Data generation method, device and system

Abstract

The invention discloses a data generation method, device and system. The method comprises the steps that a plurality of atomic forces are obtained, and the atomic forces are capabilities obtained by disassembling functions of an application; receiving a packaging instruction for the plurality of atomic forces; and in response to the packaging instruction, generating an atomic power list which is used for providing training data for an artificial intelligence large model. According to the embodiment of the invention, the burden of the terminal equipment and the inter-application function multiplexing cost brought by integration of the SDK can be reduced.

Description

Technical Field

[0001] This application belongs to the field of application technology, and in particular relates to a data generation method, apparatus and system. Background Technology

[0002] Terminal devices typically include multiple applications, each providing different functions to the user. As users' demands for terminal devices continue to increase, different applications can reuse functions to provide users with richer features.

[0003] Currently, the common approach to reusing functionality between applications is by integrating the software development kits (SDKs) of other applications into a single application. However, this current method often requires integrating multiple SDKs into a single application, leading to an increase in the package size on the terminal device, which in turn increases the burden on the terminal device and the cost of inter-application functionality reuse. Summary of the Invention

[0004] This application provides a data generation method, apparatus, system, device, storage medium, and vehicle, which can reduce the burden on terminal devices and the cost of inter-application function reuse brought about by integrated SDKs.

[0005] In a first aspect, embodiments of this application provide a data generation method, the method comprising:

[0006] Acquire multiple atomic capabilities, which are capabilities obtained by breaking down the functions of an application;

[0007] Receive a packing instruction for the plurality of atomic capabilities;

[0008] In response to the packaging instruction, an atomic capability list is generated, which is used to provide training data for large artificial intelligence models.

[0009] In one possible implementation, before receiving the packing instruction for the plurality of atomic capabilities, the method further includes:

[0010] Receive instructions to issue commands for a plurality of first atomic capabilities among the plurality of atomic capabilities;

[0011] In response to the release instruction, the plurality of first atomic capabilities are released;

[0012] The receiving of the packing instruction for the plurality of atomic capabilities includes:

[0013] Receive a packing instruction for a plurality of second atomic capabilities among the plurality of first atomic capabilities.

[0014] In one possible implementation, before receiving the packing instruction for a plurality of second atomic capabilities among the plurality of first atomic capabilities, the method further includes:

[0015] Receive instructions to display multiple second atomic capabilities among the plurality of first atomic capabilities;

[0016] In response to the display instruction, a list generation preview content corresponding to the plurality of second atomic capabilities is displayed, the list generation preview content including at least one of the list generation process and the list generation prediction result;

[0017] The step of receiving a packing instruction for multiple second atomic capabilities among the plurality of first atomic capabilities includes:

[0018] Receive a packaging instruction for the plurality of second atomic capabilities when the manifest generation preview content meets the manifest generation conditions.

[0019] In one possible implementation, before receiving the packing instruction for the plurality of atomic capabilities, the method further includes:

[0020] The atomic capabilities are verified to obtain the verification results;

[0021] The receiving of the packing instruction for the plurality of atomic capabilities includes:

[0022] If the verification result is successful, a packing instruction for the plurality of atomic capabilities is received.

[0023] Secondly, embodiments of this application provide a data generation apparatus, the apparatus comprising:

[0024] The acquisition module is used to acquire multiple atomic capabilities, which are capabilities obtained by decomposing the functions of the application;

[0025] The first receiving module is used to receive the packing instruction for the plurality of atomic capabilities;

[0026] A generation module is used to generate an atomic capability list in response to the packaging instruction, the atomic capability list being used to provide training data for large artificial intelligence models.

[0027] Thirdly, embodiments of this application provide a data generation system, including an atomic capability management platform, an object storage platform, a large model module, and a terminal device;

[0028] The atomic capability management platform is used to acquire multiple atomic capabilities, and upon receiving a packaging instruction for the multiple atomic capabilities, to generate an atomic capability list in response to the packaging instruction, and to send the atomic capability list to the object storage platform. The atomic capability is a capability obtained by decomposing the functions of an application.

[0029] The large model module is used to obtain the atomic capability list from the object storage platform and use the atomic capability list as training data for the artificial intelligence large model, so that the artificial intelligence large model can determine the target atomic capability corresponding to the task instruction.

[0030] The terminal device is configured to, upon receiving the task instruction, acquire the target atomic capability determined by the artificial intelligence big data model, and invoke the target atomic capability to execute the task instruction.

[0031] In one possible implementation, the atomic capability management platform is also used for at least one of the following:

[0032] Display first attribute information corresponding to the multiple atomic capabilities, and list generation preview content and second attribute information corresponding to the atomic capability list. The first attribute information includes at least one of the following: the number of atomic capabilities, usage status, version, and upload record uploaded to the atomic capability management platform. The list generation preview content includes at least one of the list generation process and the list generation estimate result. The second attribute information includes at least one of the following: the list generation time, the list generation result, and the list version.

