Prompt management method and apparatus, and device, storage medium and program product

Abstract

Provided in the embodiments of the present disclosure are a prompt management method and apparatus, and a device, a storage medium and a program product. The method comprises: in response to a resource creation request for a prompt, providing a creation page for creating the prompt; receiving via a creation interface a created target prompt and identification information of the target prompt; and in response to an acknowledgement indication, adding the received target prompt and identification information to a prompt library, wherein the prompt library comprises at least one prompt, each prompt can be selected to be associated with a function based on a machine learning model, and the associated prompt is provided as an input of the machine learning model.

Description

Methods, devices, equipment, storage media, and program products for prompt word management

[0001] This application claims priority to Chinese Patent Application No. 202411605436.3, filed on November 11, 2024, entitled "Method, Apparatus, Device, Storage Medium and Program Product for Prompt Word Management", the entire contents of which are incorporated herein by reference. Technical Field

[0002] The exemplary embodiments disclosed herein generally relate to the field of computers, and particularly to methods, apparatus, devices, computer-readable storage media, and computer program products for prompt word management. Background Technology

[0003] With the rapid development of computer technology, the application of machine learning models is becoming increasingly widespread. Among these, prompt words are a key tool for interacting with machine learning models. Accurate prompt words are not only crucial for optimizing task generation but also fundamentally determine the quality of machine learning model output. For example, digital assistants based on machine learning models can understand and respond to user interactions in natural language. Therefore, high-quality prompt words can significantly enhance the depth of machine learning models' understanding of tasks, thereby driving the generation of high-quality content. Summary of the Invention

[0004] In a first aspect of this disclosure, a method for managing prompt words is provided. The method includes: providing a creation page for prompt word creation in response to a prompt word resource creation request; receiving a target prompt word and its identifier information via the creation interface; and adding the received target prompt word and identifier information to a prompt word library in response to a confirmation instruction. The prompt word library includes at least one prompt word, each prompt word being selectable to be associated with a function based on a machine learning model, and the associated prompt word being provided as input to the machine learning model.

[0005] In a second aspect of this disclosure, an apparatus for managing prompt words is provided. The apparatus includes: a creation interface providing module configured to provide a creation page for prompt word creation in response to a prompt word resource creation request; a receiving module configured to receive a target prompt word and its identification information via the creation interface; and an adding module configured to add the received target prompt word and identification information to a prompt word library in response to a confirmation instruction. The prompt word library includes at least one prompt word, each prompt word being selectable to be associated with a function based on a machine learning model, and the associated prompt word being provided as input to the machine learning model.

[0006] In a third aspect of this disclosure, an electronic device is provided. The device includes at least one processor; and at least one memory coupled to the at least one processor and storing instructions for execution by the at least one processor. When executed by the at least one processor, the instructions cause the electronic device to perform the method of the first aspect.

[0007] In a fourth aspect of this disclosure, a computer-readable storage medium is provided. The computer-readable storage medium stores computer-executable instructions that can be executed by a processor to implement the method of the first aspect.

[0008] In a fifth aspect of this disclosure, a computer program product is provided. The computer program product includes computer-executable instructions that, when executed by a processor, implement the method of the first aspect.

[0009] It should be understood that the description in this section is not intended to limit the key or essential features of the embodiments of this disclosure, nor is it intended to restrict the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0010] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. In the drawings, the same or similar reference numerals denote the same or similar elements, wherein:

[0011] Figure 1 shows a schematic diagram of an example environment in which embodiments of the present disclosure can be implemented;

[0012] Figures 2A to 2G illustrate schematic diagrams of example interfaces for prompt word management according to some embodiments of the present disclosure;

[0013] Figures 3A to 6 show schematic diagrams of example interfaces for creating new prompts or editing specific functions according to some embodiments of the present disclosure;

[0014] Figures 7A to 7E show schematic diagrams of example interfaces for commenting on prompt words according to some embodiments of the present disclosure;

[0015] Figure 8 shows a flowchart of a process for prompt word management according to some embodiments of the present disclosure;

[0016] Figure 9 shows a schematic structural block diagram of a device for prompt word management according to some embodiments of the present disclosure; and

[0017] Figure 10 shows a block diagram of an electronic device in which one or more embodiments of the present disclosure may be implemented. Detailed Implementation

[0018] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0019] In the description of embodiments of this disclosure, the term "comprising" and similar terms should be understood as open-ended inclusion, i.e., "including but not limited to". The term "based on" should be understood as "at least partially based on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The term "some embodiments" should be understood as "at least some embodiments". Other explicit and implicit definitions may also be included below.

[0020] In this document, unless explicitly stated otherwise, performing a step in response to A does not mean that the step is performed immediately after A, but may include one or more intermediate steps.

[0021] It is understood that the data involved in this technical solution (including but not limited to the data itself, the acquisition, use, storage or deletion of the data) shall comply with the requirements of relevant laws, regulations and related provisions.

[0022] It is understood that before using the technical solutions disclosed in the various embodiments of this disclosure, relevant users should be informed of the type, scope of use, and usage scenarios of the information involved in this disclosure through appropriate means in accordance with relevant laws and regulations, and authorization should be obtained from the relevant users. Among them, relevant users may include any type of rights holder, such as individuals, enterprises, and groups.

[0023] For example, in response to receiving an active request from a user, a prompt message is sent to the relevant user to clearly inform the user that the requested operation will require obtaining and using the user's information, thereby enabling the relevant user to choose whether to provide information to the software or hardware such as the electronic device, application, server, or storage medium that performs the operation of the technical solution disclosed herein based on the prompt message.

[0024] As an optional but non-restrictive implementation, in response to a user's active request, a prompt message can be sent to the user, such as a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide information to the electronic device.

[0025] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of this disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of this disclosure.

[0026] As used in this paper, the term "model" refers to a model that learns the relationship between inputs and outputs from training data, enabling it to generate corresponding outputs for a given input after training. Model generation can be based on machine learning techniques. Deep learning is a machine learning algorithm that processes inputs and provides corresponding outputs using multiple layers of processing units. A neural network model is an example of a deep learning-based model. In this paper, "model" may also be referred to as a "machine learning model," "learning model," "machine learning network," or "learning network," and these terms are used interchangeably.

[0027] As briefly described above, digital assistants can understand and respond to user interactions in natural language based on machine learning models. Digital assistants can serve as effective tools for people's work, study, and daily life. Typically, the development of digital assistants is similar to that of general applications, requiring developers with programming skills to define the various capabilities of the digital assistant by writing complex code and deploying it on an appropriate operating platform so that users can download, install, and use it.

[0028] With the diversification of application scenarios and the increasing availability of machine learning technology, there is a growing demand for developing more machine learning model-based functions with varying capabilities (e.g., digital assistants or workflow nodes) to support task processing in various sub-fields or to meet the personalized needs of different users. Users can create different prompts and provide them to the machine learning model to determine different digital assistants or workflow nodes. Conventionally, users often need to manually write different prompts to create digital assistants or workflow nodes with different functions. However, when users want to create digital assistants or workflow nodes with different functions, they need to rewrite the prompts.

