Automated unit reliable curing method, apparatus, equipment and storage medium

By defining rules, conducting trial runs, and standardizing the automated units generated by the large language model, the problem of unstable large language model service functions was solved, and long-term reliable automated units were solidified, ensuring the stability and reliability of the service.

CN123068253APending Publication Date: 2026-11-10BEIJING DARK SIDE OF THE MOON TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611254053.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-18
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

The service functions generated by large language models are difficult to use stably in the long term, and there are problems such as unexecutable code, excessively broad permissions, unverifiable output results, and state drift.

Method used

By defining the original automation unit rules, conducting trial runs and standardization conversions, and ensuring that the trial run results meet the standards before verification, the original automation units are solidified into target automation units, ensuring that their inputs, outputs, and execution processes meet the requirements.

Benefits of technology

The service function of generating large language models has been enabled to be used stably for a long time in the form of automated units, ensuring the reliable solidification and stability of the service.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123068253A_ABST
    Figure CN123068253A_ABST
Patent Text Reader

Abstract

This application relates to the field of artificial intelligence automation technology, and in particular to a method, apparatus, device, and storage medium for the reliable solidification of automated units. The method includes: determining original automated unit rules and corresponding original automated units based on an acquired automation target and a preset large language model system; conducting a trial run of the original automated units in a preset isolated environment and determining the trial run results; performing a normalization transformation on the trial run results to obtain a normalized result; and solidifying the original automated unit as a target automated unit in response to the normalization result passing result verification. This application facilitates the long-term stable use of the large language model generation service function in the form of automated units.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of artificial intelligence automation technology, and in particular to a reliable method, apparatus, device and storage medium for the solidification of automated units. Background Technology

[0002] With the continuous improvement of large language model technology, current large language model systems can generate service functions such as reminders, scripts, workflows, web artifacts, data dashboards, or temporary agent tasks based on natural language. These service functions can execute corresponding functions when trigger conditions are met to meet user needs; for example, this function can collect news in a specific field for users every day.

[0003] However, large language models may generate unworkable or overly permissive code; input forms and execution parameters may lack a common contract; output results may be free text rather than verifiable objects; the result display page and execution entry permissions may be mixed up; failed runs may overwrite the previous valid results; state drift may occur between background scheduled tasks and user-visible automated objects; thus, the service functions generated by large language models are difficult to use stably in the long term. Summary of the Invention

[0004] To facilitate the long-term stable use of the large language model generation service function, this application provides an automated unit trusted solidification method, apparatus, device, and storage medium.

[0005] In a first aspect, this application provides an automated unit trusted solidification method, comprising: Based on the acquired automation goals and the preset large language model system, the original automation unit rules and the original automation units corresponding to the original automation unit rules are determined. The original automated unit was tested in a pre-set isolated environment to determine the test results; The trial operation results are then normalized to obtain normalized results. In response to the normalization result passing the result verification, the original automation unit is solidified into the target automation unit.

[0006] Secondly, this application provides an automated unit-based reliable curing device, comprising: The original unit determination module is used to determine the original automation unit rules and the original automation units corresponding to the original automation unit rules based on the acquired automation target and the preset large language model system. The trial run result determination module is used to conduct a trial run of the original automated unit in a preset isolated environment and determine the trial run result; The standardized result generation module is used to perform standardized transformation on the trial operation results to obtain standardized results; The target unit solidification module is used to solidify the original automation unit into a target automation unit in response to the normalization result passing the result verification.

[0007] Thirdly, this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the method described above.

[0008] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the above-described method.

[0009] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.