[0033] Display multiple interactive controls corresponding to the multiple atomic capabilities. The multiple interactive controls are used to publish, delist, generate a list of atomic capabilities, and upload the list of atomic capabilities to at least one of the object storage platforms.

[0034] Fourthly, embodiments of this application provide an electronic device, which includes: a processor and a memory storing computer program instructions;

[0035] When the processor executes the computer program instructions, it implements any of the possible implementations of the first aspect described above.

[0036] Fifthly, embodiments of this application provide a computer-readable storage medium storing computer program instructions, which, when executed by a processor, implement the method in any of the possible implementations of the first aspect described above.

[0037] Sixthly, embodiments of this application provide a vehicle that includes at least one of the following:

[0038] Data generation apparatus as described in any embodiment of the second aspect;

[0039] The data generation system as described in any embodiment of the third aspect;

[0040] The electronic device as described in any embodiment of the fourth aspect;

[0041] Computer-readable storage medium as described in any embodiment of the fifth aspect.

[0042] In this embodiment, since atomic capabilities are capabilities obtained by decomposing the functions of an application, by responding to the packaging instruction of multiple atomic capabilities, generating an atomic capability list, and using the atomic capability list as training data for an artificial intelligence model, the artificial intelligence model can learn and understand the usage methods and usage scenarios of multiple atomic capabilities. This can provide users with more user-friendly and intelligent cross-application atomic capability call execution needs, avoid the increase in package size on the terminal device caused by integrating multiple SDKs in one application, reduce the burden on the terminal device, and reduce the cost of inter-application function reuse brought about by integrating SDKs. Attached Figure Description

[0043] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 This is a schematic diagram of the structure of a data generation system provided in an embodiment of this application;

[0045] Figure 2 This is a flowchart illustrating a data generation method provided in an embodiment of this application;

[0046] Figure 3 This is a schematic diagram of a data generation method provided in an embodiment of this application;

[0047] Figure 4 This is a schematic diagram of another data generation method provided in an embodiment of this application;

[0048] Figure 5 This is a schematic diagram of the structure of a data generation device provided in an embodiment of this application;

[0049] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0050] To better understand the above-mentioned objectives, features, and advantages of this application, the solution of this application will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0051] Many specific details are set forth in the following description in order to provide a full understanding of this application, but this application may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some embodiments of this application, and not all embodiments.

[0052] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the term "comprising" or any other variations thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0053] As described in the background section, terminal devices typically include multiple applications, each providing different functionalities to the user. Currently, these applications operate independently, and the capabilities within a single application cannot be quickly reused. Functionality reuse requires integrating shared SDKs. However, this method of integrating corresponding SDKs for inter-application functionality reuse leads to a dramatic increase in package size, placing a significant burden on the terminal device and increasing the cost of functionality reuse. Furthermore, the lack of unified management for the SDKs output by various applications causes numerous problems when reusing new scenarios, impacting the online stability of the terminal device and the online user experience.

[0054] Therefore, to address the problems of the prior art, embodiments of this application provide a data generation method, apparatus, system, device, storage medium, and vehicle. The data generation method can be applied to scenarios where multiple applications in a vehicle infotainment system can reuse their functions.

[0055] The data generation system provided in the embodiments of this application is described below.

[0056] Figure 1 A schematic diagram of the structure of a data generation system provided in an embodiment of this application is shown. Figure 1As shown, the data generation system provided in this embodiment may include an atomic capability management platform 11, an object storage platform 12, a large model module 13, and a terminal device 14. The terminal device 14 may include a first application and a second application. The terminal device may include an in-vehicle infotainment system.

[0057] The atomic capability management platform 11 can be used to acquire multiple atomic capabilities, and when it receives a user's package operation for multiple atomic capabilities, generate a package instruction, and in response to the package instruction, generate an atomic capability list, and send the atomic capability list to the object storage platform. An atomic capability is a capability obtained by decomposing the functions of an application, and the atomic capability includes application tags.

[0058] Here, an atomic capability can be a small, independent, atomic functional module, described and implemented according to a specified format, which can be dynamically loaded and invoked to implement specific business functions or services. It has a clear interface and functional definition, and can be independently invoked and managed. Furthermore, applications can include self-developed applications and third-party applications. Self-developed applications can include applications with interactive interfaces (APPs) and system services without interactive interfaces.

[0059] As an example, an application can include multiple functions. By breaking down the application's functions according to the principle of minimizing functionality, multiple atomic capabilities can be obtained for each application. These atomic capabilities can conform to atomicity standard protocols. Additionally, a Gradle plugin can be integrated into the application. During Continuous Integration / Continuous Deployment (CI / CD), the Gradle plugin can scan bytecode annotations and other information to capture and collect atomic capabilities from the application. After collecting multiple atomic capabilities, each atomic capability can undergo protocol conversion, converting the atomicity standard protocol to the OpenAPI protocol, and then the atomic capability can be uploaded to the atomic capability management platform.

[0060] The Atomic Capability Management Platform 11 can be a platform service built between atomic capabilities and large-scale artificial intelligence models. It collects atomic capabilities during the development phase and provides verification, modification, and release capabilities. This platform can uniformly manage various versions of atomic capabilities in current terminal devices (such as in-vehicle systems), integrate the atomic capability list, and provide this list information to large-scale artificial intelligence models for training and learning, as well as for use in prompt word projects.

