Method and device for displaying evaluation result of cue word template, equipment and storage medium

Abstract

The invention discloses a cue word template evaluation result display method and device, equipment and a storage medium, and relates to the technical field of AI. The method comprises the steps that when a cue word template of an AI model is evaluated, evaluation results of the cue word template are displayed in different areas of an evaluation result display interface according to different evaluation dimensions in the form of evaluation identification, in this way, the display effect of the evaluation result is effectively improved, the related personnel can rapidly and visually know the evaluation result, and the evaluation efficiency is improved. The man-machine interaction efficiency is improved, and then the evaluation efficiency and accuracy of the cue word template are improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence (AI) technology, and in particular to a method, device, equipment and storage medium for displaying evaluation results of a prompt word template. Background Art

[0002] A prompt is an input provided to an AI model (such as a large AI model for natural language processing) to guide and stimulate the AI ​​model to generate a corresponding response or content. A prompt template is a template text that includes variables and fixed text content. By inputting specific values ​​of variables, prompts can be dynamically generated. It should be understood that since prompts are crucial to obtaining accurate and high-quality AI model output results, developers need to carefully design the best prompt template to guide the AI ​​model to better complete the task.

[0003] In related technologies, a set of use cases of the AI ​​model is usually used to evaluate the prompt word templates designed by developers, and the corresponding evaluation results are displayed in the form of a text list on the terminal interface. Developers can view the evaluation results of each prompt word template one by one to determine the impact of different prompt word templates on the output results of the AI ​​model, and then design a better prompt word template.

[0004] However, the above method of using a text list to display the evaluation results has a poor display effect, and developers need to spend a lot of time viewing the text content, resulting in low evaluation efficiency and poor accuracy of the prompt word template. Summary of the invention

[0005] The embodiment of the present application provides a method, device, equipment and storage medium for displaying evaluation results of a prompt word template, which can effectively improve the display effect of the evaluation results and enhance the evaluation efficiency and accuracy of the prompt word template. The technical solution is as follows:

[0006] In a first aspect, a method for displaying evaluation results of a prompt word template is provided, the method comprising:

[0007] Acquire a use case set and at least one prompt word template, wherein the use case set includes a plurality of use cases, the prompt word template is used to generate a prompt word according to the use case, and the prompt word is used as an input of an artificial intelligence (AI) model to obtain an output result corresponding to the prompt word;

[0008] Based on a first use case in the use case set and at least one of the prompt word templates, displaying a display area corresponding to each of the prompt word templates in a first area of ​​an evaluation result display interface, and displaying a first evaluation identifier of the corresponding prompt word template for the first use case in each of the display areas, wherein the first evaluation identifier indicates a use effect of the prompt word template obtained based on the first use case;

[0009] The second area of ​​the evaluation result display interface displays a second evaluation identifier of each prompt word template for the use case set, each second evaluation identifier indicates a usage effect of the prompt word template obtained based on a corresponding use case in the use case set, and the first area is adjacent to the second area.

[0010] Through the above method, when the prompt word template of the AI ​​model is evaluated, the evaluation results of the prompt word template are displayed in different areas of the evaluation result display interface in the form of evaluation labels according to different evaluation dimensions, so that relevant personnel can intuitively know the local use effect of using different prompt word templates on the same use case and the global use effect of using different prompt word templates on the entire use case set. This method effectively improves the display effect of the evaluation results, enables relevant personnel to quickly and intuitively know the evaluation results, improves the efficiency of human-computer interaction, and thereby improves the evaluation efficiency and accuracy of the prompt word template.

[0011] In some embodiments, the method further comprises:

[0012] Filling the first use case into at least one of the prompt word templates to obtain at least one first prompt word;

[0013] Inputting at least one of the first prompt words into the AI ​​model to obtain an output result corresponding to at least one of the first prompt words;

[0014] Based on a difference between an output result corresponding to at least one of the first prompt words and a reference result, an evaluation result of at least one of the prompt word templates is determined, and the evaluation result is used to determine the first evaluation identifier.

[0015] In some embodiments, determining the evaluation result of at least one of the prompt word templates based on the difference between the output result corresponding to at least one of the first prompt words and the reference result includes at least one of the following:

[0016] Determining an evaluation result of at least one of the prompt word templates based on a matching degree between an output result corresponding to at least one of the first prompt words and a reference result;

[0017] Based on the similarity between the output result corresponding to at least one of the first prompt words and the reference result, an evaluation result of at least one of the prompt word templates is determined.

[0018] Through the above method, multiple methods of determining the evaluation results of the prompt word template are provided, which is convenient for developers to set according to actual needs.

[0019] In some embodiments, the first evaluation identifier includes at least one of a text identifier and a pattern identifier; and the second evaluation identifier includes at least one of a text identifier and a pattern identifier.

[0020] In some embodiments, the first evaluation identifier includes at least one of the following: a symbol for indicating the effect of use, text for indicating the effect of use, and a numerical value for indicating the effect of use; the second evaluation identifier includes at least one of the following: a symbol for indicating the effect of use, text for indicating the effect of use, and a numerical value for indicating the effect of use.

[0021] In some embodiments, different first evaluation identifiers indicate different usage effects of the prompt word template obtained based on the first use case; different second evaluation identifiers indicate different usage effects of the prompt word template obtained based on corresponding use cases in the use case set.

[0022] Through the above method, a variety of evaluation mark styles are provided, so that when these evaluation marks are displayed on the evaluation result display interface, relevant personnel can intuitively know the use effect of the prompt word template, thereby improving human-computer efficiency.

[0023] In some embodiments, the method further comprises:

[0024] In response to a drag operation performed on a target control, the areas of the first region and the second region displayed on the evaluation result display interface change as the drag operation occurs.

[0025] In some embodiments, in response to the drag operation performed on the target control, displaying that the areas of the first region and the second region change with the drag operation includes any of the following:

[0026] In response to a first drag operation performed on the target control, reducing the area of ​​the first region and increasing the area of ​​the second region, wherein the first drag operation refers to a drag operation in which the drag direction points to the first region;

[0027] In response to a second drag operation performed on the target control, the area of ​​the first region is increased and the area of ​​the second region is reduced, wherein the second drag operation refers to a drag operation in which a drag direction points to the second region.

[0028] Through the above method, a function of dynamically adjusting the area of ​​the region is provided, which can meet the personalized needs of relevant personnel to view the evaluation results. Moreover, in some scenarios, the display content can be dynamically updated according to the change of the area of ​​the region, thereby making full use of the interface space, simplifying the operation method of viewing the evaluation results, and improving the evaluation efficiency and user experience.

[0029] In some embodiments, displaying the first evaluation mark of the corresponding prompt word template for the first use case in each display area includes:

[0030] A first evaluation identifier of a corresponding prompt word template for the first use case and an output result corresponding to at least one first prompt word are displayed in each display area, where the first prompt word is generated based on the first use case and the prompt word template.

[0031] In some embodiments, the method further comprises:

[0032] A third evaluation identifier of at least one of the prompt word templates for the use case set is displayed in the second area, and each of the third evaluation identifiers indicates a comprehensive use effect of at least one of the prompt word templates obtained based on a corresponding use case in the use case set.

[0033] In some embodiments, the method further comprises:

[0034] In response to a trigger operation performed on a display area corresponding to a target prompt word template in the first area, displaying an output result and a reference result corresponding to a target prompt word, the target prompt word being generated based on the first use case and the target prompt word template;

[0035] In response to the evaluation identifier update operation performed on the target prompt word template, a first evaluation identifier of the target prompt word template for the first use case is updated.

[0036] In some embodiments, the evaluation identification update operation is any of the following:

[0037] An operation performed on an evaluation identifier adjustment track, the evaluation identifier adjustment track indicating a plurality of different evaluation identifiers;

[0038] An operation is performed on an evaluation identifier adjustment option, wherein the evaluation identifier adjustment option indicates a plurality of different evaluation identifiers.

[0039] Through the above method, a manual intervention function for the evaluation results is provided, so that relevant personnel can update the corresponding evaluation logo according to the evaluation results displayed on the interface.

[0040] In some embodiments, the method further comprises:

[0041] In response to the prompt word template comparison operation performed in the first area, displaying the evaluation result of the first prompt word template and the evaluation result of the second prompt word template;

[0042] In response to a trigger operation performed on a display area corresponding to the first prompt word template, or in response to a trigger operation performed on a display area corresponding to the second prompt word template, a first evaluation identifier of the first prompt word template and the second prompt word template for the first use case is updated.

[0043] Through the above method, an evaluation result comparison operation for prompt word templates is provided, that is, relevant personnel can compare any two prompt word templates horizontally to select a prompt word template that better meets the requirements, providing a reference for further optimizing the prompt word templates.

[0044] In a second aspect, an embodiment of the present application provides a device for displaying evaluation results of a prompt word template, the device comprising at least one functional module for executing a method for displaying evaluation results of a prompt word template as provided in the aforementioned first aspect or any possible implementation of the first aspect.

[0045] In a third aspect, an embodiment of the present application provides a computing device comprising a processor and a memory; the memory is used to store at least one piece of program code, and the at least one piece of program code is loaded by the processor and executed by a method for displaying evaluation results of a prompt word template as provided in the first aspect or any possible implementation of the first aspect.

[0046] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, the computer-readable storage medium is used to store at least one program code, the at least one program code is used to implement the evaluation result display method of the prompt word template provided by the first aspect or any possible implementation of the first aspect. The storage medium includes but is not limited to a volatile memory, such as a random access memory, a non-volatile memory, such as a flash memory, a hard disk drive (HDD), and a solid state drive (SSD).