[0029] According to embodiments of this disclosure, an improved scheme for prompt word management is provided. According to this scheme, if a prompt word resource creation request is received, a creation page for prompt word creation is provided. Accordingly, the target prompt word and its identification information are received via the creation interface. If a confirmation instruction is received, the received target prompt word and identification information are added to a prompt word library. The prompt word library includes at least one prompt word, each prompt word can be selected to be associated with a function based on a machine learning model, and the associated prompt word is provided as input to the machine learning model. In embodiments of this disclosure, prompt word management may include, but is not limited to, creating prompt words as resources, editing prompt word resources, and applying prompt word resources, etc.

[0030] Therefore, by storing the prompt words in a prompt word library, this embodiment of the disclosure allows users to conveniently and quickly reuse or continue to modify the prompt words. This helps to quickly create digital assistants or workflow nodes based on machine learning models with different functions.

[0031] Figure 1 illustrates a schematic diagram of an example environment 100 in which embodiments of the present disclosure can be implemented. Environment 100 relates to an assistant creation platform 110 and an assistant application platform 130.

[0032] As shown in Figure 1, the assistant creation platform 110 can provide user 105 with an environment for creating and publishing digital assistants or workflow nodes. In some embodiments, the assistant creation platform 110 can be a low-code platform that provides a collection of tools for creating digital assistants or workflow nodes. The assistant creation platform 110 can support visual development of digital assistants or workflow nodes, allowing developers to skip the manual coding process and accelerate application development cycles and reduce costs. The assistant creation platform 110 can support any suitable platform for users to develop digital assistants and other types of applications, such as an Application Platform as a Service (aPaaS) based platform. Such a platform enables users to efficiently develop applications, enabling operations such as application creation and application function adjustments.

[0033] The assistant creation platform 110 can be deployed locally on the user 105's terminal device and / or supported by a remote server. For example, the user 105's terminal device can run a client of the assistant creation platform 110, which can support interaction between the user and the assistant creation platform 110. When the assistant creation platform 110 runs locally on the user's terminal device, the user 105 can directly interact with the local assistant creation platform 110 using the client. When the assistant creation platform 110 runs on a server device, the server device can provide services to the client running on the terminal device based on the communication connection with the terminal device. The assistant creation platform 110 can present a corresponding interface 122 to the user 105 based on the user 105's operations, to output and / or receive information from the user 105.

[0034] In some embodiments, the assistant creation platform 110 can be associated with a corresponding database, which stores the data or information required for the creation process of a digital assistant based on a machine learning model supported by the assistant creation platform 110. For example, the database can store the code and description information corresponding to the various functional modules that make up the digital assistant. The assistant creation platform 110 can also perform operations such as calling, adding, deleting, and updating the functional modules in the database. The database can also store operations that can be performed on different functional blocks. For example, in a scenario where a digital assistant needs to be created, the assistant creation platform 110 can call the corresponding functional blocks from the database to build the digital assistant. These modules may include, but are not limited to, plugins for implementing specific functions, workflows (workflows may consist of a series of workflow nodes with sequential execution and dependencies), knowledge bases, etc.

[0035] In embodiments of this disclosure, user 105 can create and publish a digital assistant 120 on assistant creation platform 110 as needed. Digital assistant 120 can be published to any suitable assistant application platform 130, as long as the assistant application platform 130 supports the operation of digital assistant 120. After publication, digital assistant 120 can be used for conversational interaction with user 135. The client of assistant application platform 130 can present an interaction window 132 of digital assistant 120, such as a conversation window, in the client interface. Digital assistant 120, as an intelligent assistant, has intelligent dialogue and information processing capabilities. User 135 can input conversational messages in the conversation window, and digital assistant 120 can determine the reply message based on the created configuration information and present it to the user in interaction window 132. In some embodiments, depending on the configuration of digital assistant 120, interaction messages with digital assistant 120 can include multimodal messages, such as text messages (e.g., natural language text), voice messages, image messages, video messages, etc.

[0036] The assistant creation platform 110 and / or assistant application platform 130 can run on suitable electronic devices. These electronic devices can be any type of computing-capable device, including terminal devices or server devices. Terminal devices can be any type of mobile terminal, fixed terminal, or portable terminal, including mobile phones, desktop computers, laptop computers, notebook computers, netbook computers, tablet computers, media computers, multimedia tablets, personal communication system (PCS) devices, personal navigation devices, personal digital assistants (PDAs), audio / video players, digital cameras / camcorders, positioning devices, television receivers, radio receivers, e-book devices, gaming devices, or any combination of the foregoing, including accessories and peripherals of these devices or any combination thereof. Server devices can include, for example, computing systems / servers, such as mainframes, edge computing nodes, computing devices in cloud environments, and so on. In some embodiments, the assistant creation platform 110 and / or assistant application platform 130 can be implemented based on cloud services.

[0037] It should be understood that the structure and function of environment 100 are described for illustrative purposes only and do not imply any limitation on the scope of this disclosure. For example, while Figure 1 illustrates a single user interacting with assistant creation platform 110 and a single user interacting with assistant application platform 130, in practice multiple users can access assistant creation platform 110 to each create a digital assistant, and each digital assistant can be used to interact with multiple users.

[0038] The following description will detail some exemplary embodiments of this disclosure with reference to the accompanying drawings. It should be understood that the pages / interfaces shown in the drawings are merely examples, and various page designs / interfaces may actually exist. The various graphic elements in the page / interface may have different arrangements and different visual representations, one or more elements may be omitted or replaced, and one or more other elements may also be present. The embodiments of this disclosure are not limited in this respect.

[0039] The prompt word management method described in the embodiments of this disclosure can be implemented on an assistant creation platform, a terminal device with the assistant creation platform installed, and / or a server corresponding to the assistant creation platform. In the examples below, for the sake of discussion, the description is from the perspective of the assistant creation platform, such as the assistant creation platform 110 shown in FIG1. ​​The interface presented by the assistant creation platform 110 can be presented via the terminal device of user 105, and user input can be received via the terminal device of user 105. In this document, the user 105 who creates the digital assistant is sometimes also referred to as the assistant creator, assistant developer, etc.

[0040] The following description, for ease of understanding, will refer to the accompanying drawings and will primarily illustrate the example of prompt words being associated with a machine learning model-based digital assistant. However, this is merely exemplary and is not intended to limit the scope of the disclosure. For example, each prompt word may also be associated with a machine learning model-based workflow or workflow node. The method for managing prompt words according to this disclosure will now be described with reference to Figures 2A to 2G. Figures 2A to 2G illustrate schematic diagrams of example interfaces 200A to 200G for prompt word management according to some embodiments of this disclosure. It should be understood that users can create prompt words and save them in a prompt word library as resources. Users can also edit prompt word resources. Additionally, users can also apply prompt word resources created by themselves or other users.

[0041] In embodiments of this disclosure, if the assistant creation platform 110 receives a resource creation request for a prompt word, it provides a creation interface for prompt word creation. In some embodiments, the assistant creation platform 110 can provide the addition of prompt word resource types in a resource addition interface. Upon receiving a selection of a prompt word resource type, the assistant creation platform 110 can receive a resource creation request for the prompt word. As shown in the example interfaces 200A to 200B of Figures 2A to 2B, if the assistant creation platform 110 detects that user 105 clicks the "Prompt Word" control 212 in the panel corresponding to "Create Resource" 211, it presents the creation interface 220 for prompt word creation shown in Figure 2B. Thus, on the assistant creation platform of the digital assistant, prompt words can be regarded as resources similar to other resources (e.g., workflows, image streams, plugins, knowledge bases, message cards, etc.) and can be shared with other users.

