding the corresponding target function in the function library according to the user's intention.

[0048] After obtaining the application program interface of the target function, the initial rendering parameters of multiple interface components corresponding to the application program interface can be determined. The specific process is as follows:

[0049] As an implementation method, the initial rendering parameters of multiple interface components described in natural language corresponding to the application interface are determined in the function library; through the large language model, based on the set interface rendering protocol, the format of the initial rendering parameters of multiple interface components described in natural language is converted to obtain the initial rendering parameters of multiple interface components described in protocol code.

[0050] In actual applications, the contents such as "function", "function description information", "application interface", "initial rendering parameters of multiple interface components" in the function library are described in natural language. In order to realize the rendering of the front-end interface, it is necessary to convert the format of the initial rendering parameters of multiple interface components based on the set interface rendering protocol to obtain the initial rendering parameters of multiple interface components described in the protocol code. Among them, the initial rendering parameters refer to a series of configuration parameters required when the interface components are displayed in the default form. These initial rendering parameters determine the appearance and behavior of the interface components and how they interact with the user. They may include color, size, position, font, margin, padding, shadow, animation effects, etc., depending on the type of interface component and the required functions. For example, the initial rendering parameters of a button component may include: "The button background color is blue, the font size is 16px, the button text is "Submit", the button is located in the center of the front-end interface, the width is 150px, and the height is 50px".

[0051] During specific implementation, the large language model can be trained using the set training samples and the supervision information corresponding to the training samples to obtain a trained large language model, and then the initial rendering parameters of multiple interface components described in natural language are input into the trained large language model. After receiving the initial rendering parameters, the large language model will parse them to obtain the specific parameter information of the initial rendering parameters, and map the specific parameter information to the data structure defined by the set interface rendering protocol, so that the large language model generates codes that comply with the set interface rendering protocol, and these codes contain the initial rendering parameters of multiple interface components described in the protocol code. In this way, the accuracy of the final initial rendering parameters is improved, laying the foundation for the subsequent smooth completion of the rendering of the customized interface based on the initial rendering parameters of multiple interface components.

[0052] As another implementation method, after determining the initial rendering parameters of multiple interface components described in natural language corresponding to the application program interface in the function library, the initial rendering parameters of the multiple interface components are parsed by using a pre-written parser to obtain the initial rendering parameters of the multiple interface components described in the protocol code. For example, assuming that the interface rendering protocol is in JSON format, then a parser that can parse the natural language description and convert it into the initial rendering parameters of the interface components in JSON format can be pre-written. After obtaining the initial rendering parameters of the multiple interface components described in natural language, it can be directly input into the pre-written parser.

[0053] Afterwards, a customized interface is generated based on the initial rendering parameters of the above-mentioned multiple interface components, and the customized interface may include multiple different types of components. Among them, the customized interface can be expressed in the form of a form, a window, etc., which is not limited here. For the specific process of generating a customized interface, please refer to the embodiment below, which will not be described in detail here. After obtaining the customized form, the target task can be generated and executed according to the task data entered by the user through the customized interface.

[0054] The generation and execution process of the above target tasks can be found in Figure 2 , Figure 2 It shows the process of user inputting task demand information - determining the target function in the function library and the application interface corresponding to the target function - determining the initial rendering parameters of multiple interface components corresponding to the application interface - generating a customized page - generating and executing the target task. Figure 2 It should be noted that Figure 2The function library in the diagram only illustrates the functions of "knowledge access" and "knowledge base creation", but the function library can actually contain many other functions, which are not listed here one by one, but about the two functions of "knowledge access" and "knowledge base creation". Among them, the function description corresponding to "knowledge access" can be "support data access, import rich text files, audio, and video through local upload", and its corresponding application interface can be " / knowledgeCenter / knowledgeCorpus / uploadCORpus", and its corresponding interface component rendering parameters can be "belong to knowledge category / single choice; whether to parse / single choice; upload file / upload file component". The function description corresponding to "knowledge base creation" can be "support data access, import rich text files, audio, and video through local upload, and create document knowledge base and graphic knowledge base", and its corresponding application interface can be " / knowledgeCenter / knowledgeCorpus / uploadCORpus... / creatAsync", and its corresponding interface component rendering parameters can be "belong to knowledge category / single choice; whether to parse / single choice; upload file / upload file component; knowledge base name / single line".