[0047] In a fifth aspect, an embodiment of the present application provides a computer program product, which, when executed on a computing device, enables the computing device to implement the method for displaying the evaluation results of a prompt word template provided in the first aspect or any possible implementation of the first aspect. The computer program product may be a software installation package, and when it is necessary to implement the method for displaying the evaluation results of a prompt word template, the computer program product may be downloaded and executed on a computing device. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a schematic diagram of an implementation environment provided by an embodiment of the present application;

[0049] Figure 2 is a schematic diagram of the hardware structure of a computing device provided in an embodiment of the present application;

[0050] Figure 3 is a schematic diagram of the structure of a computing device cluster provided in an embodiment of the present application;

[0051] Figure 4 This is a schematic diagram of a connection method of a computing device cluster provided in an embodiment of the present application;

[0052] Figure 5 It is a flow chart of a method for displaying evaluation results of a prompt word template provided in an embodiment of the present application;

[0053] Figure 6 is a schematic diagram of an evaluation result display interface provided in an embodiment of the present application;

[0054] Figure 7 is a schematic diagram of another evaluation result display interface provided in an embodiment of the present application;

[0055] Figure 8 is a schematic diagram of another evaluation result display interface provided in an embodiment of the present application;

[0056] Fig. 9 is a schematic diagram of another evaluation result display interface provided in an embodiment of the present application;

[0057] Fig.10 is a schematic diagram of another evaluation result display interface provided in an embodiment of the present application;

[0058] Fig.11 is a schematic diagram of an evaluation identification update operation provided in an embodiment of the present application;

[0059] Fig.12 is a schematic diagram of a prompt word template comparison operation provided in an embodiment of the present application;

[0060] Fig.13 It is a structural schematic diagram of an evaluation result display device for a prompt word template provided in an embodiment of the present application. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solutions and advantages of this application clearer, the implementation methods of this application will be further described in detail below in conjunction with the accompanying drawings. It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.) and signals involved in this application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data need to comply with the relevant laws, regulations and standards of relevant countries and regions. For example, the prompt word templates involved in this application are all obtained with full authorization.

[0062] For ease of understanding, the relevant terms and concepts involved in this application are first introduced below.

[0063] Artificial intelligence (AI): The basic principle of AI is to combine massive data with super-powerful computing and processing capabilities and intelligent algorithms to build an AI model that solves specific problems, so that the AI ​​model can automatically summarize and learn potential patterns or features from the data, thereby achieving a way of thinking close to that of humans. AI models are also AI algorithms (or AI operators), which are a general term for mathematical algorithms built on the principles of AI. They are also the basis for using AI to solve specific problems, such as deep learning (DL) models.

[0064] Training and reasoning of AI models: Any AI model needs to be trained before solving specific technical problems. The training method of an AI model refers to the process of using a specified initial model to calculate the training data, and adjusting the parameters in the initial model using a certain method based on the calculation results, so that the model gradually learns certain rules and has specific functions. After training, an AI model with stable functions can be used for reasoning. The training of an AI model is the process of using a trained AI model to calculate the input data and obtain predicted reasoning results.

[0065] AI Basic Development Platform: It is a one-stop AI development platform for users. The AI ​​Basic Development Platform can provide various capabilities in the entire AI development process, including data preprocessing, model building and training, model management, model deployment, data optimization, and model optimization updates. Users can complete the development of AI models and the deployment and management of AI applications based on the AI ​​Basic Development Platform. The various capabilities in the AI ​​Basic Development Platform can be integrated for users to use throughout the AI ​​process, or they can provide users with independent functions separately.

[0066] AI services in the cloud: There are two mainstream types of AI services in the cloud, one is the platform-as-a-service (PaaS) type of AI basic development platform services, and the other is the software-as-a-service (SaaS) type of AI application cloud services. For the first type of AI basic development platform services, public cloud service providers provide customers with AI basic development platforms with the support of sufficient underlying resources and upper-level AI algorithm capabilities. The built-in AI development framework and various AI algorithms in the AI ​​basic development platform allow customers to quickly build and develop AI models or AI applications that meet personalized needs on the AI ​​basic development platform. For the second type of AI application cloud services, public cloud service providers provide ready-to-use, general AI application cloud services through cloud platforms, allowing customers to use AI capabilities in various application scenarios with zero barriers.

[0067] AI big model: It is a deep neural network composed of a large number of layers and parameters. Due to its powerful predictive ability, it is widely used in various fields such as natural language processing, computer vision, speech synthesis and speech recognition, autonomous driving, etc.

[0068] Prompt: An input provided to an AI model (such as a large AI model for natural language processing) to guide and stimulate the AI ​​model to generate corresponding responses or content. For example, a prompt is a complete paragraph (or sentence) of text content, which can be used as an input to a large language model (LLM) and other models, so that the model can output the corresponding reasoning results based on the prompt.

[0069] Prompt word template: a template text including variables and fixed text content. By inputting the specific values ​​of the variables, prompt words can be dynamically generated. For example, in actual application, by adjusting the specific values ​​of the variables in the prompt word template, multiple different prompt words can be dynamically generated. Schematically, taking the AI ​​model used to implement sentiment analysis as an example, the prompt word template is: [Judge which of the three labels "positive", "neutral" and "negative" this comment {variable} belongs to? ], where {variable} can be filled in according to actual needs to dynamically generate different prompt words. For example, if the variable is "cost-effective, can run games smoothly", the generated prompt word is: [Judge which of the three labels "positive", "neutral" and "negative" this comment {cost-effective, can run games smoothly} belongs to? ].

[0070] Use case: also known as use case or usage, is a description of how the system responds to external requests in software engineering or system engineering. In the embodiment of the present application, the use case indicates the question intention for the AI ​​model, and can be used as the variable value of the prompt word template to dynamically generate different prompt words, or in other words, different use cases are combined with the same prompt word template to generate multiple different prompt words. For example, taking the AI ​​model for sentiment analysis as an example, the use case is: "1. High cost performance, can run the game smoothly. 2. High order processing efficiency. 3. Fast logistics". If the prompt word template is: [Judge this comment {variable} belongs to which of the three labels "positive", "neutral" and "negative"? ], then combine the use case with the prompt word template, or fill the use case into the prompt word template, and get the prompt word [Judge this comment {1. High cost performance, can run the game smoothly. 2. High order processing efficiency. 3. Fast logistics} belongs to which of the three labels "positive", "neutral" and "negative"? ].

[0071] The following is an introduction to the application scenarios and implementation environment involved in this application.

[0072] The technical solution provided by the embodiment of the present application can be applied to the scenario of evaluating the prompt word template of the AI ​​model. Among them, evaluating the prompt word template of the AI ​​model refers to combining the use case of the AI ​​model with the designed prompt word template to generate prompt words, inputting the prompt words into the AI ​​model to obtain the corresponding output results, and then, by comparing the difference between the output results of the prompt words and the reference results, obtaining the evaluation results of the corresponding prompt word template. It should be understood that if the evaluation result of the prompt word template indicates that the difference between the output results of the prompt words and the reference results is smaller, it indicates that when the prompt word template is used to generate the prompt words, the output results of the AI ​​model are more accurate, that is, the use effect of the prompt word template is better. Therefore, the evaluation result of the prompt word template can indicate the use effect of the prompt word template, thereby providing guidance for relevant personnel to further optimize the prompt word template.

[0073] refer to Figure 1 , Figure 1 This is a schematic diagram of an implementation environment provided by the embodiment of the present application. Figure 1 As shown, the implementation environment includes a terminal 101 and a server 102, and the terminal 101 is directly or indirectly connected to the server 102 via a wireless network or a wired network.

[0074] Terminal 101 has a display function and can display the evaluation results of the prompt word template. Terminal 101 is at least one of a smart phone, a game console, a desktop computer, an augmented reality terminal, a tablet computer, an e-book reader and a laptop portable computer. Terminal 101 is installed and runs an application that supports displaying the evaluation results. The application can be a client, or a browser, etc., and the present application is not limited to this. Taking the application as a client as an example, a user (such as a developer) can input a use case set and a designed prompt word template through the client, triggering terminal 101 to evaluate the prompt word template, and display the corresponding evaluation results on the client interface, so that the user can further optimize the prompt word template according to the evaluation results displayed on the client interface. Schematically, terminal 101 is a terminal used by a user, and a user account is logged in in the application running on terminal 101, also referred to as an object.

[0075] Server 102 is used to provide background services for applications running on terminal 101. Server 102 is an independent physical server, or a server cluster or distributed file system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content distribution networks (CDNs), and big data and artificial intelligence platforms. Take server 102 as an example of a cloud server, a cloud server refers to a service based on hardware resources and software resources, providing computing, network and storage capabilities. Through the network "cloud", the huge data calculation process is processed and analyzed at the remote end and then returned to the user, with the characteristics of large-scale, distributed, virtualized, high availability, scalability, on-demand services and security. In the embodiment of the present application, the service provided by the cloud server is also referred to as the AI ​​service in the cloud field (see the foregoing content for details, no more details).

[0076] In an embodiment of the present application, terminal 101 can independently execute the evaluation result display method of the prompt word template involved in the following method embodiment, or can work in collaboration with server 102. For example, terminal 101 sends the acquired use case and prompt word template to server 102, and server 102 evaluates the prompt word template and feeds back the evaluation result to terminal 101, which displays the evaluation result. This application does not limit this.

[0077] The terminal 101 may generally refer to one of multiple terminals, or a set of multiple terminals; the server 102 may generally refer to one of multiple servers, or a set of multiple servers. The embodiment of the present application does not limit the number and type of each device in the implementation environment. In addition, the above implementation environment can also be understood as an AI basic development platform, based on which users can complete the development of AI models (including evaluation of prompt word templates) and the deployment and management of AI applications.

[0078] In some embodiments, the wireless network or wired network described above uses standard communication technology and / or protocol. The network includes, but is not limited to, any combination of a data center network, a storage area network (SAN), a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a dedicated network or a virtual private network. In some implementations, the data exchanged through the network is represented by technologies and / or formats including hypertext markup language (HTML), extensible markup language (XML), etc. In addition, conventional encryption technologies such as secure socketslayer (SSL), transport layer security (TLS), virtual private network (VPN), and Internet protocol security (IPsec) can be used to encrypt all or part of the links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above data communication technologies.

[0079] The hardware structure of the terminal and server in the above implementation environment is introduced below.

[0080] The present application embodiment provides a computing device that can be configured as the above-mentioned terminal or server. Figure 2 , Figure 2 Schematic diagram of the hardware structure of a computing device provided in an embodiment of the present application. Figure 2As shown, the computing device 200 includes a memory 201, a processor 202, a communication interface 203 and a bus 204. The memory 201, the processor 202 and the communication interface 203 are connected to each other through the bus 204.

[0081] The memory 201 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of an instruction or data structure and can be accessed by a computer, but is not limited thereto. Schematically, taking the computing device as the above-mentioned terminal 101 as an example, the memory 201 is used to store at least one section of program code, and when the program code stored in the memory 201 is executed by the processor 202, the processor 202 is used to execute the steps executed by the terminal in the following method embodiment.

[0082] The processor 202 may be a network processor (NP), a central processing unit (CPU), an application-specific integrated circuit (ASIC), or an integrated circuit for controlling the execution of the program of the present application. The processor 202 may be a single-CPU processor or a multi-CPU processor. The number of processors 202 may be one or more. The communication interface 203 uses a transceiver module such as a transceiver to implement communication between the computing device 200 and other devices or communication networks. For example, data may be acquired through the communication interface 203.