[0042] In some embodiments, the assistant creation platform 110 can also receive resource creation requests for prompt words within the editing interface of a specific function being edited. Here, "function" refers to a function based on a machine learning model. In typical function editing scenarios, because the function is implemented based on a machine learning model, the creator also needs to input prompt words for the machine learning model, such as system prompts, during function editing. An example of such a function includes AI applications for user interaction, such as digital assistants, where the digital assistant's responses to the user are generated based on a machine learning model specified by the creator. In some examples, for a machine learning model-based digital assistant, in a scenario where the digital assistant interacts with user A, if input from user A is received, the prompt words to be input to the machine learning model associated with the digital assistant can be determined based on user A's input and the system prompt words. The machine learning model can then determine its output based on the input prompt words, which will be used to determine the response to user A.

[0043] Another example of machine learning model-based functionality is a workflow node. During workflow editing, one or more workflow nodes can be selected as machine learning model-based workflow nodes, configured to utilize the machine learning model to process the workflow node's input and obtain its output. It should be understood that other types of machine learning model-based functionality may exist besides digital assistants or workflow nodes. The embodiments of this disclosure also apply to those functionalities.

[0044] According to some embodiments of this disclosure, creators of machine learning-based functions can store prompts created during the function editing process as prompt resources in a prompt library, which can then be conveniently used by other users or when creating other functions. The process of writing prompts in an editing interface based on an editing digital assistant or workflow will be described in detail below with reference to Figures 2C to 2G.

[0045] After providing a creation interface for prompt word creation, the assistant creation platform 110 receives the target prompt word and its identification information via the creation interface. In some embodiments, the prompt word identification information is mainly for identifying the prompt word to be created, and may include at least one of the prompt word name and description information. The prompt word name is for concise identification of the prompt word. The prompt word description information can be used to describe the scenarios in which the prompt word can be used, such as whether it is suitable for role-playing digital assistants or efficiency tool digital assistants. The prompt word description information may also additionally or alternatively describe the functions that the prompt word can achieve. For example, the description information may indicate that a machine learning model can call certain plugins, workflows, databases, knowledge bases, etc. based on the prompt word. However, this is merely exemplary, and this disclosure does not limit it. As shown in the example interface 200F of Figure 2F, after the assistant creation platform 110 adds the prompt word 262 to the prompt word library, it can present the name 265 and the description information 266 of the prompt word 262.

[0046] Accordingly, the creation interface may include a first area for inputting the prompt word. The body text of the prompt word to be created can be entered through the first area. Furthermore, the creation interface may also include a second area for inputting the prompt word name, and / or a third area for inputting descriptive information.

[0047] As shown in the example interface 200B in Figure 2B, the creation interface 220 includes an area 221 for inputting the prompt name, an area 222 for inputting the description information, and an area 223 for inputting the prompt. It can be understood that the assistant creation platform 110 can receive the prompt name input by the user 105 in area 221, the description information of the prompt input in area 222, and the prompt input in area 223 via the creation interface 220.

[0048] In some embodiments, if the assistant creation platform 110 receives a confirmation instruction, it adds the received target prompt word and identification information to the prompt word library, which includes at least one prompt word. As shown in Figure 2B, if the assistant creation platform 110 detects that the user 105 clicks the "Confirm" control 224, it adds the prompt word entered by the user 105 in area 223, the name of the prompt word entered in area 221, and the description information for the prompt word entered in area 222 to the prompt word library. In some examples, the prompt word library may be presented by the assistant creation platform 110 on interface 122 for the user 105 to manage the prompt words in the prompt word library.

[0049] In some embodiments, each prompt word may be selected to be associated with a function based on a machine learning model, and the associated prompt word is provided as input to the machine learning model. In some embodiments, as mentioned above, the function based on the machine learning model may include, but is not limited to, a digital assistant, or a workflow node in a workflow. That is, the prompt word may be associated with a digital assistant 120 or a workflow node based on a machine learning model. In this way, the digital assistant or workflow node can use the machine learning model to process the input of the function and provide the output of the function based on the output of the machine learning model. The prompt word can be used to guide the machine learning model to process the input, thereby generating the desired output.

[0050] In some embodiments, prompts may include character settings, describing the role or responsibilities of the digital assistant, its response style, etc., thereby guiding the machine learning model to process the input according to the character settings. Prompts may also include functions and workflows, describing the functions and workflows of the digital assistant and specifying how the digital assistant should answer user questions in different scenarios. This guides the machine learning model to process the input according to the functions and workflows. Prompts may also include constraints and limitations, restricting the scope of the digital assistant's responses, such as defining what the digital assistant should and should not answer. Additionally, prompts may include specifying a response format for the digital assistant, so that the digital assistant responds to user input according to that format.

[0051] The machine learning model can run locally on the assistant creation platform 110 or on a remote server. In some embodiments, the machine learning model can be based on a language model (LM). A language model, by learning from a large corpus, is capable of question-and-answering. The machine learning model can also be based on other suitable models. Configuration of prompt words can be done using natural language. This allows users to easily constrain the model's output and configure diverse digital assistants. In some embodiments, the machine learning model can also be based on any suitable model architecture, including but not limited to Transformer models, convolutional neural networks (CNNs), recurrent neural networks (RNNs), deep neural networks (DNNs), and so on.

[0052] In some embodiments, alternatively or additionally, the prompts may also indicate at least one workflow to be executed by the digital assistant 120 to be created. Each workflow may correspond to various operations performed by the digital assistant 120 when performing a specific function. That is, the user 105 may be allowed to describe in natural language how the digital assistant 120 should perform a certain function. Understandably, functions based on machine learning models (e.g., workflow nodes in the digital assistant or workflow) can be enhanced with their inherent features. For example, the digital assistant or workflow node may have the ability to call plugins, knowledge bases, databases, triggers, workflows, etc. Accordingly, the use cases and descriptions of these functions can be explained in the prompts corresponding to the digital assistant or workflow node.

[0053] It should be understood that the above are merely some examples of prompt words, and the embodiments disclosed herein are not limited in this respect. Users are free to experiment with creating different prompt words to construct a digital assistant whose responses meet user expectations. For example, in the prompt words, user 105 may be allowed to input requirements for the language of the response to digital assistant 120, and constraints on the content of the response from digital assistant 120 (e.g., the number of characters for different types of responses, the type of response content, etc.).

[0054] The following description, referring to Figures 2C to 2G, details the process of transforming prompt words in the editing interface of a digital assistant or workflow being edited. For ease of understanding, embodiments of this disclosure will primarily refer to Figures 2A to 2G, and will be described using an example of a digital assistant where a prompt word is selected and associated with a machine learning model.

[0055] In some embodiments, if the assistant creation platform 110 receives a trigger to the prompt word library in the first editing interface of the first function, it presents at least one prompt word from the prompt word library. In some examples, if the assistant creation platform 110 receives a trigger to the prompt word library in the first editing interface corresponding to the digital assistant or workflow node being edited, it presents at least one prompt word from the prompt word library.

[0056] The following description, referring to Figures 2C to 2D and 2E, illustrates how the assistant creation platform 110 receives triggers on the prompt word library.