[0061] Therefore, to ensure the validity of multiple atomic capabilities in the atomic capability management platform 11, the platform can verify the collected atomic capability information and obtain verification results. Verification may include at least one of integrity verification and format verification. If the verification passes, the atomic capabilities can be categorized according to application tags, resulting in atomic capabilities corresponding to multiple applications. In addition, the atomic capability management platform 11 can also perform processing such as sorting and deduplication on multiple atomic capabilities to ensure their validity.

[0062] In addition, the Atomic Capability Management Platform 11 can provide users with visualization services.

[0063] Based on this, in order to improve the user experience of the atomic capability management platform 11, in some embodiments, the atomic capability management platform 11 may also be used for at least one of the following:

[0064] Display the first attribute information corresponding to multiple atomic capabilities, as well as the list generation preview content and second attribute information corresponding to the atomic capability list. The first attribute information includes at least one of the following: the number of atomic capabilities, usage status, version information, and upload records uploaded to the atomic capability management platform. The list generation preview content includes at least one of the following: the list generation process and the list generation estimate result. The second attribute information includes at least one of the following: the list generation time, the list generation result, and the list version.

[0065] Display multiple interactive controls corresponding to multiple atomic capabilities. These interactive controls can be used to publish, delist, generate an atomic capability list, or upload the atomic capability list to an object storage platform, among other things.

[0066] Here, the Atomic Capability Management Platform 11 can organize and arrange multiple atomic capabilities as needed, and provide users with editable operation controls and monitoring dashboards. Users can monitor the quantity, usage status, version, and upload records of multiple atomic capabilities through the monitoring dashboard. The usage status of an atomic capability indicates whether it has been used for training large-scale AI models, and upload records can be CI / CD scan records. Version information can include version number, changes between adjacent versions, and version release time. The development team can use version information to understand the evolution of atomic capabilities, obtain timely updates, and trace version history.

[0067] After being organized, atomic capabilities can enter a pending state. Users can "publish" or "remove" the atomic capabilities they need, based on their version and definition. The first published atomic capability can enter a pending packaging state, waiting for the administrator of the atomic capability management platform 11 to perform the "packaging" operation.

[0068] Therefore, as an example, multiple first atomic capabilities may include multiple second atomic capabilities. The packaging instruction may be generated in response to an administrator's operation of packaging multiple second atomic capabilities into an atomic capability manifest. Specifically, the administrator may use the OpenAPI protocol to package multiple second atomic capabilities, obtain an atomic capability manifest, and send the atomic capability manifest to the object storage platform 12 (Object Storage Service).

[0069] Building upon the above examples, the Atomic Capability Management Platform 11 can implement **permission management** and **operation logging**, ensuring that only authorized personnel can publish, delist, and package atomic capabilities, thereby guaranteeing system security and data integrity. The Atomic Capability Management Platform 11 also provides a self-service atomic capability **subscription** function, allowing development teams to independently **subscribe** to the atomic capabilities they need. Notifications are sent via communication software when atomic capabilities change, improving development efficiency and flexibility. Furthermore, the manifest generation preview can include the compilation status of atomic capabilities and possible packaging results. The Atomic Capability Management Platform 11 also provides a packaging compilation preview function, allowing development teams to preview the compilation status and possible packaging results of atomic capabilities before packaging, enabling timely identification and resolution of potential problems. The Atomic Capability Management Platform 11 can also record and display packaging history, including secondary attribute information such as the time, version, and packaging result of each packaging session, facilitating traceability and auditing of the packaging process by development teams, ensuring system traceability and security.

[0070] In addition, the large model module 13 can be used to obtain an atomic capability list from the object storage platform 12 and use the atomic capability list as training data for the artificial intelligence large model, so that the artificial intelligence large model can determine the target atomic capabilities corresponding to the task instructions.

[0071] Specifically, the large-scale AI model can learn and understand the usage methods and scenarios of multiple atomic capabilities based on training data. When the first application receives a task instruction, it determines the target atomic capability corresponding to the task instruction. The target atomic capability is the atomic capability among the multiple atomic capabilities that corresponds to the second application. The task instruction can include user instructions issued by the user to the first application. Furthermore, the target atomic capability can be located within the second application or independently of it; this is not limited here.

[0072] Here, the large model module can include large artificial intelligence (AI) models. Large AI models refer to high-performance AI models built with massive training samples and computational resources. They can learn vast amounts of language knowledge, image features, and speech patterns, and can reason and generate outputs similar to humans, finding wide applications in natural language processing, image recognition, and speech recognition. Examples of large AI models include Large Language Models (LLMs), ChatGPT (Chat Generative Pre-trained Transformer), multimodal large models, and multimodal cognitive large models. The large model module 13 can download the latest list of atomic capabilities from the object storage platform 12 and use this list as training data for the large AI model, enabling it to learn and understand the usage and application scenarios of multiple atomic capabilities. Thus, when a task instruction is input into the large AI model, the model can determine the target atomic capability corresponding to the task instruction. The target atomic capability can then be used to execute the task instruction.