[0083] The memory 201 and the processor 202 may be separately provided or integrated together.

[0084] Bus 204 may include a pathway for transmitting information between various components of computing device 200 (eg, memory 201 , processor 202 , communication interface 203 ).

[0085] The present application also provides a computing device cluster, which includes multiple computing devices. The multiple computing devices include a server and a terminal, wherein the server can be a server, such as a central server, an edge server, or a local server in a local data center, and the terminal can be a desktop computer, a laptop computer, or a smart phone. Figure 3 , Figure 3 Schematic diagram of a computing device cluster provided in an embodiment of the present application. Figure 3 As shown, the computing device cluster includes multiple computing devices 200. The same instructions for executing the evaluation result display method of the prompt word template may be stored in the memory 201 of the multiple computing devices 200 in the computing device cluster. In some embodiments, the memory 201 of the multiple computing devices 200 in the computing device cluster may also respectively store partial instructions for executing the evaluation result display method of the prompt word template. In other words, the combination of multiple computing devices 200 is used to jointly execute the instructions of the evaluation result display method of the prompt word template.

[0086] In some embodiments, multiple computing devices 200 in a computing device cluster may be connected via a network, which may be a wide area network or a local area network. Figure 4 Schematic diagram of a connection method of a computing device cluster provided in an embodiment of the present application. Figure 4 As shown, two computing devices 200 are connected via a network. Specifically, the network is connected via the communication interface 203 in each computing device 200. It should be understood that Figure 4 The functionality of the computing device 200 shown in FIG. 2 may also be performed by multiple computing devices 200 .

[0087] The following is an introduction to the method for displaying the evaluation results of the prompt word template provided in this application.

[0088] Figure 5 1 is a flow chart of a method for displaying evaluation results of a prompt word template provided in an embodiment of the present application. Figure 5 As shown, the method is applied to a terminal in the above implementation environment as an example for introduction. The method includes the following steps 501 to 507.

[0089] 501. The terminal obtains a use case set and at least one prompt word template, where the use case set includes multiple use cases, and the prompt word template is used to generate prompt words according to the use cases, and the prompt words are used as inputs of an AI model to obtain output results corresponding to the prompt words.

[0090] In the embodiments of the present application, the AI ​​model refers to an AI model based on natural language processing, and the present application does not limit the functions provided by the AI ​​model. For example, the AI ​​model is used to provide functions such as sentiment analysis, machine translation, question answering, language generation, article creation, or text classification. Schematically, the terminal is installed and runs a client with a display function, and the terminal can display the corresponding content on the client interface in response to various operations performed on the client interface. For example, developers can perform various operations on the client interface to trigger the terminal to evaluate the prompt word template of the AI ​​model.

[0091] Schematically, the client running on the terminal provides a prompt word template evaluation function for the AI ​​model. For example, the terminal displays a prompt word template evaluation interface for the AI ​​model, and in response to a use case set and at least one prompt word template input on the prompt word template evaluation interface, obtains a corresponding use case set and at least one prompt word template. The prompt word template evaluation interface can provide multiple input methods such as keyboard input and voice input, which is not limited in this application.

[0092] In addition, in the embodiment of the present application, the prompt word template obtained by the terminal may be an initial prompt word template designed by the developer, or a new prompt word template obtained by arbitrarily combining multiple initial prompt word templates (the combination process may be manually combined by the developer, or automatically combined by the terminal, which is not limited). It should be understood that the purpose of evaluating the prompt word template in the present application is to select a prompt word template that better meets the requirements (i.e., to guide the AI ​​model to output a result that is more in line with the expected result), or to further optimize the designed prompt word template. Therefore, at least one prompt word template obtained by the terminal may be regarded as at least one candidate set. For any candidate set, the candidate set may be an initial prompt word template designed by the developer, or a new prompt word template obtained by combining multiple initial prompt word templates. Thus, after the terminal obtains the evaluation result of at least one prompt word template (i.e., at least one candidate set) in the future, it may display the evaluation result, so that the developer may select the prompt word template with the best use effect, or further optimize the designed prompt word template, thereby improving the evaluation efficiency.

[0093] The following uses the AI ​​model used to provide sentiment analysis function as an example to illustrate the use case set and prompt word template involved in this application.

[0094] The use case set includes multiple use cases (the explanation of use cases is detailed in the above terminology introduction, which will not be repeated here). It should be understood that for the same use case, if different prompt word templates are combined, the generated prompt words are different, and the final output results of the AI ​​model are often different. For example, the use case set obtained by the terminal includes use case 1 and use case 2:

[0095] Use case 1: "1. High cost-effectiveness, can run games smoothly. 2. High order processing efficiency. 3. Fast logistics";

[0096] Use case 2: "1. Screen freezes. 2. Order processing efficiency is low. 3. After-sales service attitude is average."

[0097] The prompt word template is used to generate prompt words according to the use case (for the explanation of prompt words and prompt word templates, please refer to the above terminology introduction, which will not be repeated here). It should be understood that prompt word templates are usually designed by developers, and this application does not limit the number of prompt word templates. For example, the terminal obtains 3 prompt word templates of the AI ​​model:

[0098] Prompt word template a: [Determine which of the three labels "positive", "neutral" and "negative" does this comment {variable} belong to? ];

[0099] Prompt word template b: [I will provide you with a comment information. Please determine the corresponding emotional label based on the content of the comment. Possible emotional labels are positive, neutral, and negative. Do not make up other labels. You can refer to the following examples: "Community, beautiful, large keyboard" is positive; "How to buy this book?" is neutral; "The room has a strong smell, poor service, and poor attitude" is negative. The comment is: {variable}. What is the emotional label of this comment? ];

[0100] Prompt word template c: [Please determine the emotional label of this comment based on the keywords of this comment {variable}. Emotional labels include positive, neutral, and negative. Do not make up other labels].

[0101] Furthermore, if the use case is combined with the prompt word template, the corresponding prompt words can be generated as shown below:

[0102] Based on use case 1 and prompt word template ac, the following prompt words 1-a to 1-c can be generated:

[0103] Prompt word 1-a: [Judge whether this review {1. High cost performance, can run games smoothly. 2. High order processing efficiency. 3. Fast logistics} belongs to the three labels of "positive", "neutral" and "negative"? ];

[0104] Prompt word 1-b: [I will provide you with a review information. Please judge the corresponding emotional label based on the content of the review. Possible emotional labels are positive, neutral, and negative. Do not make up other labels. You can refer to the following examples: "Community, beautiful, large keyboard" is positive; "How to buy this book?" is neutral; "The room has a strong smell, poor service, and poor attitude" is negative. The review is: {1. High cost performance, can run games smoothly. 2. High order processing efficiency. 3. Fast logistics}, what is the emotional label of this review? ];

[0105] Prompt word 1-c: [Please use the keywords of this review {1. High cost performance, can run games smoothly. 2. High order processing efficiency. 3. Fast logistics} to determine the emotional label of this review. Emotional labels include positive, neutral, and negative. Do not make up other labels].

[0106] Based on use case 2 and prompt word template ac, the following prompt words 2-a to 2-c can be generated:

[0107] Prompt word 2-a: [Judge whether this review {1. Screen freezes. 2. Order processing efficiency is low. 3. After-sales service attitude is average} belongs to the three labels of "positive", "neutral" and "negative"? ];

[0108] Prompt word 2-b: [I will provide you with a review information. Please judge the corresponding emotional label based on the content of the review. Possible emotional labels are positive, neutral, and negative. Do not make up other labels. You can refer to the following examples: "Community, beautiful, large keyboard" is positive; "How to buy this book?" is neutral; "The room has a strong smell, poor service, and poor attitude" is negative. The review is: {1. The screen is stuck. 2. The order processing efficiency is low. 3. The after-sales service attitude is average}. What is the emotional label of this review? ];

[0109] Prompt word 2-c: [Please use the keywords of this review {1. Screen freezes. 2. Order processing is inefficient. 3. After-sales service attitude is average} to determine the emotional label of this review. Emotional labels include positive, neutral, and negative. Do not make up other labels].

[0110] It should be understood that the above introduction to the use case set and prompt word template is only for illustration and does not constitute a limitation of the present application.

[0111] 502. The terminal obtains an evaluation result of at least one prompt word template based on the use case set and at least one prompt word template.

[0112] In an embodiment of the present application, for any use case in the use case set, the terminal combines the use case with each prompt word template to generate a corresponding prompt word, and inputs the generated prompt word into the AI ​​model to obtain a corresponding output result, and then obtains the evaluation result of each prompt word template by comparing the difference between the output result corresponding to the prompt word and the reference result.

[0113] Among them, the reference result corresponding to the prompt word can be understood as the standard result or expected result corresponding to the prompt word. For example, taking the example "prompt word 1-a" in the above step 501 as an example, the reference result corresponding to the prompt word is "positive". If "prompt word 1-a" is input into the AI ​​model, the output result obtained is also "positive". Correspondingly, the evaluation result of prompt word template a is "match", indicating that the use of prompt word template a to generate prompt words can guide the AI ​​model to output the expected result, that is, it indicates that the use effect of prompt word template a meets the requirements. It should be noted that the reference result corresponding to the prompt word can be obtained by the terminal in response to the input operation implemented on the prompt word template evaluation interface during the execution of step 501, or it can be obtained by the terminal in response to the input operation implemented on the prompt word template evaluation interface during the execution of this step 502, and it can also be obtained from the server, etc. This application does not limit the way and time for the terminal to obtain the reference result.

[0114] Taking any one use case in the use case set (hereinafter referred to as the first use case) as an example, the process of obtaining the evaluation result of at least one prompt word template by the terminal is introduced, including the following steps 1 to 3:

[0115] Step 1: Fill the first use case into at least one prompt word template to obtain at least one first prompt word.

[0116] The terminal fills the first use case into each prompt word template respectively, and obtains the first prompt word when each prompt word template is used in the first use case. In addition, this process can refer to the example of prompt words in the above step 501, which will not be repeated here.

[0117] Step 2: Input at least one first prompt word into the AI ​​model to obtain an output result corresponding to the at least one first prompt word.

[0118] Among them, the terminal inputs each first prompt word into the AI ​​model respectively, calculates through the built-in logic of the AI ​​model, and obtains the output result corresponding to each first prompt word.

[0119] Step 3: Determine an evaluation result of at least one prompt word template based on a difference between an output result corresponding to at least one first prompt word and a reference result.