[0057] In some embodiments, the assistant creation platform 110 can receive triggers for the prompt dictionary based on the first prompt dictionary entry point in the first editing interface. As shown in the example interfaces 200C to 200D of Figures 2C to 2D, if the assistant creation platform 110 detects that the user 105 clicks the debug control 231 for debugging the created digital assistant, it presents example interface 200D, which includes the editing interface 241 corresponding to the digital assistant. In some examples, the assistant creation platform 110 can receive triggers for the prompt dictionary based on the entry point 242 of the prompt dictionary included in the editing interface 241.

[0058] As shown in the example interface 200E in Figure 2E, if the assistant creation platform 110 detects that user 105 clicks on a workflow node in the workflow corresponding to the machine learning model, it will display the details page 200E corresponding to that workflow node. The assistant creation platform 110 can receive triggers for the prompt word library based on the entry 250 of the prompt word library included in the details page 200E.

[0059] In some embodiments, if the assistant creation platform 110 receives a trigger on the prompt word library, it presents at least one prompt word from the prompt word library. In some embodiments, if a prompt word among the at least one prompt word is selected, the assistant creation platform 110 may present a preview of the selected prompt word. As shown in the example interfaces 200E to 200F of Figures 2E to 2F, if the assistant creation platform 110 detects that the user 105 clicks on the prompt word library entry 242, it presents the prompt word library in interface 260. Accordingly, the assistant creation platform 110 may also present prompt words included in the prompt word library, such as AA prompt word 262, BB prompt word, CC prompt word, etc., in interface 260. In some examples, the assistant creation platform 110 may also present the creator's identification information, creation time, etc., of at least one prompt word.

[0060] In some examples, the assistant creation platform 110 may present at least one prompt word in list form on the left side of the prompt word library presentation interface 260. However, this is merely exemplary and is not a limitation of this disclosure. Furthermore, if the assistant creation platform 110 detects that user 105 clicks on prompt word AA 262, it may present the content of prompt word AA 262 on the right side of the prompt word library presentation interface 260 for user 105 to preview.

[0061] In some embodiments, the assistant creation platform 110 may present at least one prompt if it detects a trigger on the prompt word library. Each prompt word is categorized into at least one of multiple types. In some examples, the assistant creation platform 110 may display at least one prompt word according to different types. The types of prompt words can be configured according to various criteria; for example, some prompt words may be divided into recommended prompt words and other prompt words. For example, the assistant creation platform 110 may present at least one prompt word recommended by the user, and may also present at least one prompt word created by the developer and / or the developer's team. In some examples, the assistant recommendation platform 110 may present at least one prompt word recommended by the user 105 based on information about the digital assistant / workflow node that the user 105 is currently creating (e.g., the name of the digital assistant / workflow node). In some embodiments, the types of prompt words may also be categorized based on the scenario in which the prompt word is used and / or the function it targets. For example, prompt words for the "character avatar" category may be applicable to digital assistants or workflow nodes that generate character avatars, and prompt words for efficiency tools may be applied to digital assistants or workflow nodes that are efficiency tools. In addition, different prompt word types can be categorized for the functions being created (e.g., plugins, workflows, databases, knowledge bases, etc.).

[0062] In some embodiments, the assistant creation platform 110 can display at least one prompt word included in the prompt word library according to different types. In some examples, the assistant creation platform 110 can categorize the prompt words in the prompt word library according to their source (e.g., from a recommendation or from the team to which user 105 belongs). In some examples, the assistant creation platform 110 can also divide the prompt words in the prompt word library into several categories based on their identification information. Such categorization and display methods facilitate users in quickly locating the desired prompt word.

[0063] In some embodiments, if the assistant creation platform 110 receives an application request for a first prompt among at least one prompt words, it can insert the first prompt word into the area corresponding to the prompt word editing in the editing interface. As shown in example interfaces 200F and 200D in Figures 2F and 2D, if the assistant creation platform 110 detects that the user 105 has selected prompt word AA 262 and clicked the "Use" control 264, it will insert prompt word AA 262 into the area 243 corresponding to the prompt word editing in the editing interface 241.

[0064] Subsequently, the assistant creation platform 110 creates a first function based at least on the first prompt word or the edited first prompt word. In some examples, the assistant creation platform 110 creates a digital assistant 120 or a workflow node based at least on the AA prompt word 262. In other examples, the assistant creation platform 110 may also create a digital assistant 120 or a workflow node based at least on the edited AA prompt word 262. Understandably, after selecting to insert the AA prompt word 262, the user 105 may also write the AA prompt word 262 based on the AA prompt word 262.

[0065] In some embodiments, the first prompt word or an edited first prompt word can be input into a first machine learning model associated with a first function, and the output of the first function is determined based on the output of the first machine learning model. It is understood that the AA prompt word 262 or an edited AA prompt word 262 can be associated with a function based on the first machine learning model (e.g., digital assistant 120 or workflow node). In this way, the digital assistant 120 or workflow node can use the first machine learning model to process the input of the function and provide the output of the function based on the output of the first machine learning model. For example, for a machine learning model-based digital assistant, the first machine learning model can determine the user's needs corresponding to the user input based on the AA prompt word 262 or an edited AA prompt word 262, and output an output that will be used to determine a response to the user.

[0066] In some embodiments, to allow the user 105 who creates the digital assistant 120 based on prompt words to conveniently test the operation of the created digital assistant 120 during the creation process, a debugging area for the digital assistant 120 can also be provided in the example interface 200D, as shown in the debugging area 244 of FIG2D. The debugging area 244 includes an input area 245 for receiving debugging requests for the digital assistant 120, and also includes a presentation area 246 for providing debugging results for the debugging requests (and providing the received debugging requests). The debugging area 244 can be configured as an interactive window, simulating the interactive interface seen by the user of the digital assistant 120.

[0067] The following description, with reference to Figures 2E to 2G, continues the description of the prompt word management scheme of this disclosure during the editing process of a digital assistant or workflow node.

[0068] In some embodiments, the assistant creation platform 110 may also present a second editing interface for a second function, the second editing interface including at least the entered second prompt word. As shown in the example interfaces 200E to 200F of Figures 2E to 2F, if the current prompt word for the digital assistant or workflow node has been edited, and the assistant creation platform 110 detects that the user 105 clicks on the entry 242 of the prompt word library, then it presents the prompt word library interface 260.

[0069] Furthermore, if the assistant creation platform 110 receives a resource creation request for a prompt word via the second editing interface of the second function, it can provide a creation interface for prompt word creation, which includes at least an import control. As shown in the example interfaces 200F to 200G of Figures 2F to 2G, if the assistant creation platform 110 detects that the user 105 clicks on the "New Prompt Word" control 261 for resource creation included in the prompt word library interface 260, it presents a creation interface 270 for prompt word creation. The creation interface 270 includes an import control 271, an area 272 for inputting the prompt word name, an area 273 for inputting the prompt word's description information, etc.

[0070] Furthermore, if the assistant creation platform 110 detects a trigger on the import control, it can insert a second prompt word into the prompt word input area in the creation interface. As shown in Figure 2G, if the assistant creation platform 110 detects that user 105 clicks the import control 271, it can insert the currently edited prompt word into the prompt word input area 274 in the creation interface 270. In some examples, user 105 can also write the second prompt word in area 274. Then, if the assistant creation platform 110 detects that user 105 clicks the "Confirm" control, it adds the current prompt word and its name and description information to the prompt word library.