[0010] The aforementioned method, apparatus, device, and storage medium for reliable solidification of automated units determine the original automated unit rules and the original automated units corresponding to the original automated unit rules based on the acquired automated target and a preset large language model system; conduct trial runs of the original automated units in a preset isolated environment to determine the trial run results; perform normalization transformation on the trial run results to obtain normalized results; and solidify the original automated units as target automated units in response to the normalization results passing result verification. Through the above implementation, after generating the original automated unit rules and corresponding original automated units through the large language model system, the original automated unit rules can constrain the trial operation process of the original automated units, so that the input, output, execution and other processes of the original automated units meet the corresponding requirements. Then, the trial operation results of the original automated units are converted into normalized results that can be verified. Finally, the normalized results are verified to determine whether the various trial operation indicators of the original automated units meet the requirements. When the verification is passed, that is, when the various trial operation indicators meet the requirements, the original automated unit at this time is determined to be a long-term stable automated unit, that is, the target automated unit. By generating first and then verifying, it is ensured that the final determined target automated unit is a long-term stable automated unit. This makes it easy to realize that the service functions generated by the large language model can be used stably for a long time in the form of automated units.

[0011] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent from the following description. Attached Figure Description

[0012] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a flowchart of an automated unit trusted solidification method provided in the embodiments of this application; Figure 2 This is a schematic diagram of the structure of an automated unit reliable curing device provided in the embodiments of this application; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application; Figure 4 This is an internal structural diagram of a computer-readable storage medium provided in an embodiment of this application. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of this disclosure clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this disclosure.

[0015] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings herein are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, apparatus, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0016] In this article, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0017] Example 1 Figure 1This is a flowchart of an automated unit trusted solidification method provided in Embodiment 1 of this application, with reference to... Figure 1 The method can be executed by a device that performs the method, which can be implemented in software and / or hardware, and the method includes: S110. Based on the acquired automation target and the preset large language model system, determine the original automation unit rules and the original automation units corresponding to the original automation unit rules.

[0018] It should be noted that with the continuous advancement of large language model technology, large language models have acquired a strong ability to automatically serve users and can provide corresponding automated services according to user needs, such as periodically outputting recommended news in specific fields. However, such automated services are currently difficult to maintain in a long-term and stable manner to provide users with long-term and stable services. This embodiment intends to solidify such automated service processes in the form of functional units to provide users with long-term and stable services, and such functional units are referred to as automation units.

[0019] To ensure the long-term stability of the automated unit, the triggering, input, execution, and output processes of the automated unit must be clearly defined, and these defined processes are recorded as the original automated unit rules. To generate these original automated unit rules, this embodiment pre-defines a corresponding large language model system. For example, this large language model system can be a specific large language model, such as Kimi, without further limitation. It can also be an agent applying a specific large language model. When a user wants to generate an automated unit that can stably perform a certain function over a long period, they can first input the corresponding automation goal into the large language model system in the form of text and / or voice. For example, the automation goal is "to obtain 10 news articles in a certain field every day and provide stock recommendations." The large language model system processes this automation goal and can first generate unit specifications for standardizing the automated unit, which are recorded as the original automated unit rules.

[0020] For example, the original automation unit rule includes multiple fields, specifically: identity and lifecycle, trigger contract, input contract, execution contract, output contract, capability surface strategy, result materialization strategy, scheduling synchronization strategy, and auditing and security strategy. Among these, identity and lifecycle are used to determine the automation unit's identity, domain, current state, and version, and are used for auditing and concurrency control. The trigger contract describes when the automation unit can run; triggers can be manual, timed, event-driven, conditional, or external webhooks. The input contract validates the runtime input, restricting top-level objects, required fields, field types, and default values. The execution contract describes the execution using an agent, code, workflow, remote function, or container, and limits available tools and resources. The output contract validates the execution result object, preventing arbitrary text or incomplete objects from entering the latest (any output from the most recent run). The capability surface strategy declares which interfaces or APIs are read-only, which can submit input, and which can modify configurations. The result materialization strategy declares successful results, failed results, and the latest... Pointers, history retention and overwrite rules, and scheduling synchronization policies are used to declare how derived scheduling objects are created, paused, deleted and reconciled. Auditing and security policies are used to record initiators, permissions, key usage, network access boundaries and auditing requirements.

[0021] Furthermore, given the original automated unit rules, the large language model system can also generate automated units that match the original automated unit rules, denoted as the original automated unit.