[0073] Additionally, terminal device 14 can be used to obtain the target atomic capabilities determined by the artificial intelligence big data model when receiving a task instruction, and to invoke the target atomic capabilities to execute the task instruction. Specifically, when terminal device 14 receives a task instruction through the first application, it can obtain the target atomic capabilities determined by the artificial intelligence big data model, and invoke the target atomic capabilities corresponding to the second application to execute the user instruction.

[0074] As an example, when terminal device 14 receives a user instruction through the first application, it can send the user instruction to the large model module 13. The large model module 13 can determine the task information corresponding to the user instruction through an artificial intelligence large model and send the task information corresponding to the user instruction to terminal device 14. The task information can represent the execution of a target task through a target atomic capability in a second application. Terminal device 14 can parse the task information to obtain the second application, the target atomic capability, and the target task, and call the target atomic capability in the second application to execute the target task, thereby executing the user instruction.

[0075] Therefore, by establishing an atomic capability management platform, the automated collection, unified management, and output of terminal atomic capabilities are achieved. This significantly improves the management efficiency of atomic capabilities, reduces manual operation costs, and ensures the real-time updates and legitimacy of atomic capabilities. Furthermore, the unified atomic capability management platform enables rapid reuse of atomic capabilities across different applications. Compared to traditional integration methods, this reduces the complexity and cost of inter-application integration, avoids the problem of rapidly increasing package size, and thus improves the overall performance and stability of terminal devices. In addition, this embodiment of the application outputs the collected atomic capabilities to downstream algorithm services through a unified protocol, achieving unified output of atomic capabilities. This allows algorithm models to more easily call atomic capabilities, providing more convenient and efficient support for intelligent interaction of terminal devices.

[0076] The data generation method provided in the embodiments of this application is described below. This data generation method can be applied to a data generation system and is executed by the server corresponding to the atomic capability management platform.

[0077] Figure 2 A flowchart illustrating a data generation method provided in an embodiment of this application is shown. Figure 2 As shown, the data generation method provided in this application includes the following steps:

[0078] S210. Acquire multiple atomic capabilities. Atomic capabilities are the capabilities obtained by breaking down the functions of an application.

[0079] S220, Receive packing instructions for multiple atomic capabilities;

[0080] S230, in response to the packing instruction, generates an atomic capability list, which is used to provide training data for large artificial intelligence models.

[0081] In the data generation method of this application embodiment, since atomic capabilities are capabilities obtained by decomposing the functions of an application, by responding to the packaging instruction of multiple atomic capabilities, an atomic capability list is generated, and the atomic capability list is used as training data for an artificial intelligence big model. This enables the artificial intelligence big model to learn and understand the usage methods and usage scenarios of multiple atomic capabilities, and can provide users with more humanized and intelligent cross-application atomic capability call execution needs. This avoids the increase in package size on the terminal device caused by integrating multiple SDKs in one application, reduces the burden on the terminal device, and reduces the cost of inter-application function reuse brought about by integrating SDKs.

[0082] The specific implementation methods for each of the above steps are described below.

[0083] In some embodiments, in S210, acquiring multiple atomic capabilities can be achieved by continuously capturing and collecting the application's corresponding atomic capabilities during the CI / CD process, and uploading the collected atomic capabilities to the atomic capability management platform. The atomic capabilities can reside within the application or be independent of the application; this is not limited here.

[0084] Because the atomic capabilities in the application are continuously captured and collected during the CI / CD process, the atomic capability management platform can obtain different versions of atomic capabilities. For example, for atomic capability A, the atomic capability management platform can obtain versions V1.0.0, V2.0.0, V3.0.0, etc. Specifically, atomic capability A in version V2.0.0 can be an atomic capability obtained by updating the atomic capability in version V1.0.0. Similarly, atomic capability A in version V3.0.0 can be an atomic capability obtained by updating the atomic capability in version V2.0.0.

[0085] Since atomic capabilities are capabilities derived from breaking down the functions of an application, atomic capabilities can be generated in advance before collecting the atomic capabilities corresponding to an application.

[0086] Based on this, in some embodiments, the method may further include the following before collecting the atomic capabilities corresponding to the application:

[0087] Obtain functional information from the application;

[0088] The functional information is broken down into multiple sub-functional information. Each sub-functional information includes a plugin name, multiple commands corresponding to the plugin name, and multiple command parameters corresponding to each command.

[0089] Each sub-function information is encapsulated according to the atomization standard protocol to obtain multiple sub-function information corresponding to function plug-ins, and each function plug-in includes multiple atomic capabilities.