[0071] In some embodiments, permission to apply, edit, or delete at least one of the prompts in the prompt library is based on the user's role. It is understood that, depending on the user's role, prompts in the prompt library can be copied or directly used in the digital assistant or workflow node. Correspondingly, depending on the user's role, prompts in the prompt library can also be edited, deleted, etc. For example, the creator of the prompt (e.g., user 105) can edit, delete, etc., prompts in the prompt library. Other users belonging to the same organization / team / workspace as user 105 can access, apply, and copy prompts in the prompt library. In some embodiments, the creator of the prompt can configure which users have the ability to access, apply, and / or copy the prompt, or the ability of users to access, apply, and / or copy individual prompts can be configured based on a default policy.

[0072] In summary, according to the embodiments of this disclosure, by storing the prompt words in a prompt word library, users can conveniently and quickly reuse the prompt words. This enables the rapid creation of digital assistants or workflow nodes based on machine learning models with different functions.

[0073] In some embodiments, during the editing of a prompt, such as creating a new prompt from a prompt library or editing a specific function (e.g., a digital assistant or workflow node), an edit block feature for the prompt can be provided. If a prompt contains an edit block, the user can create new prompt content within the provided edit block when the prompt is used or edited again. During the editing of a digital assistant or workflow node, the user can edit the prompt content corresponding to that digital assistant or workflow node within the edit block. In some embodiments, where an edit block exists for a prompt resource, the user 105 can edit the content contained in that prompt. Alternatively, the user can edit the content contained in an edit block within the prompt. In some examples, the content contained in the edit block can be provided to the user (e.g., a developer) using the prompt resource in the form of a form to be filled out.

[0074] The editing block function will be described below with reference to Figures 3A to 6. Figures 3A to 6 show schematic diagrams of example interfaces 300 to 600 for creating new prompts or editing specific functions according to some embodiments of the present disclosure.

[0075] In some embodiments, to facilitate intuitive user identification of edit blocks, the assistant creation platform 110 can present edit blocks or their content, as well as non-edit block content in prompts, in different visual styles. Referring to Figures 3A to 3E, Figures 3A to 3E illustrate an example 300 of an editing interface for writing prompts for a prompt library according to some embodiments of the present disclosure. Example 300 includes a region 310 for editing prompt content, which can display the prompt content. It should be noted that this example description only assumes that the prompts (including edit blocks and non-edit blocks) consist only of text. In actual applications, prompts can include any appropriate type of content, such as images or code. Different content can correspond to different types of edit blocks. For example, text content can correspond to a text edit block, image content can correspond to an image edit block, API descriptions can correspond to an API edit block, and so on. The triggering methods, creation methods, etc., of different types of edit blocks can be the same or different.

[0076] Specifically, area 310 can display content with at least one editable block (e.g., editable blocks 311, 312, 313, 314, and 315) and non-editable blocks (e.g., the text "You will play a task role," the text "The following are detailed settings for this role; please construct your answer based on this information," the text "Basic task information," the text "Role background and context," etc., as shown in the figure). The assistant creation platform 110 can display the non-editable block content in a regular text visual style, and can display the editable blocks in visual styles such as bolding, italics, using different colors, different fonts, and / or adding borders. It should be understood that the visual styles of the editable blocks shown in the figure are only examples, and the visual styles to be used can be selected according to actual needs.

[0077] The assistant creation platform 110 can display an input cursor 320 and an add control 301 for text editing blocks in area 310. If no selection of at least part of the prompt word is received, the assistant creation platform 110 can determine that a trigger for the text editing block function has been detected if it receives a trigger for the add control 301 (e.g., clicking the add control 301). The assistant creation platform 110 can display the editing interface shown in Figure 3B.

[0078] In Figure 3B, the assistant creation platform 110 can present an inserted text editing block 330 at the input cursor 320. In some embodiments, the assistant creation platform 110 can also display fill guide text associated with the text editing block 330 (e.g., the text "Please enter the prompt text when the editing block content is empty" shown in the figure, displayed in the text editing block 330). The assistant creation platform 110 can also present a text input interface 340 for the text editing block 330 associated with the text editing block 330. The text input interface 340 includes at least one of an input area 342 and an input area 344.

[0079] In some embodiments, when a text input interface 340 is presented in association with the text editing block 330, it can be determined that the text editing block 330 is in a text editing state; when the text input interface 340 is not presented, it can be determined that the text editing block 330 is not in a text editing state. The assistant creation platform 110 can present the text editing block 330 in a text editing state and the text editing block 330 not in a text editing state with different visual styles. As an example, referring also to FIG3C, the assistant creation platform 110 can present the text editing block 330 with different visual styles in FIG3B and FIG3C.

[0080] Referring again to the editing interface shown in Figure 3D, in some embodiments, if the assistant creation platform 110 detects that the text "this role" in the prompt (not an editable block) is selected, it can directly present a window 360 including at least one operation control. The at least one control in window 360 includes at least an editing control 362. In response to the triggering of the editing control 362, the assistant creation platform 110 can configure a text editing block 350 based on the text "this role," with the text "this role" filled in the text editing block 350.

[0081] In some embodiments, the assistant creation platform 110 may also present a text input interface 370 associated with the text editing block 350. The text input interface 370 includes at least one of an input area 372 and an input area 374. Similarly, if the assistant creation platform 110 receives fill guide text for a given text editing block in input area 372, the fill guide text may be displayed if no fill text is present in the given text editing block. If the assistant creation platform 110 receives specified text in input area 374, the specified text may replace the already filled text (e.g., "this role") in the given text editing block with the specified text.

[0082] Similarly, if no specified text is received from the user via the input area 374, the assistant creation platform 110 may present the guiding text in the input area 372 in the text editing block 350. If the specified text is received from the user via the input area 374, the assistant creation platform 110 will prioritize presenting the specified text in the text editing block 350.

[0083] Similarly, when the text input interface 370 is presented in association with the text editing block 350, it can be determined that the text editing block 350 is in a text editing state; when the text input interface 370 is not presented, it can be determined that the text editing block 350 is not in a text editing state. The assistant creation platform 110 can present the text editing block 350 in a text editing state and the text editing block 350 not in a text editing state with different visual styles. As an example, referring also to FIG3E, the assistant creation platform 110 can present the text editing block 350 with different visual styles in FIG3D and FIG3E.

[0084] The assistant creation platform 110 can, in response to confirmation of editing of the prompt word, determine the content filled in the edit block as part of the prompt word. For example, referring to Example 300 shown in Figures 3A to 3E, Example 300 may include a confirmation control for the prompt word. For example, if the assistant creation platform 110 receives a trigger for the confirmation control, it can determine the content filled in the edit block as part of the prompt word.

[0085] Referring again to Figures 4A and 4B, the assistant creation platform 110 can, if a triggering of a text editing block function is detected when at least a portion of the text of a prompt is selected, configure a text editing block based on that portion of the text at the location of the selected text. This text editing block can be filled with at least a portion of the text. Figures 4A and 4B illustrate an example 400 of an editing interface for a target function according to some embodiments of this disclosure. A prompt is presented in area 410 of example 400.

[0086] Referring to the editing interface shown in Figure 4A, in some embodiments, if the assistant creation platform 110 detects that the text "XXXXXXXX" (not an editable block) in the prompt word is selected, it can directly present a window 420 including at least one operation control. The at least one control in the window 420 includes at least an editing control 422. If the assistant creation platform 110 detects that the editing control 422 is triggered, it can configure a text editing block 412 according to the selected text, the text editing block 412 being filled with the selected text "XXXXXXXX".