[0022] S120. Conduct a trial run of the original automated unit in a preset isolated environment and determine the trial run results.

[0023] It should be noted that when the original automation unit is initially generated, it cannot be guaranteed that the original automation unit generated by the large language model system will be able to run normally. Therefore, before using the original automation unit, it is necessary to perform a trial run (dry-run).

[0024] In order to ensure that the execution link of the trial run is consistent with the real link, but does not pollute the production results, does not overwrite the currently valid latest results, and does not trigger a sandboxed environment that causes side effects to external systems, this embodiment intends to perform the trial run of the original automation unit in a preset isolation environment, and record the running results generated by the trial run of the original automation unit in the isolation environment as the trial run results.

[0025] It should be noted that the trial run in the isolated environment is to verify the actual feasibility of the unit in terms of code integrity, permission boundaries, resource consumption, and output contract before the automated unit generated by the large language model system is formally solidified into an object that can be scheduled for a long time and affect the production state. This is done in a sandboxed environment that is completely consistent with the real execution chain (including real runtime, real input, and real output verification paths) but does not pollute the production results, overwrite the currently valid latest results, or trigger side effects on external systems. This way, behavioral defects (such as inoperable code, unauthorized tool calls, or illegal output structures) that cannot be detected in the specification verification stage of the large language model system's output can be intercepted at the draft stage, avoiding the uncertainty of generative AI from being directly solidified into runtime failures. It also ensures that even if the trial run fails, the existing formal automated units and the latest successful results visible to users remain stable, thus truly achieving reliable solidification rather than just "format verification".

[0026] S130. The trial operation results are standardized to obtain standardized results.

[0027] It should be noted that by verifying the trial operation results, it can be determined whether the original automation unit is qualified to be solidified, that is, whether the original automation unit is qualified to be used stably for a long time. When verifying the trial operation results, the results need to be in a standardized format required for verification. Therefore, the trial operation results need to be standardized and converted first.

[0028] In order to achieve the standardized conversion of the trial operation results, this embodiment has a pre-set structured output adapter. This adapter is used to convert the trial operation results into a format that conforms to the subsequent verification specifications, and the result after the trial operation results are converted is recorded as the standardized result.

[0029] S140. In response to the normalization result passing the result verification, the original automation unit is solidified into the target automation unit.

[0030] It should be noted that the original automation unit has corresponding original automation unit rules, which include an output contract. The purpose of the output contract is to verify the execution result object and prevent any text or incomplete object from entering the latest (any output of the most recent run). Through the output contract, the normalized result can be verified to check whether the normalized result corresponding to the trial run result meets the output requirements in the original automation unit rules. If it meets the requirements, it means that the original automation unit has the capability to officially provide users with long-term stable services.

[0031] Specifically, the output contract in the original automation unit rules corresponding to the original automation unit is obtained. Using the output contract, the normalized result corresponding to the running result is verified. If the normalized result satisfies the output contract, the original automation unit is solidified into the target automation unit.

[0032] It should be noted that this embodiment determines the original automation unit rules and the original automation units corresponding to the original automation unit rules based on the acquired automation target and the preset large language model system; it performs a trial run on the original automation units in a preset isolation environment to determine the trial run results; it performs a normalization transformation on the trial run results to obtain normalized results; and in response to the normalization results passing the result verification, it solidifies the original automation units into target automation units. Through the above implementation, after generating the original automated unit rules and corresponding original automated units through the large language model system, the original automated unit rules can constrain the trial operation process of the original automated units, so that the input, output, execution and other processes of the original automated units meet the corresponding requirements. Then, the trial operation results of the original automated units are converted into normalized results that can be verified. Finally, the normalized results are verified to determine whether the various trial operation indicators of the original automated units meet the requirements. When the verification is passed, that is, when the various trial operation indicators meet the requirements, the original automated unit at this time is determined to be a long-term stable automated unit, that is, the target automated unit. By generating first and then verifying, it is ensured that the final determined target automated unit is a long-term stable automated unit. This makes it easy to realize that the service functions generated by the large language model can be used stably for a long time in the form of automated units.