[0090] Here, applications can include self-developed applications and third-party applications. Self-developed applications can include applications with interactive interfaces (APPs) and system services without interactive interfaces. Functional information can be information inherent to the application itself, used to implement the application's capabilities. Functional information can be represented as a Class within the application. For example, an application can have Class A and Class B. Class A can include methods + method A(string):int, and Class B can include methods + method B(type):type. By decomposing the functional information (i.e., multiple Classes), multiple sub-functional information can be obtained. For example, all the information in Class A can be a sub-functional information. Another example is that some information in Class A can be a sub-functional information. Yet another example is that a combination of information from Class A and Class B can be a sub-functional information. Each sub-functional information can include a plugin name, multiple commands corresponding to the plugin name, and multiple command parameters corresponding to each command. The plugin name can be a plugin identifier used to identify the provider of atomic capabilities; the same plugin can carry multiple atomic capabilities. The plugin identifier and application tags can be associated. The command can be an atomic capability identifier, used to identify a specific atomic capability. The command parameters (Content) are the input parameters of the atomic capability; they can be serialized encapsulations of the atomic capability parameters, used to pass necessary parameters when calling the atomic capability interface.

[0091] After obtaining multiple sub-functionality information, each sub-functionality information can be encapsulated according to the atomic standard protocol to obtain functional plugins corresponding to each sub-functionality. The atomic standard protocol can be a protocol derived with reference to the Uniform Resource Identifier (URI) protocol. A standard URI can be: scheme: / / domain / path?key=value. Here, scheme identifies the access method of the resource, domain represents the permission part, including user information, hostname, and port number, path identifies the specific location of the resource, and key=value can be appended to the path to pass additional parameters or information. Based on this, using TPU as the protocol header, the plugin name (including the protocol header) as the permission part, the command corresponding to the plugin name as the command part, and the command parameters corresponding to the command as the parameter part, the protocol can be defined to obtain the atomic standard protocol. TPU is short for Task Processing Unit. A plugin name could be, for example, tpu: / / media. "Media" can be associated with the application name. A command could be, for example, play. Command parameters can include, for example, `pkg=com.xx.music, id=xxx`, representing the location and identifier of the music to be played. An atomic standard protocol can be, for example, `tpu: / / media / play?pkg=com.xx.music&id=xxx`. This is because the atomic standard protocol can include the plugin name, command, and command parameters.

[0092] It should be noted that each functional plugin can include multiple atomic capabilities. For example, if a functional plugin is named "Music Plugin," and its commands include playing music and setting sound effects, with the relevant parameters for setting sound effects being sound effect A and music B, then this functional plugin can include multiple atomic capabilities: setting sound effect A, and playing music B.

[0093] By pre-generating atomic capabilities, it is possible to ensure mutual cooperation among multiple applications when implementing inter-application function reuse based on multiple atomic capabilities, thereby improving communication efficiency.

[0094] In some embodiments, in S220 and S230, the packaging instruction may be an instruction generated in response to a user's packaging operation on multiple atomic capabilities in the atomic capability management platform. In response to the user's packaging operation, the atomic capability management platform may generate an atomic capability list including multiple atomic capabilities. The user's packaging operation may, for example, be an action by the user clicking a "package" control.

[0095] Therefore, in order to ensure the validity of the atomic capability list, in some embodiments, before S220 above, the following may also be included:

[0096] The atomic capabilities are verified, and the verification results are obtained.

[0097] Based on this, the aforementioned S220 may specifically include:

[0098] If the verification result is successful, receive the packing instruction for multiple atomic capabilities.

[0099] Here, validation can include at least one of integrity validation and format validation. If the validation of an atomic capability passes, the user can select multiple first atomic capabilities from the validated atomic capabilities to publish. Among the published first atomic capabilities, the user can select multiple second atomic capabilities as needed to package them into a list of atomic capabilities. In addition, the user can also select multiple third atomic capabilities from the published first atomic capabilities to delist.

[0100] Therefore, in order to successfully generate the atomic capability list and ensure its validity, in some embodiments, before S220 above, the following may be included:

[0101] Receive instructions to issue commands for multiple first atomic capabilities among multiple atomic capabilities;

[0102] In response to the release command, multiple first-atom capabilities are released.

[0103] Here, multiple first atomic capabilities can be the same as multiple atomic capabilities, or they can be a subset of multiple atomic capabilities. Furthermore, the publish command can be a command generated in response to a user's publish operation on multiple first atomic capabilities within the atomic capability management platform. In response to the user's publish operation, the atomic capability management platform can publish multiple first atomic capabilities. For example, the user's publish operation could be the user clicking the "Publish" control.

[0104] Based on this, the aforementioned S220 may specifically include:

[0105] Receive a packing instruction for multiple second atomic capabilities among multiple first atomic capabilities. The multiple second atomic capabilities may be the same as the multiple first atomic capabilities, or may be a subset of the multiple first atomic capabilities.

[0106] By packaging multiple published second atomic capabilities, we can successfully generate an atomic capability list and ensure its validity.

[0107] In addition, to facilitate users in promptly identifying and resolving potential problems during the list generation process and to ensure the effectiveness of the atomic capability list, in some embodiments, before receiving the packing instruction for multiple second atomic capabilities among multiple first atomic capabilities, the following may also be included:

[0108] Receive instructions to display multiple second-atom capabilities among multiple first-atom capabilities;

[0109] In response to the display command, display the inventory generation preview content corresponding to multiple second atomic capabilities. The inventory generation preview content includes at least one of the inventory generation process and the inventory generation estimate result.