[0055] The task execution method provided by the embodiment of the present invention can determine the target function corresponding to the user's task demand information and the application interface corresponding to the target function in the function library corresponding to the target application. By determining the initial rendering parameters of the multiple interface components corresponding to the application interface, the rendering of the customized interface containing multiple interface components can be completed, and the customized interface that meets the corresponding task triggering requirements can be automatically generated directly based on the task demand information triggered by the user, thereby simplifying the user's operation link. Moreover, in actual application, the user can directly input the task data in the customized interface without the need to perform other complicated operations with long links, and will not be interfered with by other interface components, so the user experience is better.

[0056] As for the specific process of generating a customized interface, as an implementation method, the HTML or XML code of the interface components can be dynamically generated according to the initial rendering parameters and inserted into the corresponding position of the interface to obtain the customized interface required by the user. This method can be implemented through JavaScript or template engines (such as Handlebars, Mustache, etc.).

[0057] As another implementation, Figure 3 A flowchart of generating a customized interface based on initial rendering parameters of multiple interface components is provided in an embodiment of the present invention, such as Figure 3 As shown, a customized interface is generated based on the initial rendering parameters of multiple interface components, including:

[0058] 301. Identify a keyword corresponding to a first interface component in task demand information, where the first interface component is included in a plurality of interface components.

[0059] 302. Determine a display state of the first interface component according to the keyword.

[0060] 303. Update the initial rendering parameters of the first interface component according to the display state of the first interface component to obtain the target rendering parameters of the first interface component.

[0061] 304. Generate a customized interface based on the target rendering parameters of the first interface component and the initial rendering parameters of the remaining other interface components.

[0062] It should be noted that an interface is composed of multiple interface components, and the function library contains a one-to-one mapping relationship between "function", "function description information", "application interface", and "initial rendering parameters of multiple interface components". In actual applications, when the target function is determined based on the task requirement information and the function description information in the function library, the rendering parameters of the multiple interface components corresponding to the target function are also determined.

[0063] In specific implementation, the large language model can be used to identify keywords in the task demand information to determine the keywords corresponding to the first interface component among the multiple interface components. The first interface component here can be any one or more of the above-mentioned multiple interface components, and generally refers to the interface component corresponding to the keyword in the task demand information. For ease of understanding, for example, assuming that the task demand information is "I need to parse this document immediately", and at this time the multiple interface components include "knowledge classification component", "parsing component (the parsing component includes "postponed parsing" and "immediate parsing" options)", "file upload component", then, the keyword corresponding to the first interface component identified by the large language model can be "parse immediately", and based on this keyword, it can be determined that the current display status of the first interface component (i.e., the parsing component) is "parse immediately".

[0064] Assuming that the initial rendering parameter of the first interface component is "Whether to parse / single choice", then according to the display status of the above-mentioned first interface component, the initial rendering parameter can be updated from "Whether to parse / single choice" to the target rendering parameter "parse, and parse immediately", based on the target rendering parameter and the initial rendering parameters of the remaining other interface components (such as the initial rendering parameter of the "knowledge classification component" is "belonging to the knowledge category / single choice (knowledge system 1 / knowledge system 2 / knowledge system 3)", the initial rendering parameter of the "file upload component" is "upload document (support uploading doc, txt, pdf formats)") The interface is rendered to obtain the final customized interface.

[0065] By identifying the keywords corresponding to the first interface component in the task demand information, the display status of the first interface component among multiple interface components can be quickly determined, so that the initial rendering parameters of the first interface component can be updated according to the display status of the first interface component. It can be understood that since the task demand information does not involve changes in the display status of the remaining other interface components, it is only necessary to render the interface based on the updated target rendering parameters of the first interface component and the initial rendering parameters of the remaining other interface components, without changing the initial rendering parameters of the remaining other interface components. While ensuring that a customized interface can be generated for users, the efficiency of interface generation is improved.