[0120] Among them, the evaluation result of the prompt word template is used to determine the first evaluation identifier of the prompt word template for the first use case, and the first evaluation identifier indicates the use effect of the prompt word template obtained based on the first use case (the relevant content of the first evaluation identifier will be introduced in detail in the subsequent step 503, and will not be repeated here). For any first prompt word, the terminal determines the evaluation result of the prompt word template corresponding to the first prompt word based on the difference between the output result corresponding to the first prompt word and the reference result. Schematically, the difference between the output result corresponding to the first prompt word and the reference result can be reflected by the matching degree, similarity and other information between the two. Several implementation methods of step 3 are introduced below, including at least one of the following:

[0121] The first method is to determine an evaluation result of at least one prompt word template based on a matching degree between an output result corresponding to at least one first prompt word and a reference result.

[0122] Among them, for any first prompt word, the terminal extracts the keyword of the output result corresponding to the first prompt word, matches the keyword with the reference result corresponding to the first prompt word, and determines the evaluation result of the prompt word template corresponding to the first prompt word based on the matching result. Schematically, if the matching result indicates a successful match, the evaluation result of the prompt word template corresponding to the first prompt word is determined to be "matched", and if the matching result indicates a failed match, the evaluation result of the prompt word template corresponding to the first prompt word is determined to be "unmatched". For example, continuing to take the example "prompt word 1-a" in step 501 as an example, the reference result corresponding to the prompt word is "positive", and the "prompt word 1-a" is input into the AI ​​model. If the output result obtained is also "positive", the evaluation result of the prompt word template a is "matched"; if the output result obtained is "neutral", the evaluation result of the prompt word template is "unmatched". It should be noted that the above method of determining whether the output result matches the reference result by using keywords is only for example, and the present application is not limited thereto. For example, the terminal directly compares the output result with the reference result. If the two are the same, the evaluation result is "matched", and so on.

[0123] The second method is to determine the evaluation result of at least one prompt word template based on the similarity between the output result corresponding to at least one first prompt word and the reference result.

[0124] Among them, for any first prompt word, the terminal determines the evaluation result of the prompt word template corresponding to the first prompt word based on the similarity between the output result corresponding to the first prompt word and the reference result on the sentence vector. Schematically, the evaluation result can be embodied in the form of a score value or a percentage, and the present application is not limited to this. For example, continuing with the example "prompt word 1-a" in step 501, the reference result corresponding to the prompt word is "positive", and the "prompt word 1-a" is input into the AI ​​model. If the output result obtained is also "positive", the evaluation result of the prompt word template a is "similarity score 100 points" or "similarity 100%"; if the output result obtained is "neutral", the evaluation result of the prompt word template is "similarity score 0 points" or "similarity 0%". It should be noted that the above method of determining the similarity between the output result and the reference result through the sentence vector is only for illustration, and the present application is not limited to this.

[0125] In addition, the above step 502 is introduced by taking the execution by the terminal as an example. In some embodiments, the process of determining the evaluation result in the above step 502 is executed by the server, that is, the terminal sends an evaluation request to the server based on the use case set and at least one prompt word template, and the server determines the evaluation result of at least one prompt word template based on the evaluation request and sends the evaluation result to the terminal. In this way, the computing resources of the terminal can be saved.

[0126] After the above steps 501 and 502, for at least one prompt word template of the AI ​​model, the terminal obtains the evaluation result of each prompt word template, which provides technical support for the subsequent display of the corresponding evaluation results. The following introduces the process of how the terminal displays the evaluation results through steps 503 to 507.

[0127] 503. Based on the evaluation results of the first use case and at least one prompt word template in the use case set, the terminal displays a display area corresponding to each prompt word template in the first area of ​​the evaluation result display interface, and displays a first evaluation identifier of the corresponding prompt word template for the first use case in each display area, wherein the first evaluation identifier indicates a usage effect of the prompt word template obtained based on the first use case.

[0128] In the embodiment of the present application, the first use case refers to any one of the use case sets. For example, the first use case is a use case in a specified position in the use case set, such as the first or last position, etc., and the present application is not limited thereto. The evaluation result display interface includes a first area, and the first area displays a display area corresponding to each prompt word template. The terminal displays the first evaluation identifier of the corresponding prompt word template for the first use case in the display area corresponding to each prompt word template based on the evaluation results of the first use case and each prompt word template. Among them, the evaluation result display interface can be automatically triggered by the terminal, for example, after the terminal executes steps 501 and 502, it automatically jumps from the prompt word template evaluation interface to the evaluation result display interface. The evaluation result display interface can also be triggered by the terminal in response to a trigger operation implemented on the client interface, for example, the terminal displays an "evaluation result view" control on the prompt word template evaluation interface, and jumps to the evaluation result display interface in response to the trigger operation implemented on the control. The present application does not limit the triggering method of the evaluation result display interface. Moreover, the terminal can jump to the evaluation result display interface after obtaining the evaluation results of all prompt word templates, or jump to the evaluation result display interface after obtaining the evaluation results of some prompt word templates, and continuously update the evaluation result display interface according to the evaluation results obtained subsequently. This application does not limit this.

[0129] In some embodiments, the first evaluation mark includes at least one of a text mark and a pattern mark. Schematically, taking the evaluation result of the prompt word template determined by the matching degree as an example, if the evaluation result is "match", the first evaluation mark includes the text mark "match" and the pattern mark "√" used to represent "match"; if the evaluation result is "mismatch", the first evaluation indicates that the text mark "mismatch" and the pattern mark "×" used to represent "mismatch" are included. Taking the evaluation result of the prompt word template determined by similarity as an example, if the evaluation result is "similarity score 88 points", the first evaluation mark includes the text mark "score 88" and the pattern mark "88". Of course, for different evaluation results, pattern marks of different colors can be used to distinguish, and the present application is not limited to this. In other words, the first evaluation mark can also be understood to include at least one of the following: a symbol for indicating the use effect, a text for indicating the use effect, and a numerical value for indicating the use effect, etc. It should be understood that all marks that can distinguish different use effects can be applied to the embodiments of the present application, and the present application is not limited to this.

[0130] In addition, different first evaluation identifiers indicate different usage effects of the prompt word template obtained based on the first use case. Schematically, taking the evaluation result of the prompt word template determined by the matching degree as an example, if the evaluation result is "match", the first evaluation identifier includes the text identifier "match" and the pattern identifier "√" used to represent "match", and the pattern identifier uses a black circular pattern; if the evaluation result is "mismatch", the first evaluation representation includes the text identifier "mismatch" and the pattern identifier "×" used to represent "mismatch", and the pattern identifier uses a gray circular pattern. For another example, taking the evaluation result of the prompt word template determined by similarity as an example, different colors can be used to distinguish different first evaluation identifiers, that is, a red circular pattern is used for a score value higher than 80, a green circular pattern is used for a score value between 60-80, and a gray circular pattern is used for a score value lower than 60. In different score intervals, the depth of the color can also be used to reflect the high and low scores, and the present application is not limited to this.

[0131] It should be understood that the various implementations of the above-mentioned first evaluation identifier can be combined arbitrarily. In practical applications, the style of the first evaluation identifier can be configured according to needs, and this application does not limit this.

[0132] In some embodiments, the display area corresponding to each prompt word template in the first area can also display the output result of each prompt word, that is, the terminal displays the first evaluation identifier of the corresponding prompt word template for the first use case and the output result corresponding to at least one first prompt word in each display area. In this way, relevant personnel can be assisted in evaluating the prompt word template and the evaluation efficiency can be improved.

[0133] Reference below Figures 6 to 8 , introduce the first area of ​​the evaluation result display interface.

[0134] Figure 6 is a schematic diagram of an evaluation result display interface provided in an embodiment of the present application. Figure 6As shown, the evaluation result display interface includes a first area 601, and the use case set includes 20 use cases. Taking the 19th use case as the first use case as an example, the terminal displays a plurality of display areas corresponding to prompt word templates 1 to 3 in the first area 601, and displays the first evaluation identifier of the corresponding prompt word template for the first use case in each display area. It should be understood that since the plurality of prompt word templates 1 to 3 are prompt word templates designed by developers, developers often select prompt word templates that meet the requirements or make further optimizations therefrom. Therefore, these prompt word templates can also be understood as a plurality of candidate sets (see the aforementioned step 501 for details, which will not be repeated here). In addition, the terminal can display the use case content of the first use case, such as "1. High cost performance, can run the game smoothly. 2. High order processing efficiency. 3. Fast logistics", and can also display the template content of the prompt word template and the generated prompt word content, etc. This application does not limit this (subsequent Figures 7 and 8 Similarly, no further details are given). Schematically, Figure 6 In the example, for prompt word template 1, the first evaluation mark is displayed: the text mark "match" and the pattern mark "√", and the pattern mark is filled with black; for prompt word template 2, the first evaluation mark is displayed: the text mark "unmatched" and the pattern mark "×", and the pattern mark is filled with lines; for prompt word template 3, since its evaluation result is being calculated, the text mark "evaluating" and the dynamic pattern mark being evaluated are displayed. It should be understood that the above-mentioned pattern marks can also be filled in other ways, and any filling method that can distinguish different usage effects can be applied to this application.

[0135] Figure 7 is a schematic diagram of another evaluation result display interface provided in an embodiment of the present application. Figure 7 As shown, the evaluation result display interface includes a first area 701, and the use case set includes 20 use cases. Taking the 19th use case as the first use case as an example, the terminal displays multiple display areas corresponding to prompt word templates 1 to 3 in the first area 701, and displays the first evaluation mark of the corresponding prompt word template for the first use case in each display area. For prompt word template 1, the first evaluation mark is displayed: the text mark "score 88" and the pattern mark "88", and the pattern mark is filled with light red; for prompt word template 2, the first evaluation mark is displayed: the text mark "score 97" and the pattern mark "97", and the pattern mark is filled with dark red; for prompt word template 3, since its evaluation result is being calculated, the text mark "evaluating" and the dynamic pattern mark in evaluation are displayed. It should be understood that the above-mentioned pattern mark can also be filled in other ways, and any filling method that can distinguish different usage effects can be applied to this application.

[0136] Figure 8 is a schematic diagram of another evaluation result display interface provided in the embodiment of the present application. Figure 8 As shown, the evaluation result display interface includes a first area 801, and the use case set includes 20 use cases. Taking the 19th use case as the first use case as an example, the terminal displays multiple display areas corresponding to prompt word templates 1 to 3 in the first area 801, and displays the first evaluation mark of the corresponding prompt word template for the first use case in each display area. For prompt word template 1, the first evaluation mark is displayed: the text mark "similarity 80%" and the pattern mark "80%", and the pattern mark is filled with green; for prompt word template 2, the first evaluation mark is displayed: the text mark "similarity 100%" and the pattern mark "100%", and the pattern mark is filled with red; for prompt word template 3, since its evaluation result is being calculated, the text mark "evaluation" and the dynamic pattern mark under evaluation are displayed. It should be understood that the above-mentioned pattern mark can also be filled in other ways, and any filling method that can distinguish different usage effects can be applied to this application.