[0110] Based on this, the aforementioned instruction to receive a packing of multiple second atomic capabilities among multiple first atomic capabilities may specifically include:

[0111] Receives a packing instruction for multiple second atomic capabilities when the manifest generation preview content meets the manifest generation conditions.

[0112] Here, the display command can be a command generated in response to a user's display operation on multiple second atomic capabilities within the atomic capability management platform. In response to the user's display operation, the atomic capability management platform can display a preview of the list corresponding to the multiple second atomic capabilities. For example, the user's display operation could be the user clicking the "Display" control.

[0113] As an example, after a user selects multiple second atomic capabilities, before packaging, the atomic capability management platform can display the compilation status of the atomic capabilities (i.e., the manifest generation process) and the possible packaging results (manifest generation estimate) so that users can promptly identify and resolve potential problems in the manifest generation process and ensure the effectiveness of the atomic capability manifest.

[0114] In addition, the steps in the embodiments of this application can be found in the above description, and will not be repeated in detail here.

[0115] With the rapid deployment of generative applications across various business sectors, in order to accelerate the deployment of generative applications, a comprehensive and unified management service for plug-in atomic capabilities is provided to the business domain. This enables end-to-end calls from business-side capabilities to algorithms, task processing engines, and business processes, improving the efficiency of atomic capability development and management. Furthermore, to facilitate the external output, unified configuration, and management of atomic capabilities, an atomic capability management platform is used to uniformly manage the collection and exposure of atomic capabilities of various versions. It is also responsible for managing, verifying, and publishing generated atomic capabilities and their corresponding capability lists, ensuring that the business side can conveniently, stably, and promptly call the atomic capabilities provided by the business-side applications.

[0116] To better describe the overall solution, some specific examples are given based on the above embodiments.

[0117] For example, a schematic diagram of a data generation method provided in an embodiment of this application can be as follows: Figure 3 As shown.

[0118] like Figure 3 As shown, by breaking down the functionality of application A, we can obtain a functional plugin consisting of multiple atomic capabilities, namely TPUPlugin. TPUPlugin conforms to the TPU specification protocol. TPU is short for Task Processing Unit. The Task Processing Unit is located in the terminal device and acts as a "bridge" and central hub between tasks and atomic capabilities. During specific business operations, it manages the lifecycle of atomic capabilities and distributes tasks to specific atomic capabilities based on the user's interaction with the terminal device, responding to user interactions.

[0119] As an example, a terminal device may include a first application, a second application, a task parsing module, and a task processing engine. The second application may include target atomic capabilities. Specifically, after receiving a user instruction through the first application, the terminal device can send the user instruction to a large model module. The large model module uses an artificial intelligence model to determine the task information corresponding to the user instruction and sends this task information to the terminal device. The task information can represent the execution of a target task through the target atomic capabilities in the second application. The terminal device can parse the task information through the task parsing module to obtain the second application, the target atomic capabilities, and the target task. Furthermore, the task processing engine can then invoke the target atomic capabilities in the second application to execute the target task, thereby executing the user instruction.

[0120] In addition, Gradle plugins can be integrated into application A. During the compilation phase, Gradle plugins can capture and collect atomic capabilities within application A.

[0121] In the Continuous Integration / Continuous Deployment (CI / CD) process, the Gradle plugin can scan bytecode annotations and other information during the compilation phase to collect atomic capabilities from multiple applications. These atomic capabilities are then combined into a format conforming to the OpenAPI 3.0 protocol and uploaded to the atomic capability management platform. The platform can perform preliminary verification, organization, and deduplication of the collected atomic capability information, and categorize it by application.

[0122] Once the atomic capability verification passes, the system enters the "ready to be released" state. Users can select multiple first atomic capabilities from the multiple atomic capabilities awaiting release for release, and from the multiple released first atomic capabilities, select multiple second atomic capabilities as needed to package them into an atomic capability list. Furthermore, users can also select multiple third atomic capabilities from the multiple released first atomic capabilities to remove them from the platform. Users can package multiple versions of the atomic capability list as needed. For example, at the first moment, a user can package the atomic capability list based on multiple atomic capabilities to obtain version A. At the second moment, the atomic capabilities have been updated; therefore, the user can package the atomic capability list based on multiple updated atomic capabilities to obtain version B.

[0123] Users can upload a list of atomic capabilities to an object storage platform so that the large model module can download the latest list of atomic capabilities from the object storage platform. Multiple atomic capabilities in the list can be used as training data for the large AI model so that the large AI model can learn and understand the usage and application scenarios of multiple atomic capabilities. When a user issues a user command through the first application, the target atomic capability corresponding to the user command can be determined.

[0124] Based on this, a schematic diagram of another data generation method provided in the embodiments of this application can be shown as follows. Figure 4 As shown.