[0087] In some embodiments, the assistant creation platform 110 may also present a text input interface 430 associated with the text editing block 412. The text input interface 430 includes at least one of an input area 432 (which may be referred to, for example, as a second input area corresponding to the text editing block 412) and an input area 434 (which may be referred to, for example, as a first input area corresponding to the text editing block 412). The selected text “XXXXXXXX” may be displayed by default in the input area 434.

[0088] Similarly, when the text input interface 430 is presented in association with the text editing block 412, it can be determined that the text editing block 412 is in a text editing state; when the text input interface 430 is not presented, it can be determined that the text editing block 412 is not in a text editing state. The assistant creation platform 110 can present the text editing block 412 in a text editing state and the text editing block 412 not in a text editing state with different visual styles. As an example, the assistant creation platform 110 can present the text editing block 412 with different visual styles in FIG4A and FIG4B.

[0089] Referring again to Figures 5A to 5D, the assistant creation platform 110 can insert an API edit block at the selected location of a prompt if a trigger on the application programming interface (API) edit block function is detected. In some examples, this API edit block can be populated with identification information for a predefined API, which may include the API's name, image identifier, brief description, or any other appropriate information. Figures 5A to 5D illustrate example 500 of an editing interface for a target function according to other embodiments of this disclosure.

[0090] Referring to Figure 5A, in Example 500, the assistant creation platform 110 may, for example, determine that a trigger for the API edit block function has been detected in response to receiving a predefined symbol 530 (e.g., curly braces {}) input by the user in the configuration area 510. In this case, the assistant creation platform 110 may determine that a trigger for the API edit block has been detected and insert the API edit block at the predefined symbol 530.

[0091] In some examples, the assistant creation platform 110 may display at least one API (e.g., plugin A) associated with the target function in area 520 of the interface 500. The assistant creation platform 110 may also display a window 540 at a predetermined symbol 530, which may display the identification information of plugin A and an add control for plugin A. Subsequently, in example 500 shown in Figure 5B, if the assistant creation platform 110 receives a trigger for the add control for plugin A in window 540, it may populate the identification information of plugin A into the API edit block 550.

[0092] Referring again to Figure 5C, the prompt includes the API editing block 560. Since the API (i.e., plugin 123) is not associated with the target function, the assistant creation platform 110 can overlay a strikethrough on the API editing block to present the identification information of plugin 123 in a visual style corresponding to the disabled state. The visual style corresponding to the disabled state could be, for example, grayed out, but this is merely exemplary and not limited thereto. Further, if the assistant creation platform 110 detects a trigger on the API editing block 560, it presents window 562, which displays the fillable plugin 123 and the add control for plugin 123. In example 500 shown in Figures 5C to 5D, if the assistant creation platform 110 receives a trigger on the add control for plugin 123, it can associate plugin 123 with the target function, and the assistant creation platform 110 can present plugin 123 in area 520. Plugin 123 associated with the target function is then switched to the enabled state. The assistant creation platform 110 can also display the identification information of plugin 123 in a visual style corresponding to the enabled state by removing the strikethrough.

[0093] Continuing with reference to Figure 6, the assistant creation platform 110 can present the annotation content in different visual styles if it determines that a portion of the input content in the editing interface is marked as annotation content. This annotation content will not be input into the machine learning model. Figure 6 shows an example 600 of an editing interface according to some embodiments of this disclosure. Example 600 shown in Figure 6 includes an area 610 for presenting prompts. If the assistant creation platform 110 determines that content 612 is identified as annotation content, it can present the annotation content in area 610 using visual styles such as italic text and gray font. In other examples, if the assistant creation platform 110 determines that content 614 is computer language code, it can present the symbol "%" in content 614 and the word "set" in the code in a bold visual style.

[0094] In some embodiments, for the editing interface of a prompt word-based function, a prompt word commenting function may also be provided for the prompt word area. For example, the prompt word commenting function may be provided in the editing interface for prompt words in the prompt word library, or in the editing interface for a specific function (e.g., a digital assistant or workflow node). This is because different users may develop and maintain the same function, and providing a commenting function helps these users share opinions on the prompt words, provide annotations, and help users better understand the function of the prompt words. Figures 7A to 7E illustrate an example interface 700 for the prompt word commenting function according to some embodiments of this disclosure. The commenting function here is sometimes also referred to as an annotation function. Figures 7A to 7E illustrate the provision of the prompt word commenting function in the prompt word input area for a digital assistant. It should be understood that the prompt word commenting function may be provided during the prompt word editing process for other functions (e.g., workflow nodes), or during the process of adding new prompt words to the prompt word library, or when viewing prompt words in the prompt word library.

[0095] In some embodiments, in response to a comment trigger on at least a portion of the prompt word, the assistant creation platform 110 presents a user interface for comment input. In some embodiments, in response to at least a portion of the prompt word being selected, a comment control is presented, and in response to a trigger on the comment control, a user interface for comment input is presented. The user interface for comment input includes input controls. The input controls for comment input may include input boxes, or may also include one or more other input controls that support voice input, image input, file import, etc. Comments on at least a portion of the selected content can be received via the input controls, such as input boxes.

[0096] As shown in Figure 7A, in the prompt input area of ​​the digital assistant, in response to the selection of a portion 712 in the prompt, a panel of operable controls can be presented, including at least a comment control 712. In response to detecting a trigger on the comment control 710, a user interface 720 for comment input is presented, including an input box 722. The user can enter a comment on the portion 712 in the input box 722. In response to detecting confirmation of the comment, such as a trigger on the "Submit" control 724 in Figure 7A (or confirmation triggered in other ways), the received comment can be associated with the selected portion 712.

[0097] In some embodiments, the comment control used to trigger the comment function can be presented in association with the unit content of the prompt word, for example, the comment control can be presented in association with each paragraph or each line of prompt word content. As shown in Figure 7B, the comment control 710 is presented at each paragraph of the prompt word. The comment control can be fixed in presentation, or it can be presented after a hover operation on that section is detected (e.g., the mouse hovers over that section). Similar to the example in Figure 7B, by triggering the comment control, a user interface for comment input can be presented for the user to enter comment content.

[0098] In some embodiments, the commenting functionality for prompts can be determined based on the user's role. For example, a digital assistant, workflow node, or the creator of prompts in a prompt library can add comments to prompts, and the scope of users who can comment on corresponding prompts can be configured.

[0099] In some embodiments, comments associated with at least a portion of the prompt words can be presented to the user. In some embodiments, comments associated with at least a portion of the prompt words can be fixedly displayed in a specific comment display area. In some embodiments, comments associated with at least a portion of the prompt words can be collapsed and expanded to be presented to the user upon detecting a trigger to view the comments. As shown in FIG7C, a comment panel 730 can be presented by clicking the comment control 710 or by hovering over the comment control 710, where comments on the associated prompt word content are presented. In some embodiments, if there are multiple comments on at least a portion of the prompt word content, the comments can also be presented in a collapsed state and can be presented to the user upon being triggered. As shown in FIG7D, multiple comments can be presented in the comment panel 730 after triggering the comment control 710. In some embodiments, the comment panel can also provide similar comment controls or other comment input triggering methods to trigger the presentation of comment input controls. As shown in FIG7C, a comment input box 722 can also be presented while other comments are being presented, so that the current user can enter comment content.