[0137] It should be noted that the above Figures 6 to 8 The layout of the evaluation result display interface and the style of the first evaluation mark are shown for illustrative purposes only and do not constitute limitations on the present application. For example, multiple prompt word templates may be arranged horizontally or vertically. For example, the pattern mark in the first evaluation mark may be circular, rectangular, diamond-shaped, and so on.

[0138] After the above step 503, the evaluation result of each prompt word template for the first use case is displayed in the first area of ​​the evaluation result display interface in the form of an evaluation mark, so that relevant personnel can intuitively know the local use effect of using different prompt word templates on the same use case, thereby improving the human-computer interaction efficiency and further improving the evaluation efficiency and accuracy of the prompt word template.

[0139] 504. The terminal displays a second evaluation identifier of each prompt word template for the use case set in a second area of ​​the evaluation result display interface, each second evaluation identifier indicates a use effect of the prompt word template obtained based on a corresponding use case in the use case set, and the first area is adjacent to the second area.

[0140] In an embodiment of the present application, the evaluation result display interface also includes a second area, for example, the first area and the second area are adjacent to each other up and down, or to each other left and right, and the present application does not limit this. It should be understood that this step 504 and the above-mentioned step 503 can be executed simultaneously. In addition, the second evaluation mark is the same as the aforementioned first evaluation mark, so it will not be repeated here. That is, the second evaluation mark includes at least one of a text mark and a pattern mark; in other words, the second evaluation mark includes at least one of the following: a symbol for indicating the effect of use, a text for indicating the effect of use, and a numerical value for indicating the effect of use. Moreover, different second evaluation marks indicate different use effects of the prompt word template obtained based on the corresponding use case in the use case set.

[0141] In some embodiments, the terminal displays the second evaluation identification of each prompt word template for multiple use cases in the use case set in the second area in the form of a two-dimensional table. In this way, the global use effect of using different prompt word templates on the entire use case set can be clearly displayed, which is convenient for relevant personnel to make quick comparisons and improve evaluation efficiency. In other embodiments, the terminal displays the comprehensive evaluation information of each prompt word template for multiple use cases in the use case set in the second area. For example, if the use case set includes 19 use cases, after prompt word template 1 is used in these 19 use cases, 15 of the evaluation results obtained are "matched" and 4 are "unmatched". This information is displayed in the second area as comprehensive evaluation information.

[0142] In some embodiments, the second area can also display at least one prompt word template for multiple use cases in the use case set. The third evaluation mark indicates the comprehensive use effect of at least one prompt word template obtained based on the corresponding use case in the use case set. In this way, it can assist relevant personnel to evaluate the prompt word template as a whole and improve the evaluation efficiency. For example, for a certain use case, the evaluation result corresponding to the prompt word template 1 is "similarity score 80", the evaluation result corresponding to the prompt word template 2 is "similarity score 90", and the evaluation result corresponding to the prompt word template 3 is "similarity score 70". These three prompt word templates are merged into a total prompt word template, and the comprehensive evaluation result is "similarity score average 80". Accordingly, the third evaluation mark is a pattern mark "80". It should be understood that this is only for example and does not constitute a limitation of the present application. The third evaluation mark can also be displayed in other ways. In addition, the second area can simultaneously display the second evaluation mark and the third evaluation mark, or it can display one of them, and the present application does not limit this.

[0143] In some embodiments, in response to a trigger operation performed on a second use case in the second area, the terminal displays at least one prompt word template and a fourth evaluation identifier of each prompt word template for the second use case in the first area, wherein the fourth evaluation identifier indicates the use effect of the prompt word template obtained based on the second use case, and the second use case refers to other use cases in the use case set except the first use case. That is, relevant personnel can perform a trigger operation on the use case in the second area to switch the use case displayed in the first area. In other words, the second area can provide a navigation function so that relevant personnel can view the corresponding evaluation results by selecting different use cases.

[0144] Continue to refer to Figures 6 to 8 , the second area of ​​the evaluation result display interface is introduced.

[0145] like Figure 6 As shown, the evaluation result display interface also includes a second area 602. The terminal displays the second evaluation identifications of prompt word templates 1 to 3 for the 19 processed use cases in the second area 602 in the form of a two-dimensional table, and displays the corresponding comprehensive evaluation information, that is, for prompt word template 1, the number of use cases with an evaluation result of "match" is 15, and the number of use cases with an evaluation result of "mismatch" is 4; for prompt word template 2, the number of use cases with an evaluation result of "match" is 17, and the number of use cases with an evaluation result of "mismatch" is 2; for prompt word template 3, the number of use cases with an evaluation result of "match" is 17, and the number of use cases with an evaluation result of "mismatch" is 1.

[0146] like Figure 7 As shown, the evaluation result display interface also includes a second area 702. The terminal displays the second evaluation identification of the 19 used cases processed by the prompt word templates 1 to 3 in the second area 702 in the form of a two-dimensional table, and displays the corresponding comprehensive evaluation information, that is, for the prompt word template 1, the average value of the similarity score corresponding to the 19 use cases is 81 (it should be understood that the average value is only for example, and tolerance, etc. can also be used, and the present application is not limited to this); for the prompt word template 2, the average value of the similarity score corresponding to the 19 use cases is 97; for the prompt word template 3, the average value of the similarity score corresponding to the 18 use cases (the evaluation result corresponding to the 19th use case is being processed) is 67. In addition, the prompt information is displayed in the second area 702 to remind the relevant personnel of the meaning of the second evaluation identification, for example, multiple pattern identifications of 0, 20, 40, 60, 80, and 100 are used as prompt information to indicate that different scores correspond to different color types and different depths. It should be understood that each second evaluation identification in the figure is distinguished by different colors, and the specific colors are not drawn in the figure, and are explained by the bottom annotation.

[0147] like Figure 8As shown, the evaluation result display interface also includes a second area 802. The terminal displays the second evaluation identification of the prompt word templates 1 to 3 for the 19 used cases that have been processed in the second area 802 in the form of a two-dimensional table, and displays the corresponding comprehensive evaluation information, that is, for the prompt word template 1, the average value of the similarity percentage corresponding to the 19 use cases is 81%; for the prompt word template 2, the average value of the similarity percentage corresponding to the 19 use cases is 97%; for the prompt word template 3, the average value of the similarity score corresponding to the 18 use cases (the evaluation result corresponding to the 19th use case is being processed) is 67%. In addition, the prompt information is displayed in the second area 802 to remind the relevant personnel of the meaning of the second evaluation identification, for example, multiple pattern identifications of 0, 20, 40, 60, 80, and 100 are used as prompt information to indicate that different scores correspond to different color types and different depths. It should be understood that each second evaluation identification in the figure is distinguished by different colors, and the specific colors are not drawn in the figure, and are explained by the bottom annotation.

[0148] After the above step 504, in the second area of ​​the evaluation result display interface, the evaluation results of each prompt word template for each use case in the use case set are displayed in the form of an evaluation mark, so that relevant personnel can intuitively know the global use effect of using different prompt word templates on the entire use case set, improve the efficiency of human-computer interaction, and then improve the evaluation efficiency and accuracy of the prompt word template. In addition, combined with the above step 503, it can be seen that the present application divides the evaluation result display interface into two adjacent areas, and displays the evaluation results of the prompt word template in different areas of the evaluation result display interface according to different evaluation dimensions in the form of an evaluation mark, so that relevant personnel can quickly and intuitively know the entire evaluation result.

[0149] Based on the above steps 503 and 504, it can be known that the present application provides a variety of display types of evaluation identifiers, for example, an evaluation identifier based on "match" or "mismatch", and for example, an evaluation identifier based on a similarity score value or percentage. In some embodiments, the terminal determines the type of evaluation identifier that meets the evaluation result based on the evaluation result of the prompt word template. That is, the terminal can automatically select the corresponding evaluation identifier type based on the evaluation result of the prompt word template, thereby improving the display efficiency. For example, if the evaluation result is determined by the degree of match, the evaluation identifier adopts an evaluation identifier based on "match" or "mismatch", and if the evaluation result is determined by similarity, the evaluation identifier adopts an evaluation identifier based on a similarity score value or percentage, and the present application is not limited to this. Of course, the type of evaluation identifier can also be manually set by relevant personnel.

[0150] In some embodiments, the evaluation result display interface also provides a display area adjustment function for the first area and the second area and a manual intervention function for the evaluation results to further improve the human-computer interaction efficiency and evaluation efficiency. These functions are introduced below through steps 505 to 507.

[0151] 505. In response to the dragging operation performed on the target control, the terminal displays on the evaluation result display interface that the areas of the first region and the second region change as the dragging operation occurs.

[0152] In an embodiment of the present application, the target control may be displayed in the form of a dividing line between the first area and the second area, or in the form of a bar at any position of the evaluation result display interface, but the present application is not limited thereto. Displaying that the areas of the first area and the second area change with the drag operation on the evaluation result display interface means that the area change process of the first area and the second area is dynamically displayed as the drag operation is performed. Schematically, the terminal reduces the area of ​​the first area and increases the area of ​​the second area in response to a first drag operation implemented on the target control, and the first drag operation refers to a drag operation in which the drag direction points to the first area; the terminal increases the area of ​​the first area and reduces the area of ​​the second area in response to a second drag operation implemented on the target control, and the second drag operation refers to a drag operation in which the drag direction points to the second area.

[0153] Reference below Fig. 9 , an example is given to illustrate this step 505.

[0154] Fig. 9 is a schematic diagram of another evaluation result display interface provided in an embodiment of the present application. Fig. 9 As shown in Figure (a), the target control is displayed in the form of a dividing line between the first area and the second area. In response to the first drag operation implemented on the target control, that is, the upward drag operation, the terminal reduces the area of ​​the first area and increases the area of ​​the second area; in response to the second drag operation implemented on the target control, that is, the downward drag operation, the terminal increases the area of ​​the first area and reduces the area of ​​the second area. Fig. 9As shown in Figure (b), the target control is in the form of a bar and is displayed at any position of the evaluation result display interface (the position of the target control can be adjusted as needed). The target control can be understood as an adjustment track. The terminal responds to the first drag operation implemented on the target control, that is, the upward drag operation, to reduce the area of ​​the first area and increase the area of ​​the second area; responds to the second drag operation implemented on the target control, that is, the downward drag operation, to increase the area of ​​the first area and reduce the area of ​​the second area. It should be understood that the form of the target control shown in the figure is only for illustration and does not constitute a limitation of the present application. For example, the target control can also be an edit box for adjusting the area of ​​the area. The terminal can dynamically adjust the area of ​​the first area and the second area in response to the area of ​​the area entered in the target control, and so on.