[0125] Therefore, this application embodiment, by automatically collecting atomic capabilities of terminal devices (such as in-vehicle systems), can ensure that the capabilities of terminal devices are perceived by external entities in a timely and accurate manner. Unified management of the collected atomic capabilities ensures the legitimacy of all atomic capabilities, laying the foundation for unified scheduling of capabilities across the entire domain. By aggregating the capabilities of terminal devices into a single atomic capability list and outputting it to a large-scale artificial intelligence model, a standardized channel for post-consumption of capabilities can be provided.

[0126] Based on the data generation method provided in the above embodiments, this application also provides specific implementations of the data generation apparatus. Please refer to the following embodiments.

[0127] like Figure 5 As shown, the data generation apparatus 500 provided in this application embodiment includes the following modules:

[0128] The acquisition module 510 is used to acquire multiple atomic capabilities, which are capabilities obtained by breaking down the functions of an application.

[0129] The first receiving module 520 is used to receive a packing instruction for multiple atomic capabilities;

[0130] The generation module 530 is used to generate an atomic capability list in response to the packaging instruction. The atomic capability list is used to provide training data for large artificial intelligence models.

[0131] The data generation device 500 described above will be explained in detail below:

[0132] In some embodiments, the data generation apparatus 500 may further include:

[0133] Based on this, the first receiving module 520 may specifically include:

[0134] The second receiving module is used to receive the release instructions for multiple first atomic capabilities among multiple atomic capabilities;

[0135] The publish module is used to publish multiple first atomic capabilities in response to publish commands.

[0136] Based on this, the first receiving module 520 may specifically include:

[0137] The first receiving submodule is used to receive packing instructions for multiple second atomic capabilities among multiple first atomic capabilities.

[0138] In some embodiments, the data generation apparatus 500 may further include:

[0139] The third receiving module is used to receive a display instruction for multiple second atomic capabilities among multiple first atomic capabilities before receiving a packing instruction for multiple second atomic capabilities among multiple first atomic capabilities.

[0140] The display module is used to respond to display commands and display the manifest generation preview content corresponding to multiple second atomic capabilities. The manifest generation preview content includes at least one of the manifest generation process and manifest generation prediction results.

[0141] Based on this, the first receiving submodule may specifically include:

[0142] The receiving unit is used to receive the packing instruction for multiple second atomic capabilities when the manifest generation preview content meets the manifest generation conditions.

[0143] In some embodiments, the data generation apparatus 500 may further include:

[0144] The verification module is used to verify the atomic capabilities before receiving the packing instruction for multiple atomic capabilities, and obtain the verification result.

[0145] Based on this, the first receiving module 520 may specifically include:

[0146] The second receiving submodule is used to receive a packing instruction for multiple atomic capabilities if the verification result is successful.

[0147] In the data generation apparatus of this application embodiment, since atomic capabilities are capabilities obtained by disassembling the functions of an application, by responding to the packaging instruction of multiple atomic capabilities, an atomic capability list is generated, and the atomic capability list is used as training data for an artificial intelligence big model, enabling the artificial intelligence big model to learn and understand the usage methods and usage scenarios of multiple atomic capabilities. This can provide users with more user-friendly and intelligent cross-application atomic capability call execution requirements, avoid the increase in package size in the terminal device caused by integrating multiple SDKs in one application, reduce the burden on the terminal device, and reduce the cost of inter-application function reuse brought about by integrating SDKs.

[0148] Based on the data generation method provided in the above embodiments, this application also provides specific implementation methods for electronic devices. Figure 6 A schematic diagram of an electronic device 600 provided in an embodiment of this application is shown.

[0149] Electronic device 600 may include processor 610 and memory 620 storing computer program instructions.

[0150] Specifically, the processor 610 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0151] Memory 620 may include mass storage for data or instructions. For example, and not limitingly, memory 620 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where suitable, memory 620 may include removable or non-removable (or fixed) media. Where suitable, memory 620 may be internal or external to electronic device 600. In a particular embodiment, memory 620 is a non-volatile solid-state memory.

[0152] Memory may include read-only memory (ROM), random access memory (RAM), disk storage media devices, optical storage media devices, flash memory devices, and electrical, optical, or other physical / tangible memory storage devices. Therefore, typically, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to the first aspect of this application.

[0153] The processor 610 implements any of the data generation methods described in the above embodiments by reading and executing computer program instructions stored in the memory 620.

[0154] In one example, electronic device 600 may further include communication interface 630 and bus 640. For example, Figure 6 As shown, the processor 610, memory 620, and communication interface 630 are connected via bus 640 and communicate with each other.

[0155] The communication interface 630 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0156] Bus 640 includes hardware, software, or both, that couples components of an electronic device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 640 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, any suitable bus or interconnect is contemplated herein.

[0157] For example, the electronic device 600 can be a mobile phone, tablet computer, laptop computer, handheld computer, in-vehicle electronic device, ultra-mobile personal computer (UMPC), netbook, or personal digital assistant (PDA), etc.

[0158] The electronic device can execute the data generation method described in the embodiments of this application, thereby achieving the combination Figures 1 to 5 The data generation system, method, and apparatus described.