[0033] Example 2 This application provides a second embodiment of a trusted method for solidifying automated units. This method optimizes the "determining the original automated unit rules and the original automated units corresponding to the original automated unit rules based on the acquired automated target and the preset large language model system" in the first embodiment. It should be noted that for parts not described in detail in this embodiment, please refer to the descriptions in other embodiments. The method includes: S211. Based on the pre-set large language model system processing, the obtained automation target is used to determine the original automation unit rules.

[0034] It should be noted that before using the preset large language model system to generate automated units, the triggering, input, execution, and output processes of the automated unit to be generated must be clearly defined. That is, the unit rules corresponding to the automated unit must be generated first, and the unit rules are recorded as the original automated unit rules draft UnitSpec.

[0035] Among them, the large language model system can process the automation goals issued by the user, thereby generating the original automation unit rules corresponding to the automation units to be generated by the automation goals; for example, the automation goal is "to obtain 10 news articles in a certain field every day and provide stock recommendations".

[0036] S212. In response to the original automation unit rule passing the rule verification, the original automation unit rule is solidified into the target automation unit rule.

[0037] It should be noted that the original automation unit rules include multiple fields, such as: identity and lifecycle, trigger contract, input contract, execution contract, output contract, capability surface strategy, result materialization strategy, scheduling synchronization strategy, and auditing and security strategy. To ensure the standardization and correctness of each field in the original automation unit rules, the initially generated original automation unit rules need to be validated.

[0038] In order to realize the rule verification of the original automation unit rule, this embodiment has a pre-set corresponding rule verification tool JSON Schema. The rule verification tool can be used to verify each field of the original automation unit rule. If all fields pass the verification, it means that the corresponding original automation unit rule has passed the rule verification, which means that the original automation unit rule has the qualification to guide the corresponding automation unit to perform processes such as triggering, input, execution and output in the subsequent specification. At this time, the original automation unit rule can be used as the formal automation unit rule, that is, the original automation unit rule can be solidified into the target automation unit rule.

[0039] S213. Based on the large language model system, process the target automation unit rules to determine the original automation unit.

[0040] The generated target automation unit rules are used to input into the large language model system to guide the system to generate corresponding automation units, and these automation units are recorded as the original automation units. The original automation units are constrained by the corresponding target automation unit rules during execution.

[0041] Through the above implementation, after generating the original automation unit rules, this embodiment does not directly use them as the unit rules for the corresponding automation unit in the guide specification. Instead, it first performs rule verification to ensure that the initially generated original automation unit rules themselves conform to the specification. In this way, the accuracy of the automation unit generated based on the verified original automation unit rules (i.e., the target automation unit rules) can be guaranteed.

[0042] S220. Conduct a trial run of the original automated unit in a preset isolated environment and determine the trial run results.

[0043] S230. The trial operation results are standardized to obtain standardized results.

[0044] S240. In response to the normalization result passing the result verification, the original automation unit is solidified into the target automation unit.

[0045] Example 3 This application provides a method for the reliable solidification of automated units in Embodiment 3, which supplements the method shown in Embodiment 2. It should be noted that for parts not detailed in this embodiment, please refer to the descriptions in other embodiments. The method includes: S311. Based on the preset large language model system processing, the obtained automation target is used to determine the original automation unit rules.

[0046] S312. In response to the original automation unit rule passing the rule verification, the original automation unit rule is solidified into the target automation unit rule.

[0047] S313. Based on the large language model system, process the target automation unit rules to determine the original automation unit.

[0048] S320. Conduct a trial run of the original automated unit in a preset isolated environment and determine the trial run results.

[0049] S330. The trial operation results are standardized and converted to obtain standardized results.

[0050] S340. In response to the normalization result passing the result verification, the original automation unit is solidified into the target automation unit.