[0155] In some embodiments, when the area of ​​the first area and the second area displayed changes with the dragging operation, the terminal dynamically updates the display content of the first area and the second area. That is, if the area of ​​the first area increases, other information is additionally displayed in the first area. For example, if the area of ​​the first area increases to a preset threshold or the area ratio of the first area to the second area is greater than the preset threshold, the output result corresponding to each first prompt word is displayed. If the area of ​​the second area increases, other information is additionally displayed in the second area. For example, if the area of ​​the second area increases to a preset threshold or the area ratio of the second area to the first area is greater than the preset threshold, the use case content of each use case is displayed. Accordingly, if the area of ​​the first area is reduced, the information displayed in the first area is reduced. For example, if the area of ​​the first area is reduced to a preset threshold or the area ratio of the first area to the second area is less than the preset threshold, the text logo in the first evaluation logo is hidden and the pattern logo is retained. If the area of ​​the second area is reduced, the information displayed in the second area is reduced. For example, if the area of ​​the second area is reduced to a preset threshold or the area ratio of the second area to the first area is less than the preset threshold, all prompt word templates are used as a total prompt word template, and at least one prompt word template is displayed in the second area. The third evaluation identifier for multiple use cases in the use case set, each third evaluation identifier indicates the comprehensive use effect of at least one prompt word template obtained based on the corresponding use case in the use case set. This method of dynamically updating the displayed content as the area of ​​the area changes can make full use of the interface space, simplify the operation method of viewing the evaluation results, and meet the personalized needs for viewing the evaluation results, which not only improves the evaluation efficiency, but also improves the user experience.

[0156] Reference below Fig.10 , another implementation method of step 505 is illustrated.

[0157] Fig.10is a schematic diagram of another evaluation result display interface provided in an embodiment of the present application. Fig.10 As shown in Figure (a), the evaluation result display interface includes a first area 1001 and a second area 1002. The first area 1001 includes a display area for each prompt word template. In each display area, a first evaluation identifier of the corresponding prompt word template for a first use case is displayed. In the second area 1002, a second evaluation identifier of each prompt word template for multiple use cases in the use case set is displayed. A target control 1003 "dividing line" is displayed between the first area 1001 and the second area 1002. In response to a downward drag operation on the target control 1003, the terminal increases the area of ​​the first area and decreases the area of ​​the second area. If the area of ​​the first area increases to a preset threshold or the area ratio of the first area to the second area is greater than the preset threshold, the evaluation result display interface is as shown in FIG. Fig.10 As shown in the figure (b), that is, in the display area of ​​each prompt word template in the first area 1001, the first evaluation mark of the corresponding prompt word template for the first use case and the output result corresponding to at least one first prompt word are displayed. For example, taking the sentiment analysis function shown in the aforementioned step 501 as an example, "positive" corresponds to an output value of 1, "neutral" corresponds to an output value of 2, and "negative" corresponds to an output value of 3, then the output value of prompt word template 1 is 1, the output value of prompt word template 2 is 3, and the output of prompt word template 3 is being calculated; in the second area 1002, the third evaluation mark of each prompt word template for multiple use cases in the use case set is displayed, and each third evaluation mark indicates the comprehensive use effect of all prompt word templates obtained based on the corresponding use case in the use case set. Taking the evaluation result determined by matching as an example, the same third evaluation mark can be displayed for each use case to provide a navigation function, that is, the relevant personnel can view the corresponding evaluation result by selecting different use cases. This process refers to the aforementioned step 504 and will not be repeated. Of course, in some embodiments, taking the evaluation result determined by similarity as an example, the corresponding third evaluation mark can be displayed for each use case according to the average similarity score of all prompt word templates for a certain use case, and the present application is not limited to this. It should be understood that the above Fig.10 The evaluation result display interface shown in Figure (b) can also be understood as a current use case viewing mode, that is, relevant personnel can view in detail the horizontal comparison of a certain use case in different prompt word templates to compare the effects of using different prompt word templates.

[0158] 506. The terminal displays the output result and the reference result corresponding to the target prompt word in response to the trigger operation implemented on the display area corresponding to the target prompt word template in the first area, where the target prompt word is generated based on the first use case and the target prompt word template.

[0159] In the embodiment of the present application, the target prompt word template refers to any prompt word template, and the trigger operation implemented on the display area corresponding to the target prompt word template can be a click operation, or a click operation on the prompt word template viewing control in the first area, and the present application is not limited thereto. In response to the trigger operation implemented on the display area corresponding to the target prompt word template, the terminal displays the output result and reference result corresponding to the target prompt word in the first area of ​​the evaluation result display interface for relevant personnel to view.

[0160] 507 . The terminal updates a first evaluation identifier of the target prompt word template for the first use case in response to the evaluation identifier update operation performed on the target prompt word template.

[0161] In an embodiment of the present application, the terminal provides a manual intervention function for the evaluation result, that is, the relevant personnel can update the first evaluation identifier of the target prompt word template according to the output result and the reference result of the target prompt word displayed by the terminal, so as to update the first evaluation identifier of the target prompt word template for the first use case, that is, by performing an evaluation identifier update operation on the target prompt word template. Schematically, the evaluation identifier update operation is any of the following:

[0162] The first type is an operation performed on an evaluation identifier adjustment track, where the evaluation identifier adjustment track indicates a plurality of different evaluation identifiers.

[0163] Among them, the first area displays the evaluation mark adjustment track, and the two ends of the evaluation mark adjustment track respectively indicate the upper limit and lower limit of the evaluation mark adjustment. Schematically, the evaluation mark update operation includes at least one of the following: triggering operation at any position on the evaluation mark adjustment track; sliding on the evaluation mark adjustment track to reach any position. For example, the two ends of the evaluation mark adjustment track respectively display the prompt words of the upper limit "compliant" and the lower limit "non-compliant", that is, the evaluation mark indicated by the evaluation mark adjustment track is not fixed, and the form of the evaluation mark can be set according to actual needs (for example, different evaluation marks can be distinguished by color depth). Relevant personnel select any evaluation mark by implementing a trigger operation for any position, or implementing a sliding operation on the evaluation mark adjustment track, triggering the terminal to update according to the corresponding evaluation mark. In this way, the personalized needs of relevant personnel for updating the evaluation mark can be met, thereby improving the user experience.

[0164] The second type is an operation performed on an evaluation identifier adjustment option, where the evaluation identifier adjustment option indicates a plurality of different evaluation identifiers.

[0165] Among them, the first area displays multiple evaluation mark adjustment options, and different evaluation mark adjustment options correspond to different evaluation marks. For example, the multiple evaluation mark adjustment options are 100, 80, 60, 40, 20, and 0. That is, the evaluation mark adjustment option indicates multiple fixed evaluation marks, and the relevant personnel can implement the trigger operation for a certain adjustment option, and the trigger terminal is updated according to the corresponding evaluation mark. In this way, it is convenient for relevant personnel to intuitively understand the available evaluation marks, thereby improving the efficiency of human-computer interaction.

[0166] Schematically, refer to Fig.11 , Fig.11 Schematic diagram of an evaluation identification update operation provided by an embodiment of the present application. Fig.11 As shown in Figure (a), the first area 1101 displays an evaluation mark adjustment track, and the upper limit "compliant" and lower limit "non-compliant" prompt words are displayed at both ends of the track. Relevant personnel can select any evaluation mark by performing a trigger operation at any position, or performing a sliding operation on the evaluation mark adjustment track. Fig.11 As shown in the figure (b), a plurality of evaluation mark adjustment options are displayed in the first area 1101, which are 100, 80, 60, 40, 20, 0 respectively. The relevant personnel can implement a trigger operation for a certain adjustment option to select the corresponding evaluation mark. In addition, the first area 1101 also displays a prompt word template switching control, and the terminal can respond to the trigger operation of the prompt word template switching control to switch the prompt word template currently displayed in the first area 1101. It should be understood that the evaluation mark update operation method shown in the figure is only for illustration and does not constitute a limitation of the present application. For example, an edit box for adjusting the evaluation mark can also be displayed on the evaluation result display interface, and the terminal can respond to the evaluation mark entered in the edit box. Update the corresponding evaluation mark, and so on.

[0167] In some embodiments, the evaluation result display interface also provides an evaluation result comparison function for the prompt word template, that is, the relevant personnel can select a prompt word template that better meets the requirements according to the evaluation results of multiple prompt word templates displayed by the terminal, and then update the evaluation identification of the corresponding prompt word template. Schematically, this process includes: in response to the prompt word template comparison operation implemented in the first area, displaying the evaluation results of the first prompt word template and the evaluation results of the second prompt word template; in response to the trigger operation implemented on the display area corresponding to the first prompt word template, or in response to the trigger operation implemented on the display area corresponding to the second prompt word template, updating the first evaluation identification of the first prompt word template and the second prompt word template for the first use case. Among them, the prompt word template comparison operation can be a trigger operation for the display area corresponding to a certain prompt word template (for example, double-click or long press), or it can be a trigger operation for the prompt word template comparison control in the first area, and the present application is not limited to this.

[0168] Schematically, refer to Fig.12 , Fig.12 is a schematic diagram of a prompt word template comparison operation provided in an embodiment of the present application. Fig.12 As shown, the evaluation results of the first prompt word template and the evaluation results of the second prompt word template are displayed in the first area 1201. In response to the trigger operation on the display area corresponding to any prompt word template, the first evaluation identifiers of the two prompt word templates are updated. For example, the first evaluation identifier of the first prompt word template was originally the text identifier "match", and the first evaluation identifier of the second prompt word template was originally "match". If the trigger operation is implemented on the display area corresponding to the first prompt word template, the first evaluation identifier of the first prompt word template is kept unchanged, and the first evaluation identifier of the second prompt word template is updated to "unmatched", that is, the relevant personnel can select one of the two prompt word templates to indicate that the selected prompt word template is more in line with the requirements than the unselected prompt word template. In some embodiments, the first area 1201 also displays a first control "Almost". In response to a trigger operation on the first control, the terminal updates the first evaluation identifiers of the two prompt word templates. For example, the first evaluation identifier of the first prompt word template was originally a text identifier "match", and the first evaluation identifier of the second prompt word template was originally "mismatch". If a trigger operation is performed on the first control, the first evaluation identifier of the first prompt word template remains unchanged, and the first evaluation identifier of the second prompt word template is updated to "match". Alternatively, the first evaluation identifier of the second prompt word template remains unchanged, and the first evaluation identifier of the first prompt word template is updated to "mismatch". This can be set according to actual needs, and the present application does not limit this. It should be understood that the above Fig.12The evaluation result display interface shown can also be understood as a prompt word template comparison mode, that is, relevant personnel can compare any two prompt word templates horizontally to select a prompt word template that better meets the requirements, providing a reference for further optimizing the prompt word template.