[0066] Further, in an optional embodiment, a customized interface is generated based on the target rendering parameters of the first interface component and the initial rendering parameters of the remaining other interface components, including the following steps: determining the logical linkage relationship between multiple interface components; in response to the target rendering parameters of the first interface component, determining the target rendering parameters of at least one second interface component linked with the first interface component according to the logical linkage relationship, wherein the at least one second interface component is included in the multiple interface components, and the target rendering parameters of the at least one second interface component are the updated results of the initial rendering parameters of the at least one second interface component; based on the target rendering parameters of the first interface component, the target rendering parameters of the at least one second interface component and the initial rendering parameters of the remaining other interface components, a customized interface is generated.

[0067] In actual applications, there may be a logical linkage relationship between multiple interface components. For ease of understanding, for example, assume that multiple interface components include: "Certain ID component: (to be filled in)" and "Date of birth component: (to be filled in)", then it can be understood that the Certain ID component and the Date of birth component are two interface components with a logical linkage relationship (because after the user fills in his or her Certain ID number, the corresponding date of birth can be automatically obtained from the Certain ID number without the user having to fill it in again).

[0068] In view of this, after obtaining the target rendering parameters of the first interface component based on the task demand information, it is possible to determine whether there is an interface component in multiple interfaces that has a logical linkage relationship with the first interface component based on the logical linkage relationship. If not, there is no need to consider the logical linkage relationship of the first interface component when rendering the customized interface subsequently. If there is at least one second interface component linked to the first interface component, the target rendering parameters of the at least one second interface component can be directly determined based on the logical linkage relationship, and then a customized interface can be generated based on the target rendering parameters of the first interface component, the target rendering parameters of at least one second interface component, and the initial rendering parameters of the remaining other interface components.

[0069] Based on the above, since there may be a logical linkage relationship between multiple interface components, by determining the logical linkage relationship between multiple interface components and then updating the initial rendering parameters of at least one second interface component that has a logical linkage relationship with the first interface, the process of obtaining the display status of at least one second interface component is eliminated, thereby further improving the generation efficiency of the customized interface.

[0070] In an embodiment of the present invention, in addition to considering the logical linkage relationship between multiple interface components in the process of generating a customized interface, after the customized interface is generated, the display results of some interface components can still be updated based on the logical linkage relationship between the multiple interface components to improve the user experience. The specific update process is as follows: determine the logical linkage relationship between multiple interface components; based on the user's operation behavior on the third interface component, determine the display status of the fourth interface component linked to the third interface component according to the logical linkage relationship, and the third interface component and the fourth interface component are included in the multiple interface components; based on the user's operation behavior on the third interface component and the determined display status of the fourth interface component, update the display results of the third interface component and the fourth interface component in the customized interface.

[0071] In specific implementation, the logical linkage relationship between multiple interface components can be pre-configured in the configuration file. When the user triggers an operation (such as a modification operation) on the third interface component among the multiple interface components, the display state of the fourth interface component linked with the third interface component is determined according to the logical linkage relationship. For example, assuming that the third interface component is a "certain certificate component" and the fourth interface component is a "date of birth component", that is, the third interface component and the fourth interface component have a logical linkage relationship. Then, at this time, when the user modifies the certificate number filled in the "certain certificate component", the display state of the "date of birth component" will also change accordingly. For example, assuming that before the third interface component (i.e., the certain certificate component) is not modified, its corresponding certificate number is "xxxxxx19990120xxxx" (at this time, the date of birth displayed by the fourth interface component is January 20, 1999). After the modification operation is performed, its corresponding certificate number is "xxxxxx20000120xxxx". At this time, the date of birth displayed by the fourth interface component is changed to January 20, 2000. Based on the above, after the final display results of the third interface component and the fourth interface component are determined, both can be updated directly in the customized interface.

[0072] By determining the logical linkage relationship between multiple interface components, when the user triggers an operation on the third interface component in the customized interface, the display status of the fourth interface component linked to the third interface component can be determined according to the logical linkage relationship, without having to obtain the display result of the fourth interface component separately. This not only improves the update speed of the customized interface, but also ensures the accuracy of the display results of each interface component in the customized interface.

[0073] Figure 4 A flowchart of a task execution method provided by an embodiment of the present invention is shown in FIG. Figure 3 As shown, the method comprises the following steps:

[0074] 401. Receive task request information input by a user.