[0051] S350. Determine the capability surface of the target automation unit for each exposed target.

[0052] It should be noted that after the target automation unit is generated, in order to transform it from a static, standardized object in the background into an interactive service that can be actually put into production, enabling different roles (such as end users viewing results, operators submitting input, and administrators adjusting configurations) or external upstream and downstream systems to interact with the automation unit in a standardized and auditable manner without directly contacting the underlying execution engine or scheduling system, the interaction entry is divided into different permission boundaries such as result browsing (read-only), input execution (controlled submission), configuration management (authorized modification), and system scheduling (derived tasks). This provides availability, integrability, and multi-user collaboration capabilities while enforcing the principle of least privilege, preventing any aspect from overstepping its authority to trigger operation, tampering with data, or drifting scheduling. Ultimately, this ensures that the automation unit can be fully utilized in real business scenarios while always remaining under the control of host isolation verification and unified standards. To this end, this embodiment needs to first determine the capability aspects of the target automation unit facing each exposed target, where the exposed targets can be other users or other systems.

[0053] Specifically, this capability surface is an abstraction of the interaction surface exposed by the target automation unit to the user or other systems. Its specific forms can be: HTML iframe, WebView, native components, plugin panels, CLI, API endpoint or other interaction carriers.

[0054] For example, the types of capability surfaces include: result browsing surface, input execution surface, configuration management surface, and system scheduling surface. When a capability sends an execution request to the host, the host performs at least the following checks: whether the source window or caller matches the current automation unit; whether the capability surface is declared as executable; whether the message type is an allowed submitInput or equivalent protocol; whether the payload is an object and satisfies inputContract; whether activeRun exists or whether the concurrency strategy is satisfied; and whether the caller has execution permissions. If any check fails, the host refuses to create a run and returns a structured error.

[0055] Example 4 This application provides a method for the reliable solidification of automated units in Embodiment 4, which supplements the method shown in Embodiment 2. It should be noted that for parts not detailed in this embodiment, please refer to the descriptions in other embodiments. The method includes: S411. Based on the preset large language model system processing, the obtained automation target is used to determine the original automation unit rules.

[0056] S412. In response to the original automation unit rule passing the rule verification, the original automation unit rule is solidified into the target automation unit rule.

[0057] S413. Based on the large language model system, process the target automation unit rules to determine the original automation unit.

[0058] S420. Conduct a trial run of the original automated unit in a preset isolated environment and determine the trial run results.

[0059] S430. The trial operation results are standardized to obtain standardized results.

[0060] S440. In response to the normalization result passing the result verification, the original automation unit is solidified into the target automation unit.

[0061] S450. In response to acquiring the unit output of the target automation unit, the unit output is materialized to obtain a target data object.

[0062] It should be noted that after the target automation unit performs the corresponding task, it will produce the corresponding output, which is recorded as the unit output. However, the unit output may be difficult to save in a certain data format for subsequent use. Therefore, this implementation intends to further materialize the results after obtaining the unit output.

[0063] Among them, result materialization means separating the unit output of the target automation unit from the operation process of the target automation unit and turning it into an addressable, verifiable, refreshable, and auditable target data object. The target data object can be saved as a file, database record, object storage object, cache key, event snapshot, etc.

[0064] Example 5 This application provides a method for the trusted solidification of automated units in Embodiment 5, which supplements the method described in Embodiment 1 or 2. It should be noted that for parts not detailed in this embodiment, please refer to the descriptions in other embodiments. Taking the supplement to Embodiment 1 as an example, the method includes: S510. Based on the acquired automation target and the preset large language model system, determine the original automation unit rules and the original automation units corresponding to the original automation unit rules.

[0065] S520. Conduct a trial run of the original automated unit in a preset isolated environment and determine the trial run results.

[0066] S530. The trial operation results are standardized and converted to obtain standardized results.

[0067] S540. In response to the normalization result passing the result verification, the original automation unit is solidified into the target automation unit.

[0068] S550. Obtain the execution status and execution version corresponding to the target automation unit.