[0169] It should be noted that the above steps 505 to 507 are optional steps, that is, the terminal may execute at least one of steps 505 to 507, or may not execute them. In addition, this application does not limit the execution order of the above steps 505 to 507.

[0170] In summary, through the evaluation result display method of the prompt word template provided in the embodiment of the present application, when the prompt word template of the AI ​​model is evaluated, the evaluation results of the prompt word template are displayed in different areas of the evaluation result display interface in the form of evaluation identifiers according to different evaluation dimensions, so that relevant personnel can intuitively know the local use effect of using different prompt word templates on the same use case and the global use effect of using different prompt word templates on the entire use case set. This method effectively improves the display effect of the evaluation results, enables relevant personnel to quickly and intuitively know the evaluation results, improves the efficiency of human-computer interaction, and thereby improves the evaluation efficiency and accuracy of the prompt word template.

[0171] Fig.13 1 is a schematic diagram of the structure of a device for displaying evaluation results of a prompt word template provided in an embodiment of the present application. The device can realize part or all of the functions of the aforementioned terminal through software, hardware or a combination of both. Fig.13 As shown, the device includes an acquisition module 1301 and a display module 1302.

[0172] An acquisition module 1301 is used to acquire a use case set and at least one prompt word template, where the use case set includes multiple use cases, and the prompt word template is used to generate prompt words according to the use cases, and the prompt words are used as inputs of the AI ​​model to obtain output results corresponding to the prompt words;

[0173] A display module 1302 is used to display a display area corresponding to each prompt word template in a first area of ​​the evaluation result display interface based on a first use case and at least one prompt word template in the use case set, and display a first evaluation identifier of the corresponding prompt word template for the first use case in each display area, wherein the first evaluation identifier indicates a use effect of the prompt word template obtained based on the first use case;

[0174] Display module 1302 is also used to display the second evaluation identifier of each prompt word template for the use case set in the second area of ​​the evaluation result display interface, each second evaluation identifier indicates the use effect of the prompt word template obtained based on the corresponding use case in the use case set, and the first area is adjacent to the second area.

[0175] In some embodiments, the apparatus further comprises an evaluation result determination module, configured to:

[0176] Filling the first use case into at least one prompt word template to obtain at least one first prompt word;

[0177] Inputting at least one first prompt word into the AI ​​model to obtain an output result corresponding to the at least one first prompt word;

[0178] Based on the difference between the output result corresponding to the at least one first prompt word and the reference result, an evaluation result of the at least one prompt word template is determined, and the evaluation result is used to determine the first evaluation identifier.

[0179] In some embodiments, the evaluation result determination module is used for at least one of the following:

[0180] Determining an evaluation result of at least one prompt word template based on a matching degree between an output result corresponding to at least one first prompt word and a reference result;

[0181] Based on the similarity between the output result corresponding to the at least one first prompt word and the reference result, an evaluation result of the at least one prompt word template is determined.

[0182] In some embodiments, the first evaluation identifier includes at least one of a text identifier and a pattern identifier; and the second evaluation identifier includes at least one of a text identifier and a pattern identifier.

[0183] In some embodiments, the first evaluation identifier includes at least one of the following: a symbol for indicating the effect of use, text for indicating the effect of use, and a numerical value for indicating the effect of use; the second evaluation identifier includes at least one of the following: a symbol for indicating the effect of use, text for indicating the effect of use, and a numerical value for indicating the effect of use.

[0184] In some embodiments, different first evaluation identifiers indicate different usage effects of the prompt word template obtained based on the first use case; different second evaluation identifiers indicate different usage effects of the prompt word template obtained based on the corresponding use case in the use case set.

[0185] In some embodiments, the display module 1302 is further configured to:

[0186] In response to a drag operation performed on the target control, the areas of the first region and the second region displayed on the evaluation result display interface change along with the drag operation.

[0187] In some embodiments, the display module 1302 is used for any of the following:

[0188] In response to a first drag operation performed on the target control, reducing the area of ​​the first region and increasing the area of ​​the second region, the first drag operation being a drag operation with the drag direction pointing to the first region;

[0189] In response to a second drag operation performed on the target control, the area of ​​the first region is increased and the area of ​​the second region is reduced, wherein the second drag operation refers to a drag operation in which the drag direction points to the second region.

[0190] In some embodiments, the display module 1302 is used to:

[0191] In each display area, a first evaluation identifier corresponding to each prompt word template for a first use case and an output result corresponding to at least one first prompt word are displayed, where the first prompt word is generated based on the first use case and the prompt word template.

[0192] In some embodiments, the display module 1302 is further used to: display in the second area a third evaluation identifier of at least one prompt word template for a use case set, each third evaluation identifier indicating a comprehensive use effect of at least one prompt word template obtained based on a corresponding use case in the use case set.

[0193] In some embodiments, the display module 1302 is further configured to:

[0194] In response to a trigger operation performed on a display area corresponding to a target prompt word template in the first area, displaying an output result and a reference result corresponding to the target prompt word, wherein the target prompt word is generated based on the first use case and the target prompt word template;

[0195] In response to an evaluation identifier update operation performed on the target prompt word template, a first evaluation identifier of the target prompt word template for the first use case is updated.

[0196] In some embodiments, the evaluation identifier update operation is any of the following: an operation performed on an evaluation identifier adjustment track, where the evaluation identifier adjustment track indicates multiple different evaluation identifiers; an operation performed on an evaluation identifier adjustment option, where the evaluation identifier adjustment option indicates multiple different evaluation identifiers.

[0197] In some embodiments, the display module 1302 is further configured to:

[0198] In response to the prompt word template comparison operation performed in the first area, displaying the evaluation result of the first prompt word template and the evaluation result of the second prompt word template;

[0199] In response to a trigger operation performed on a display area corresponding to the first prompt word template, or in response to a trigger operation performed on a display area corresponding to the second prompt word template, first evaluation identifiers of the first prompt word template and the second prompt word template for the first use case are updated.

[0200] Through the above-mentioned device, when the prompt word template of the AI ​​model is evaluated, the evaluation results of the prompt word template are displayed in different areas of the evaluation result display interface in the form of evaluation marks according to different evaluation dimensions, so that relevant personnel can intuitively know the local use effect of using different prompt word templates on the same use case and the global use effect of using different prompt word templates on the entire use case set. This method effectively improves the display effect of the evaluation results, enables relevant personnel to quickly and intuitively know the evaluation results, improves the efficiency of human-computer interaction, and thereby improves the evaluation efficiency and accuracy of the prompt word template.

[0201] In addition, in the above device, both the acquisition module 1301 and the display module 1302 can be implemented by software or by hardware. Exemplarily, the implementation of the acquisition module 1301 is described below by taking the acquisition module 1301 as an example. Similarly, the implementation of other modules can refer to the implementation of the acquisition module 1301.

[0202] As an example of a software functional unit, the acquisition module 1301 may include code running on a computing instance. Among them, the computing instance may include at least one of a physical host (computing device), a virtual machine, and a container. Further, the above-mentioned computing instance may be one or more. For example, the acquisition module 1301 may include code running on multiple hosts / virtual machines / containers. It should be noted that the multiple hosts / virtual machines / containers used to run the code may be distributed in the same region (region) or in different regions. Furthermore, the multiple hosts / virtual machines / containers used to run the code may be distributed in the same availability zone (AZ) or in different AZs, each AZ including a data center or multiple data centers with similar geographical locations. Among them, usually a region may include multiple AZs.

[0203] Similarly, multiple hosts / virtual machines / containers used to run the code can be distributed in the same virtual private cloud (VPC) or in multiple VPCs. Usually, a VPC is set up in a region. For cross-region communication between two VPCs in the same region and between VPCs in different regions, a communication gateway needs to be set up in each VPC to achieve interconnection between VPCs through the communication gateway.

[0204] As an example of a hardware functional unit, the acquisition module 1301 may include at least one computing device. Alternatively, the acquisition module 1301 may also be a device implemented using an application-specific integrated circuit (ASIC) or a programmable logic device (PLD). The PLD may be a complex programmable logical device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL) or any combination thereof.

[0205] The multiple computing devices included in the acquisition module 1301 can be distributed in the same region or in different regions. The multiple computing devices included in the acquisition module 1301 can be distributed in the same AZ or in different AZs. Similarly, the multiple computing devices included in the acquisition module 1301 can be distributed in the same VPC or in multiple VPCs. The multiple computing devices can be any combination of computing devices such as servers, ASICs, PLDs, CPLDs, FPGAs, and GALs.

[0206] It should be noted that: the evaluation result display device for the prompt word template provided in the above embodiment only uses the division of the above functional modules as an example when displaying the evaluation result of the prompt word template. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the evaluation result display device for the prompt word template provided in the above embodiment and the evaluation result display method embodiment of the prompt word template belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0207] In this application, the terms "first", "second", etc. are used to distinguish between identical or similar items with substantially the same effects and functions. It should be understood that there is no logical or temporal dependency between "first", "second", and "nth", nor is the quantity and execution order limited. It should also be understood that although the following description uses the terms first, second, etc. to describe various elements, these elements should not be limited by the terms. These terms are only used to distinguish one element from another. For example, without departing from the scope of the various described examples, the first use case may be referred to as the second use case, and similarly, the second use case may be referred to as the first use case. Both the first use case and the second use case may be use cases, and in some cases, may be separate and different use cases.

[0208] The term "at least one" in this application means one or more, and the term "plurality" in this application means two or more, for example, a plurality of use cases means two or more use cases.

[0209] The above description is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0210] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware, or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of program structure information. The program structure information includes one or more program instructions. When the program instructions are loaded and executed on a computing device, all or part of the processes or functions in the embodiments of the present application are generated.