[0100] In some embodiments, when the comment is collapsed, a summary of the comment for a portion of the prompt can be provided. The summary can indicate the number of comments, the identifier of at least some of the users who posted the comment, a portion of the comment's content, etc. This allows users to know that a comment exists for a portion of the prompt without expanding the comment details, and also to obtain at least some information about the comment. As shown in Figure 7E, a comment viewing control 740 is provided for the portion of the prompt associated with the comment. Triggering the comment viewing control 740 displays a comment panel 730 showing comments related to the associated prompt content.

[0101] In some embodiments, the presentation of comments can also be determined based on the user's role. For example, a user who can access the prompt can be configured to access comments related to the prompt. In some embodiments, editing of existing comments in the prompt can also be supported, including modifying or deleting the comment content. The comment editing functionality can also be determined based on the user's role. For example, a user who has the ability to edit the prompt can also be configured to edit comments.

[0102] In some embodiments, during the prompt editing process, the annotations and comments associated with the prompts are not used to build the prompts input into the machine learning model. In some embodiments, when creating a digital assistant or workflow, adding annotations / comments can help other developers of the digital assistant or workflow quickly understand the logic of the prompt, thereby improving development efficiency. Furthermore, after the prompts with added annotations / comments are saved in the prompt library, other developers can also quickly understand the logic of the prompt when reusing it.

[0103] Figure 8 shows a flowchart of a process 800 for prompt word management according to some embodiments of the present disclosure. Process 800 can be implemented at the assistant creation platform 110. Process 800 is described below with reference to Figure 1.

[0104] In box 810, the assistant creation platform 110 responds to the resource creation request for the prompt word and provides a creation page for prompt word creation.

[0105] In box 820, the assistant creation platform 110 receives the created target prompt word and the target prompt word's identification information via the creation interface.

[0106] In box 830, the assistant creation platform 110 responds to the confirmation instruction by adding the received target prompt word and identification information to the prompt word library, which includes at least one prompt word. Each prompt word can be selected to be associated with a function based on a machine learning model, and the associated prompt word is provided as input to the machine learning model.

[0107] In some embodiments, process 800 further includes: in a first editing interface of a first function, in response to a triggering of a prompt word library, presenting at least one prompt word from the prompt word library; in response to an application request for a first prompt word among the at least one prompt word, inserting the first prompt word into an area in the editing interface corresponding to the prompt word editing; and creating a first function based at least on the first prompt word or the edited first prompt word, wherein the first prompt word or the edited first prompt word can be input into a first machine learning model associated with the first function, and the output of the first function is determined based on the output of the first machine learning model.

[0108] In some embodiments, presenting at least one prompt word includes: presenting at least one prompt word in response to a triggering of a prompt word library; and presenting a preview of the selected prompt word in response to a prompt word being selected from the at least one prompt word.

[0109] In some embodiments, presenting at least one cue word includes presenting at least one cue word by type in response to a cue word library, wherein each cue word is categorized into at least one of a plurality of types.

[0110] In some embodiments, triggering of the prompt word library is initiated via: triggering a first prompt word library entry in a first editing interface; or selecting the type of prompt word resource in a resource addition interface.

[0111] In some embodiments, providing a creation interface for prompt word creation includes: presenting a second editing interface for a second function, the second editing interface including at least the entered second prompt word; in response to receiving a resource creation request for the prompt word via the second editing interface for the second function, providing a creation interface for prompt word creation, the creation interface including at least an import control; and in response to detecting a triggering of the import control, inserting the second prompt word into the area of ​​the creation interface for prompt word input.

[0112] In some embodiments, permission to apply, edit, or delete at least one of the prompt words in the prompt word library is based on the user's role.

[0113] In some embodiments, the identification information includes at least one of a prompt name and a description, and the creation interface includes at least a first area for prompt input, and the creation interface also includes at least one of the following: a second area for inputting the prompt name and a third area for inputting the description.

[0114] In some embodiments, the functionality based on the machine learning model includes at least one of the following: a digital assistant, a workflow node in a workflow.

[0115] Embodiments of this disclosure also provide corresponding apparatus for implementing the methods or processes described above. Figure 8 shows a schematic structural block diagram of an apparatus 800 for prompt word management according to some embodiments of this disclosure. The apparatus 800 may be implemented in or included in an assistant creation platform 110, for example. The various modules / components in the apparatus 800 may be implemented by hardware, software, firmware, or any combination thereof.

[0116] As shown in the figure, device 900 includes a creation interface providing module 910, configured to provide a creation page for creating prompt words in response to a resource creation request for prompt words. Device 900 also includes a receiving module 920, configured to receive the created target prompt word and its identification information via the creation interface. Device 900 further includes an adding module 930, configured to add the received target prompt word and identification information to a prompt word library in response to a confirmation instruction. The prompt word library includes at least one prompt word, each prompt word can be selected to be associated with a function based on a machine learning model, and the associated prompt word is provided as input to the machine learning model.

[0117] In some embodiments, the device 900 further includes a function creation module configured to, in a first editing interface of a first function, in response to a triggering of a prompt word library, present at least one prompt word from the prompt word library; in response to an application request for a first prompt word among the at least one prompt words, insert the first prompt word into an area in the editing interface corresponding to the prompt word editing; and create a first function based at least on the first prompt word or the edited first prompt word, wherein the first prompt word or the edited first prompt word can be input into a first machine learning model associated with the first function, and the output of the first function is determined based on the output of the first machine learning model.

[0118] In some embodiments, the device 900 further includes a presentation module configured to present at least one prompt word in response to a triggering of a prompt word library; and to present a preview of the selected prompt word in response to a prompt word being selected from the at least one prompt word library.

[0119] In some embodiments, the presentation module is further configured to present at least one cue word by type in response to a triggering of the cue word library, wherein each cue word is categorized into at least one of a plurality of types.

[0120] In some embodiments, triggering of the prompt word library is initiated via: triggering a first prompt word library entry in a first editing interface; or selecting the type of prompt word resource in a resource addition interface.

[0121] In some embodiments, the creation interface providing module 910 is further configured to present a second editing interface with a second function, the second editing interface including at least an entered second prompt word; in response to receiving a resource creation request for the prompt word via the second editing interface with the second function, a creation interface for prompt word creation is provided, the creation interface including at least an import control; and in response to detecting a trigger on the import control, the second prompt word is inserted into the area of ​​the creation interface for prompt word input.

[0122] In some embodiments, permission to apply, edit, or delete at least one of the prompt words in the prompt word library is based on the user's role.

[0123] In some embodiments, the identification information includes at least one of a prompt name and a description, and the creation interface includes at least a first area for prompt input, and the creation interface also includes at least one of the following: a second area for inputting the prompt name and a third area for inputting the description.

[0124] In some embodiments, the functionality based on the machine learning model includes at least one of the following: a digital assistant, a workflow node in a workflow.