[0075] 402. In the human-computer dialogue interface, task demand information is displayed.

[0076] 403. In a function library corresponding to the target application, determine an application program interface of the target function, where the target function corresponds to the task requirement information.

[0077] 404. Determine initial rendering parameters of multiple interface components corresponding to the application program interface.

[0078] 405. Generate a customized interface based on initial rendering parameters of multiple interface components.

[0079] 406. Generate response information corresponding to the task request information through the large language model, and display the response information in the human-computer dialogue interface. The response information includes natural language task generation prompt information and task cards.

[0080] 407. In response to the user's viewing operation on the task card, a customized interface is displayed.

[0081] 408. Generate and execute the target task according to the task data input by the user through the customized interface.

[0082] In actual applications, for the interfaces viewed by users (such as web page interfaces, application interfaces, game interfaces, AI assistant interfaces, etc., which are not listed here one by one), they can be divided into human-computer dialogue interfaces and customized interfaces. For ease of understanding, the following is combined with Figure 5 , take the AI ​​assistant interface as an example for subsequent explanation:

[0083] exist Figure 5 In the example, the left interface is the human-computer dialogue interface, and the right interface is the customized interface. In the specific implementation, suppose the task request information entered by the user in the AI ​​assistant is "I have a document to ask questions and answers". Figure 5The task requirement information will be displayed in the human-computer dialogue interface on the left. During this process, the AI ​​assistant will determine the application interface of the target function in the function library corresponding to the target application, and determine the initial rendering parameters of multiple interface components corresponding to the application interface, and then generate a customized interface based on the initial rendering parameters of multiple interface components.

[0084] Afterwards, the AI ​​assistant generates a reply message corresponding to the task request information through the large language model, and displays the reply message in the human-computer dialogue interface. It should be noted that the reply message includes the task generation prompt information in natural language (such as Figure 5 "OK, we have created a document question answering assistant agent task for you. You can initiate the task in just 3 steps") and task cards (such as Figure 5 The specific execution steps will be recorded in the task card below the task generation prompt information).

[0085] In specific implementation, the user only needs to click on the task card, and the AI ​​assistant will respond to the user's viewing operation on the task card and display Figure 5 The customized interface on the right. Continuing with the above example, the three steps required to initiate a task are "knowledge access", "knowledge base creation" and "agent orchestration" (i.e., three customized interfaces). Figure 5 The customized interface of "Knowledge Access" is displayed in Figure 5 In the interface, the components include "knowledge classification", "whether to parse", and "upload files". The initial rendering parameters of "knowledge classification" include knowledge system 1, knowledge system 2, and knowledge system 3. Users can select them through the drop-down box. The initial rendering parameter of "whether to parse" is "parse immediately". The initial rendering parameters of uploaded files include support for uploading doc, txt, and pdf formats. After the user confirms the parameters of each interface component in the above "knowledge access" interface, click "Next" to jump to Figure 6 (That is, the "Knowledge Base Creation" interface).

[0086] exist Figure 6 In the interface components, the "knowledge base name" and "vectorized model" are included. The initial rendering parameters of the knowledge base name are "zhishiku123" and the initial rendering parameters of the "vectorized model" are "moxing123". Based on the initial rendering parameters of the above two interface components, the customized "knowledge access" interface can be generated. After that, click "Next" to jump to Figure 7 (i.e. the “agent orchestration” interface).

[0087] exist Figure 7In the example, the interface components include "agent name", "agent description", "agent avatar", "selection tool", "select large model", and "prompt word". The initial rendering parameter of "agent name" is "document question and answer assistant", and the initial rendering parameter of "agent description" is "this is a document question and answer assistant, which automatically parses the description information based on the uploaded document". The initial rendering parameters of the remaining other interface components are all in the state of waiting to be selected or added. For specific forms of expression, see Figure 7 After confirming the parameters of each interface component, click "Next" to complete the generation and execution of the target task.

[0088] Based on the above, by displaying the task requirement information in the human-computer dialogue interface, generating reply information corresponding to the task requirement information through a large language model, and then displaying the reply information in the human-computer dialogue interface, users can more intuitively understand the steps required to generate the task. They only need to fill in the corresponding task data according to the guidance, thereby improving the user experience.