[0211] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0212] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for displaying evaluation results of a prompt word template, characterized in that: The method comprises: Acquire a use case set and at least one prompt word template, wherein the use case set includes a plurality of use cases, the prompt word template is used to generate a prompt word according to the use case, and the prompt word is used as an input of an artificial intelligence (AI) model to obtain an output result corresponding to the prompt word; Based on a first use case in the use case set and at least one of the prompt word templates, displaying a display area corresponding to each of the prompt word templates in a first area of ​​an evaluation result display interface, and displaying a first evaluation identifier of the corresponding prompt word template for the first use case in each of the display areas, wherein the first evaluation identifier indicates a use effect of the prompt word template obtained based on the first use case; The second area of ​​the evaluation result display interface displays a second evaluation identifier of each prompt word template for the use case set, each second evaluation identifier indicates a usage effect of the prompt word template obtained based on a corresponding use case in the use case set, and the first area is adjacent to the second area.

2. The method according to claim 1, characterized in that The method further comprises: Filling the first use case into at least one of the prompt word templates to obtain at least one first prompt word; Inputting at least one of the first prompt words into the AI ​​model to obtain an output result corresponding to at least one of the first prompt words; Based on a difference between an output result corresponding to at least one of the first prompt words and a reference result, an evaluation result of at least one of the prompt word templates is determined, and the evaluation result is used to determine the first evaluation identifier.

3. The method according to claim 2, characterized in that The determining of the evaluation result of at least one of the prompt word templates based on the difference between the output result corresponding to at least one of the first prompt words and the reference result includes at least one of the following: Determining an evaluation result of at least one of the prompt word templates based on a matching degree between an output result corresponding to at least one of the first prompt words and a reference result; Based on the similarity between the output result corresponding to at least one of the first prompt words and the reference result, an evaluation result of at least one of the prompt word templates is determined.

4. The method according to any one of claims 1 to 3, characterized in that The first evaluation mark includes at least one of a text mark and a pattern mark; The second evaluation mark includes at least one of a text mark and a pattern mark.

5. The method according to any one of claims 1 to 4, characterized in that The first evaluation mark includes at least one of the following: a symbol for indicating the use effect, a word for indicating the use effect, and a numerical value for indicating the use effect; The second evaluation identifier includes at least one of the following: a symbol for indicating the use effect, a word for indicating the use effect, and a numerical value for indicating the use effect.

6. The method according to any one of claims 1 to 5, characterized in that Different first evaluation identifiers indicate different use effects of the prompt word template obtained based on the first use case; Different second evaluation identifiers indicate different usage effects of the prompt word template obtained based on corresponding use cases in the use case set.

7. The method according to any one of claims 1 to 6, characterized in that The method further comprises: In response to a drag operation performed on a target control, the areas of the first region and the second region displayed on the evaluation result display interface change as the drag operation occurs.

8. The method according to claim 7, characterized in that In response to the dragging operation performed on the target control, displaying on the evaluation result display interface that the areas of the first region and the second region change as the dragging operation changes, includes any of the following: In response to a first drag operation performed on the target control, reducing the area of ​​the first region and increasing the area of ​​the second region, wherein the first drag operation refers to a drag operation in which the drag direction points to the first region; In response to a second drag operation performed on the target control, the area of ​​the first region is increased and the area of ​​the second region is reduced, wherein the second drag operation refers to a drag operation in which a drag direction points to the second region.

9. The method according to any one of claims 1 to 8, characterized in that The displaying in each display area a first evaluation mark of a corresponding prompt word template for the first use case includes: In each of the display areas, a first evaluation identifier of the prompt word template for the first use case and an output result corresponding to at least one first prompt word are displayed, where the first prompt word is generated based on the first use case and the prompt word template.

10. The method according to any one of claims 1 to 9, characterized in that The method further comprises: A third evaluation identifier of at least one of the prompt word templates for the use case set is displayed in the second area, and each of the third evaluation identifiers indicates a comprehensive use effect of at least one of the prompt word templates obtained based on a corresponding use case in the use case set.

11. The method according to any one of claims 1 to 10, characterized in that The method further comprises: In response to a trigger operation performed on a display area corresponding to a target prompt word template in the first area, displaying an output result and a reference result corresponding to a target prompt word, the target prompt word being generated based on the first use case and the target prompt word template; In response to the evaluation identifier update operation performed on the target prompt word template, a first evaluation identifier of the target prompt word template for the first use case is updated.

12. The method according to claim 11, characterized in that The evaluation identification update operation is any of the following: An operation performed on an evaluation identifier adjustment track, the evaluation identifier adjustment track indicating a plurality of different evaluation identifiers; An operation is performed on an evaluation identifier adjustment option, wherein the evaluation identifier adjustment option indicates a plurality of different evaluation identifiers.

13. The method according to any one of claims 1 to 12, characterized in that The method further comprises: In response to the prompt word template comparison operation performed in the first area, displaying the evaluation result of the first prompt word template and the evaluation result of the second prompt word template; In response to a trigger operation performed on a display area corresponding to the first prompt word template, or in response to a trigger operation performed on a display area corresponding to the second prompt word template, a first evaluation identifier of the first prompt word template and the second prompt word template for the first use case is updated.

14. A device for displaying evaluation results of a prompt word template, characterized in that: The device comprises: An acquisition module, used to acquire a use case set and at least one prompt word template, wherein the use case set includes multiple use cases, the prompt word template is used to generate a prompt word according to the use case, and the prompt word is used as an input of the AI ​​model to obtain an output result corresponding to the prompt word; A display module, configured to display, based on a first use case in the use case set and at least one of the prompt word templates, a display area corresponding to each of the prompt word templates in a first area of ​​an evaluation result display interface, and display a first evaluation identifier of the corresponding prompt word template for the first use case in each of the display areas, wherein the first evaluation identifier indicates a use effect of the prompt word template obtained based on the first use case; The display module is also used to display a second evaluation identifier of each prompt word template for the use case set in a second area of ​​the evaluation result display interface, each second evaluation identifier indicates a usage effect of the prompt word template obtained based on a corresponding use case in the use case set, and the first area is adjacent to the second area.

15. The device according to claim 14, characterized in that The device also includes an evaluation result determination module, which is used to: Filling the first use case into at least one of the prompt word templates to obtain at least one first prompt word; Inputting at least one of the first prompt words into the AI ​​model to obtain an output result corresponding to at least one of the first prompt words; Based on a difference between an output result corresponding to at least one of the first prompt words and a reference result, an evaluation result of at least one of the prompt word templates is determined, and the evaluation result is used to determine the first evaluation identifier.

16. The device according to claim 15, characterized in that The evaluation result determination module is used for at least one of the following: Determining an evaluation result of at least one of the prompt word templates based on a matching degree between an output result corresponding to at least one of the first prompt words and a reference result; Based on the similarity between the output result corresponding to at least one of the first prompt words and the reference result, an evaluation result of at least one of the prompt word templates is determined.

17. The device according to any one of claims 14 to 16, characterized in that The first evaluation mark includes at least one of a text mark and a pattern mark; The second evaluation mark includes at least one of a text mark and a pattern mark.

18. The device according to any one of claims 14 to 17, characterized in that The first evaluation mark includes at least one of the following: a symbol for indicating the use effect, a word for indicating the use effect, and a numerical value for indicating the use effect; The second evaluation identifier includes at least one of the following: a symbol for indicating the use effect, a word for indicating the use effect, and a numerical value for indicating the use effect.

19. The device according to any one of claims 14 to 18, characterized in that Different first evaluation identifiers indicate different use effects of the prompt word template obtained based on the first use case; Different second evaluation identifiers indicate different usage effects of the prompt word template obtained based on corresponding use cases in the use case set.

20. The device according to any one of claims 14 to 19, characterized in that The display module is also used for: In response to a drag operation performed on a target control, the areas of the first region and the second region displayed on the evaluation result display interface change as the drag operation occurs.

21. The device according to claim 20, characterized in that The display module is used for any of the following: In response to a first drag operation performed on the target control, reducing the area of ​​the first region and increasing the area of ​​the second region, wherein the first drag operation refers to a drag operation in which the drag direction points to the first region; In response to a second drag operation performed on the target control, the area of ​​the first region is increased and the area of ​​the second region is reduced, wherein the second drag operation refers to a drag operation in which a drag direction points to the second region.

22. The device according to any one of claims 14 to 21, characterized in that The display module is used for: In each of the display areas, a first evaluation identifier of the prompt word template for the first use case and an output result corresponding to at least one first prompt word are displayed, where the first prompt word is generated based on the first use case and the prompt word template.

23. The device according to any one of claims 14 to 22, characterized in that The display module is also used for: A third evaluation identifier of at least one of the prompt word templates for the use case set is displayed in the second area, and each of the third evaluation identifiers indicates a comprehensive use effect of at least one of the prompt word templates obtained based on a corresponding use case in the use case set.

24. The device according to any one of claims 14 to 23, characterized in that The display module is also used for: In response to a trigger operation performed on a display area corresponding to a target prompt word template in the first area, displaying an output result and a reference result corresponding to a target prompt word, the target prompt word being generated based on the first use case and the target prompt word template; In response to the evaluation identifier update operation performed on the target prompt word template, a first evaluation identifier of the target prompt word template for the first use case is updated.

25. The device according to claim 24, characterized in that The evaluation identification update operation is any of the following: An operation performed on an evaluation identifier adjustment track, the evaluation identifier adjustment track indicating a plurality of different evaluation identifiers; An operation is performed on an evaluation identifier adjustment option, wherein the evaluation identifier adjustment option indicates a plurality of different evaluation identifiers.

26. The device according to any one of claims 14 to 25, characterized in that The display module is also used for: In response to the prompt word template comparison operation performed in the first area, displaying the evaluation result of the first prompt word template and the evaluation result of the second prompt word template; In response to a trigger operation performed on a display area corresponding to the first prompt word template, or in response to a trigger operation performed on a display area corresponding to the second prompt word template, a first evaluation identifier of the first prompt word template and the second prompt word template for the first use case is updated.

27. A computing device, characterized in that The computing device includes a processor and a memory, the memory is used to store at least one section of program code, and the at least one section of program code is loaded by the processor and executes the evaluation result display method of the prompt word template according to any one of claims 1 to 13.

28. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store at least one section of program code, and the at least one section of program code is used to execute the evaluation result display method of the prompt word template according to any one of claims 1 to 13.

29. A computer program product, characterized in that When the computer program product is executed on a computing device, the computing device is enabled to execute the method for displaying evaluation results of a prompt word template according to any one of claims 1 to 13.

Images