[0125] The units and / or modules included in device 900 can be implemented in various ways, including software, hardware, firmware, or any combination thereof. In some embodiments, one or more units and / or modules can be implemented using software and / or firmware, such as machine-executable instructions stored on a storage medium. In addition to or as an alternative to machine-executable instructions, some or all of the units and / or modules in device 900 can be implemented at least partially by one or more hardware logic components. By way of example and not limitation, exemplary types of hardware logic components that can be used include field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0126] Figure 10 illustrates a block diagram of an electronic device 1000 in which one or more embodiments of the present disclosure may be implemented. It should be understood that the electronic device 1000 shown in Figure 10 is merely exemplary and should not constitute any limitation on the functionality and scope of the embodiments described herein. The electronic device 1000 shown in Figure 10 may include or be implemented as the assistant creation platform 110 of Figure 1, or the device 900 of Figure 9.

[0127] As shown in Figure 10, the electronic device 1000 is in the form of a general-purpose electronic device. Components of the electronic device 1000 may include, but are not limited to, one or more processing units or processors 1010, memory 1020, storage devices 1030, one or more communication units 1040, one or more input devices 1050, and one or more output devices 1060. The processor 1010 may be a physical or virtual processor and is capable of performing various processes according to programs stored in the memory 1020. In a multiprocessor system, multiple processors execute computer-executable instructions in parallel to improve the parallel processing capability of the electronic device 1000.

[0128] Electronic device 1000 typically includes multiple computer storage media. Such media can be any accessible media that is accessible to electronic device 1000, including but not limited to volatile and non-volatile media, removable and non-removable media. Memory 1020 can be volatile memory (e.g., registers, cache, random access memory (RAM)), non-volatile memory (e.g., read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory), or some combination thereof. Storage device 1030 can be removable or non-removable media and may include machine-readable media, such as flash drives, disks, or any other media that can be used to store information and / or data and can be accessed within electronic device 1000.

[0129] Electronic device 1000 may further include additional removable / non-removable, volatile / non-volatile storage media. Although not shown in FIG. 10, disk drives for reading from or writing to removable, non-volatile disks (e.g., "floppy disks") and optical disk drives for reading from or writing to removable, non-volatile optical disks may be provided. In these cases, each drive may be connected to a bus (not shown) via one or more data media interfaces. Memory 1020 may include computer program product 1025 having one or more program modules configured to perform various methods or actions of various embodiments of this disclosure.

[0130] The communication unit 1040 enables communication with other electronic devices via a communication medium. Additionally, the functionality of the components of the electronic device 1000 can be implemented using a single computing cluster or multiple computing machines capable of communicating via communication connections. Therefore, the electronic device 1000 can operate in a networked environment using logical connections to one or more other servers, network personal computers (PCs), or another network node.

[0131] Input device 1050 can be one or more input devices, such as a mouse, keyboard, trackball, etc. Output device 1060 can be one or more output devices, such as a monitor, speaker, printer, etc. Electronic device 1000 can also communicate with one or more external devices (not shown) via communication unit 1040 as needed. These external devices include storage devices, display devices, etc., and can communicate with one or more devices that enable user interaction with electronic device 1000, or with any device that enables electronic device 1000 to communicate with one or more other electronic devices (e.g., network card, modem, etc.). Such communication can be performed via input / output (I / O) interface (not shown).

[0132] According to an exemplary implementation of this disclosure, a computer-readable storage medium is provided that stores computer-executable instructions thereon, wherein the computer-executable instructions are executed by a processor to implement the methods described above. According to an exemplary implementation of this disclosure, a computer program product is also provided, which is tangibly stored on a non-transitory computer-readable medium and includes computer-executable instructions, which are executed by a processor to implement the methods described above.

[0133] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatuses, devices, and computer program products implemented according to this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0134] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processing unit of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner. Thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0135] Computer-readable program instructions can be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions that execute on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0136] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction, which contains one or more executable instructions for implementing the specified logical function. In some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0137] Various implementations of this disclosure have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed implementations. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described implementations. The terminology used herein is chosen to best explain the principles, practical applications, or improvements to technology in the market, or to enable others skilled in the art to understand the various implementations disclosed herein.

Claims

1. A method for managing prompt words, comprising: In response to a resource creation request for a prompt word, a creation interface is provided for creating prompt words. The system receives the created target prompt word and its identification information via the creation interface. as well as In response to a confirmation instruction, the received target prompt word and the identification information are added to a prompt word library, which includes at least one prompt word. Each prompt word can be selected to be associated with a function based on a machine learning model, and the associated prompt word is provided as input to the machine learning model.

2. The method according to claim 1, further comprising: In the first editing interface of the first function, in response to the triggering of the prompt word library, at least one prompt word in the prompt word library is presented; In response to an application request for the first prompt word among the at least one prompt words, the first prompt word is inserted into the area of ​​the editing interface corresponding to the prompt word editing; as well as The first function is created at least based on the first prompt word or the edited first prompt word, the first prompt word or the edited first prompt word being input into a first machine learning model associated with the first function, and the output of the first function being determined based on the output of the first machine learning model.

3. The method of claim 2, wherein presenting the at least one prompt word comprises: In response to a trigger on the prompt word library, at least one prompt word is presented; as well as In response to the selection of one of the at least one prompt words, a preview of the selected prompt word is displayed.

4. The method of claim 2, wherein presenting the at least one prompt word comprises: In response to a trigger on the cue word library, the at least one cue word is presented by type, wherein each cue word is categorized into at least one of a plurality of types.

5. The method of claim 2, wherein triggering the prompt word library is initiated via: Triggering the first prompt dictionary entry in the first editing interface; Select the type of prompt word resource in the resource addition interface.

6. The method according to any one of claims 1 to 5, wherein providing a creation interface for prompt word creation includes: A second editing interface that presents a second function, the second editing interface including at least the second prompt word that has been entered; In response to receiving a resource creation request for a prompt word via a second editing interface of the second function, a creation interface for creating the prompt word is provided, the creation interface including at least an import control; as well as In response to the detection of a trigger on the import control, the second prompt word is inserted into the area for prompt word input in the creation interface.

7. The method according to any one of claims 1 to 6, wherein permission for at least one of the application, editing, or deletion of prompt words in the prompt word library is based on the user's role.

8. The method according to any one of claims 1 to 7, wherein the identification information includes at least one of a prompt name and descriptive information, and The creation interface includes at least a first area for inputting prompt words, and the creation interface also includes at least one of the following: a second area for inputting prompt word names and a third area for inputting descriptive information.

9. The method according to any one of claims 1 to 8, wherein the function based on the machine learning model includes at least one of the following: a digital assistant, a workflow node in the workflow.

10. An apparatus for managing prompt words, comprising: The creation interface module is configured to respond to resource creation requests for prompt words and provide a creation interface for prompt word creation. The receiving module is configured to receive the created target prompt word and the identification information of the target prompt word via the creation interface; as well as An add module is configured to add the received target prompt word and the identification information to a prompt word library in response to a confirmation instruction. The prompt word library includes at least one prompt word, and each prompt word can be selected to be associated with a function based on a machine learning model. The associated prompt word is provided as input to the machine learning model.

11. An electronic device, comprising: At least one processor; as well as At least one memory coupled to the at least one processor and storing instructions for execution by the at least one processor, the instructions causing the electronic device to perform the method according to any one of claims 1 to 9 when executed by the at least one processor.

12. A computer-readable storage medium having stored thereon computer-executable instructions that can be executed by a processor to implement the method according to any one of claims 1 to 9.

13. A computer program product comprising computer-executable instructions, wherein the computer-executable instructions, when executed by a processor, implement the method according to any one of claims 1 to 9.

Images