[0089] The task execution device of one or more embodiments of the present invention will be described in detail below. Those skilled in the art will appreciate that these devices can be configured using commercially available hardware components through the steps taught in this solution.

[0090] Figure 8 A schematic diagram of a task execution device provided by an embodiment of the present invention is shown in FIG. Figure 8 As shown, the device includes: a receiving module 81, a determining module 82, a generating module 83 and an executing module 84.

[0091] The receiving module 81 is used to receive task demand information input by the user.

[0092] The determination module 82 is used to determine an application program interface of a target function in a function library corresponding to a target application, the target function corresponding to the task requirement information, and determine initial rendering parameters of a plurality of interface components corresponding to the application program interface.

[0093] The generating module 83 is used to generate a customized interface based on the initial rendering parameters of the multiple interface components.

[0094] The execution module 84 is used to generate and execute the target task according to the task data input by the user through the customized interface.

[0095] Among them, optionally, the generation module 83 is specifically used to: identify the keyword corresponding to the first interface component in the task demand information, and the first interface component is included in the multiple interface components; determine the display state of the first interface component according to the keyword; update the initial rendering parameters of the first interface component according to the display state of the first interface component to obtain the target rendering parameters of the first interface component; generate a customized interface based on the target rendering parameters of the first interface component and the initial rendering parameters of the remaining other interface components.

[0096] Among them, optionally, the generation module 83 is also specifically used to: determine the logical linkage relationship between multiple interface components; in response to the target rendering parameters of the first interface component, determine the target rendering parameters of at least one second interface component linked with the first interface component according to the logical linkage relationship, the at least one second interface component is included in the multiple interface components, and the target rendering parameters of the at least one second interface component are the updated results of the initial rendering parameters of the at least one second interface component; based on the target rendering parameters of the first interface component, the target rendering parameters of the at least one second interface component and the initial rendering parameters of the remaining other interface components, generate a customized interface.

[0097] Among them, optionally, the device also includes a display module, which is used to: determine the logical linkage relationship between the multiple interface components, and based on the user's operation behavior on the third interface component, determine the display status of the fourth interface component linked to the third interface component according to the logical linkage relationship, the third interface component and the fourth interface component are included in the multiple interface components, and according to the user's operation behavior on the third interface component and the determined display status of the fourth interface component, update the display results of the third interface component and the fourth interface component in the customized interface.

[0098] Among them, optionally, the determination module 82 is specifically used to: determine in the function library the initial rendering parameters of the multiple interface components described in natural language corresponding to the application interface; through the large language model, based on the set interface rendering protocol, format the initial rendering parameters of the multiple interface components described in natural language to obtain the initial rendering parameters of the multiple interface components described in protocol code, and determine through the large language model in the function library corresponding to the target application program, target function description information matching the task requirement information, wherein the function library contains multiple functions provided by the target application program, each function has corresponding function description information, and the target function description information is the function description information corresponding to the target function among the multiple functions; based on the target function description information, determine the application interface of the target function.

[0099] Among them, optionally, the display module is specifically used to: display the task requirement information in the human-computer dialogue interface; generate reply information corresponding to the task requirement information through a large language model, and display the reply information in the human-computer dialogue interface, wherein the reply information includes task generation prompt information and task cards in natural language; and display the customized interface in response to the user's viewing operation of the task card.

[0100] Figure 8 The device shown can execute the steps in the aforementioned embodiments. For detailed execution process and technical effects, please refer to the description in the aforementioned embodiments, which will not be repeated here.

[0101] In one possible design, the above Figure 8 The structure of the task execution device shown can be implemented as an electronic device. Fig. 9 As shown, the electronic device may include: a processor 91, a memory 92, and a communication interface 93. The memory 92 stores executable codes, and when the executable codes are executed by the processor 91, the processor 91 can at least implement the task execution method provided in the above embodiments.

[0102] In addition, an embodiment of the present invention provides a non-temporary machine-readable storage medium, on which executable code is stored. When the executable code is executed by a processor of an electronic device, the processor can at least implement the task execution method provided in the aforementioned embodiment.