[0069] Once the target automation unit is generated, its execution status can be determined based on whether it is currently executing. This execution status includes: currently executing and currently not executing. In addition, after the current version of the target automation unit is generated, in order to meet the new needs of users, the current version of the target automation unit can be updated according to the steps S510-S540 above. Each version of the target automation unit is matched with a corresponding execution version.

[0070] It should be noted that multiple target automation units may be running simultaneously at any given time. The execution status helps users understand the current actual execution status of the corresponding target automation unit, and the execution version helps users understand whether the target automation unit currently being used is the latest version.

[0071] Example 6 This application provides a method for trusted solidification of automated units in Embodiment Six. This method supplements the method shown in Embodiment One or Two. It should be noted that for parts not described in detail in this embodiment, please refer to the descriptions in other embodiments. Taking the supplement to Embodiment One as an example, the method includes: S610. Based on the acquired automation target and the preset large language model system, determine the original automation unit rules and the original automation units corresponding to the original automation unit rules.

[0072] S620. Conduct a trial run of the original automated unit in a preset isolated environment and determine the trial run results.

[0073] S630. The trial operation results are standardized to obtain standardized results.

[0074] S640. In response to the normalization result passing the result verification, the original automation unit is solidified into the target automation unit.

[0075] S650. Save the target automation unit locally.

[0076] It should be noted that large language model systems generally execute the functions of target automation units by retrieving local files. Therefore, after generating the target automation unit, it is also necessary to store the target automation unit in the ontology file so that the large language model system can retrieve and use it later.

[0077] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0078] Example 7 Based on the same inventive concept, this embodiment also provides an automated unit trusted curing apparatus for implementing the automated unit trusted curing method described above. The solution provided by this apparatus is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the automated unit trusted curing apparatus provided below can be found in the limitations of the automated unit trusted curing method described above, and will not be repeated here.

[0079] In this embodiment, as Figure 2 As shown, an automated unit reliable curing device is provided, comprising: The original unit determination module is used to determine the original automation unit rules and the original automation units corresponding to the original automation unit rules based on the acquired automation target and the preset large language model system. The trial run result determination module is used to conduct a trial run of the original automated unit in a preset isolated environment and determine the trial run result; The standardized result generation module is used to perform standardized transformation on the trial operation results to obtain standardized results; The target unit solidification module is used to solidify the original automation unit into a target automation unit in response to the normalization result passing the result verification.

[0080] Each module in the aforementioned automated unit's trusted solidification device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of the electronic device in hardware form or independent of it, or stored in the memory of the electronic device in software form, so that the processor can call and execute the operations corresponding to each module.

[0081] It should be noted that this embodiment determines the original automation unit rules and the original automation units corresponding to the original automation unit rules based on the acquired automation target and the preset large language model system; it performs a trial run on the original automation units in a preset isolation environment to determine the trial run results; it performs a normalization transformation on the trial run results to obtain normalized results; and in response to the normalization results passing the result verification, it solidifies the original automation units into target automation units. Through the above implementation, after generating the original automated unit rules and corresponding original automated units through the large language model system, the original automated unit rules can constrain the trial operation process of the original automated units, so that the input, output, execution and other processes of the original automated units meet the corresponding requirements. Then, the trial operation results of the original automated units are converted into normalized results that can be verified. Finally, the normalized results are verified to determine whether the various trial operation indicators of the original automated units meet the requirements. When the verification is passed, that is, when the various trial operation indicators meet the requirements, the original automated unit at this time is determined to be a long-term stable automated unit, that is, the target automated unit. By generating first and then verifying, it is ensured that the final determined target automated unit is a long-term stable automated unit. This makes it easy to realize that the service functions generated by the large language model can be used stably for a long time in the form of automated units.

[0082] In an optional embodiment, determining the original automation unit rules and the original automation units corresponding to the original automation unit rules based on the acquired automation target and the preset large language model system includes: Based on the pre-set large language model system, the obtained automation targets are processed to determine the original automation unit rules; In response to the original automation unit rule passing rule verification, the original automation unit rule is solidified into the target automation unit rule; The target automation unit rules are processed based on the large language model system to determine the original automation unit.