[0159] Furthermore, in conjunction with the data generation methods in the above embodiments, this application embodiment can provide a computer-readable storage medium for implementation. This computer-readable storage medium stores computer program instructions; when executed by a processor, these computer program instructions implement any of the data generation methods in the above embodiments.

[0160] In addition, this application embodiment also provides a vehicle, which may include at least one of the following:

[0161] Data generation apparatus as described in any embodiment of the second aspect;

[0162] The data generation system as described in any embodiment of the third aspect;

[0163] The electronic device as described in any embodiment of the fourth aspect;

[0164] Computer-readable storage media as described in any embodiment of the fifth aspect. Further details will not be provided here.

[0165] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0166] The functional blocks shown in the above-described structural diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0167] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0168] The aspects of this application have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by dedicated hardware performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0169] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A data generating method characterized by comprising: The method comprises: obtaining a plurality of atomic capabilities, the atomic capabilities being capabilities obtained by disassembling functions of an application; receiving a packaging instruction for the plurality of atomic capabilities; in response to the packaging instruction, generating an atomic capability list, the atomic capability list being used to provide training data for an artificial intelligence large model.

2. The method of claim 1, wherein, Before the receiving the packaging instruction for the plurality of atomic capabilities, the method further comprises: receiving a publishing instruction for a plurality of first atomic capabilities in the plurality of atomic capabilities; in response to the publishing instruction, publishing the plurality of first atomic capabilities; the receiving the packaging instruction for the plurality of atomic capabilities comprises: receiving a packaging instruction for a plurality of second atomic capabilities in the plurality of first atomic capabilities.

3. The method of claim 2, wherein, Before the receiving the packaging instruction for the plurality of second atomic capabilities in the plurality of first atomic capabilities, the method further comprises: receiving a display instruction for the plurality of second atomic capabilities in the plurality of first atomic capabilities; in response to the display instruction, displaying list generation preview content corresponding to the plurality of second atomic capabilities, the list generation preview content comprising at least one of a list generation process and a list generation estimated result; the receiving the packaging instruction for the plurality of second atomic capabilities in the plurality of first atomic capabilities comprises: receiving a packaging instruction for the plurality of second atomic capabilities when the list generation preview content meets a list generation condition.

4. The method according to any one of claims 1 to 3, characterized in that, Before the receiving the packaging instruction for the plurality of atomic capabilities, the method further comprises: verifying the atomic capabilities to obtain a verification result; the receiving the packaging instruction for the plurality of atomic capabilities comprises: in a case where the verification result is a verification pass, receiving the packaging instruction for the plurality of atomic capabilities.

5. A data generating apparatus characterized by comprising: The apparatus comprises: an obtaining module configured to obtain a plurality of atomic capabilities, the atomic capabilities being capabilities obtained by disassembling functions of an application; a first receiving module configured to receive a packaging instruction for the plurality of atomic capabilities; a generating module configured to, in response to the packaging instruction, generate an atomic capability list, the atomic capability list being used to provide training data for an artificial intelligence large model.

6. A data generating system characterized by comprising: The apparatus comprises an atomic capability management platform, an object storage platform, a large model module, and a terminal device; the atomic capability management platform is configured to obtain a plurality of atomic capabilities, and in a case where a packaging instruction for the plurality of atomic capabilities is received, in response to the packaging instruction, generate an atomic capability list, and send the atomic capability list to the object storage platform, the atomic capabilities being capabilities obtained by disassembling functions of an application; the large model module is configured to obtain the atomic capability list from the object storage platform, and use the atomic capability list as training data for an artificial intelligence large model, so that the artificial intelligence large model determines a target atomic capability corresponding to a task instruction; the terminal device is configured to, in a case where the task instruction is received, obtain the target atomic capability determined by the artificial intelligence large model, and invoke the target atomic capability to execute the task instruction.

7. The system of claim 6, wherein, The atomic capability management platform is further configured to at least one of: display first attribute information corresponding to the plurality of atomic capabilities, and list generation preview content and second attribute information corresponding to the atomic capability list, the first attribute information including at least one of a number, a usage state, a version, and an upload record uploaded to the atomic capability management platform of the atomic capability, the list generation preview content including at least one of a list generation process and a list generation estimated result, and the second attribute information including at least one of a list generation time, a list generation result, and a list version of the atomic capability list; display a plurality of interaction controls corresponding to the plurality of atomic capabilities, the plurality of interaction controls being used for at least one of publishing, delisting, generating the atomic capability list, and uploading the atomic capability list to the object storage platform.

8. An electronic device, comprising: The electronic device comprises a processor and a memory storing computer program instructions; The processor executes the computer program instructions to implement the data generation method of any one of claims 1-4.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer program instructions, and the computer program instructions are executed by the processor to implement the data generation method of any one of claims 1-4.

10. A vehicle characterized by comprising: comprising at least one of: The data generation apparatus of claim 5; The data generation system of any one of claims 6-7; The electronic device of claim 8; The computer readable storage medium of claim 9.

Images