[0103] In addition, an embodiment of the present invention provides a computer program product, which includes: a computer program, when the computer program is executed by a processor of an electronic device, the processor can at least implement the task execution method provided in the above embodiment.

[0104] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. Those of ordinary skill in the art may understand and implement the present invention without creative effort.

[0105] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by adding a necessary general hardware platform, and of course can also be implemented by combining hardware and application programs. Based on such an understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a computer product, and the present invention can be in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0106] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A task execution method, characterized in that: include: Receive task request information input by the user; Determine, in a function library corresponding to a target application, an application program interface of a target function, wherein the target function corresponds to the task requirement information; Determining initial rendering parameters of a plurality of interface components corresponding to the application program interface; generating a customized interface based on initial rendering parameters of the plurality of interface components; Generate and execute a target task according to the task data input by the user through the customized interface; The generating of a customized interface based on the initial rendering parameters of the multiple interface components includes: identifying a keyword corresponding to a first interface component in the task demand information, the first interface component being included in the multiple interface components; determining a display state of the first interface component according to the keyword; updating the initial rendering parameters of the first interface component according to the display state of the first interface component to obtain target rendering parameters of the first interface component; and generating a customized interface based on the target rendering parameters of the first interface component and the initial rendering parameters of the remaining other interface components.

2. The method according to claim 1, characterized in that The generating a customized interface based on the target rendering parameter of the first interface component and the initial rendering parameters of the remaining other interface components includes: Determine the logical linkage between multiple interface components; In response to the target rendering parameter of the first interface component, according to the logical linkage relationship, determine the target rendering parameter of at least one second interface component linked with the first interface component, the at least one second interface component is included in the multiple interface components, and the target rendering parameter of the at least one second interface component is an update result of the initial rendering parameter of the at least one second interface component; A customized interface is generated based on the target rendering parameters of the first interface component, the target rendering parameters of the at least one second interface component, and the initial rendering parameters of the remaining other interface components.

3. The method according to claim 1, characterized in that Also includes: Determining the logical linkage relationship between the multiple interface components; Based on the user's operation behavior on the third interface component, according to the logical linkage relationship, determine the display state of a fourth interface component linked with the third interface component, the third interface component and the fourth interface component being included in the multiple interface components; According to the user's operation behavior on the third interface component and the determined display state of the fourth interface component, the display results of the third interface component and the fourth interface component are updated in the customized interface.

4. The method according to any one of claims 1 to 3, characterized in that The determining of initial rendering parameters of a plurality of interface components corresponding to the application program interface comprises: Determining in the function library initial rendering parameters of the plurality of interface components described in a natural language corresponding to the application program interface; Through the large language model, based on the set interface rendering protocol, the initial rendering parameters of the multiple interface components described in natural language are format converted to obtain the initial rendering parameters of the multiple interface components described in protocol code.

5. The method according to any one of claims 1 to 3, characterized in that: Also includes: In the human-computer dialogue interface, display the task demand information; Generate reply information corresponding to the task request information through a large language model, and display the reply information in the human-computer dialogue interface, wherein the reply information includes task generation prompt information and task cards in natural language; In response to the user's viewing operation on the task card, the customized interface is displayed.

6. The method according to any one of claims 1 to 3, characterized in that Determining the application program interface of the target function in the function library corresponding to the target application program includes: Determine, by means of a large language model, target function description information matching the task requirement information in a function library corresponding to the target application, wherein the function library contains a plurality of functions provided by the target application, each function having corresponding function description information, and the target function description information is function description information corresponding to the target function among the plurality of functions; Based on the target function description information, an application program interface of the target function is determined.

7. An electronic device, characterized in that: include: A memory, a processor, and a communication interface; wherein the memory stores executable code, and when the executable code is executed by the processor, the processor executes the task execution method as described in any one of claims 1 to 6.

8. A non-transitory machine-readable storage medium, characterized in that: The non-transitory machine-readable storage medium stores executable code, and when the executable code is executed by a processor of an electronic device, the processor is caused to execute the task execution method as described in any one of claims 1 to 6.

9. A computer program product, characterized in that include: A computer program, when executed by a processor of an electronic device, causes the processor to execute the task execution method according to any one of claims 1 to 6.

Images