[0083] In an optional embodiment, the automated unit reliable curing device further includes: The capability surface determination module is used to determine the capability surface of the target automation unit for each exposed target.

[0084] In an optional embodiment, the automated unit reliable curing device further includes: The result materialization module is used to materialize the unit output of the target automation unit in response to obtaining the unit output, thereby obtaining the target data object.

[0085] In an optional embodiment, the automated unit reliable curing device further includes: The unit information acquisition module is used to acquire the execution status and execution version corresponding to the target automation unit.

[0086] In an optional embodiment, the automated unit reliable curing device further includes: A local storage module is used to save the target automation unit locally.

[0087] Example 8 Figure 3 A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0088] like Figure 3 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0089] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0090] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as the trusted solidification method for automated units.

[0091] In some embodiments, the trusted hardening method for automated units can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the trusted hardening method for automated units described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the trusted hardening method for automated units by any other suitable means (e.g., by means of firmware).

[0092] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0093] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0094] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0095] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0096] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0097] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0098] Example 9 In this embodiment, a computer-readable storage medium is provided, such as... Figure 4 As shown, a computer program is stored thereon, and when the computer program is executed by the processor, it implements the steps in the above-described method embodiments.

[0099] Example 10 In this embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0100] It should be noted that the information collected is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data all comply with the relevant laws, regulations and standards of the relevant countries and regions, necessary confidentiality measures have been taken, and it does not violate public order and good morals. Corresponding operation portals are provided for users to choose to authorize or refuse.

[0101] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this disclosure can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this disclosure may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this disclosure may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0102] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0103] The embodiments described above are merely illustrative of several implementations of this disclosure, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent disclosure. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this disclosure, and these all fall within the protection scope of this disclosure. Therefore, the protection scope of this disclosure should be determined by the appended claims.

Claims

1. An automated unit reliable solidification method, characterized in that, include: Based on the acquired automation goals and the preset large language model system, the original automation unit rules and the original automation units corresponding to the original automation unit rules are determined. The original automated unit was tested in a pre-set isolated environment to determine the test results; The trial operation results are then normalized to obtain normalized results. In response to the normalization result passing the result verification, the original automation unit is solidified into the target automation unit.

2. The method according to claim 1, characterized in that, The system, based on the acquired automation target and the preset large language model, determines the original automation unit rules and the original automation units corresponding to the original automation unit rules, including: Based on the pre-set large language model system, the obtained automation targets are processed to determine the original automation unit rules; In response to the original automation unit rule passing rule verification, the original automation unit rule is solidified into the target automation unit rule; The target automation unit rules are processed based on the large language model system to determine the original automation unit.

3. The method according to claim 2, characterized in that, Also includes: Determine the capability aspects of the target automation unit for each exposed target.

4. The method according to claim 2, characterized in that, Also includes: In response to acquiring the unit output of the target automation unit, the unit output is materialized to obtain a target data object.

5. The method according to any one of claims 1 or 2, characterized in that, Also includes: Obtain the execution status and execution version corresponding to the target automation unit.

6. The method according to any one of claims 1 or 2, characterized in that, Also includes: Save the target automation unit locally.

7. An automated unit-based reliable curing device, characterized in that, The device includes: The original unit determination module is used to determine the original automation unit rules and the original automation units corresponding to the original automation unit rules based on the acquired automation target and the preset large language model system. The trial run result determination module is used to conduct a trial run of the original automated unit in a preset isolated environment and determine the trial run result; The standardized result generation module is used to perform standardized transformation on the trial operation results to obtain standardized results; The target unit solidification module is used to solidify the original automation unit into a target automation unit in response to the normalization result passing the result verification.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the automated unit trusted solidification method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that are used to cause a processor to execute the trusted solidification method for automated units as described in any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-6.