A / B test question and answer method and device based on large model agent

Abstract

The invention relates to an A / B test question answering method and device based on a large model agent, and relates to the technical field of large models, agents and computers. The method comprises the following steps: displaying an intelligent interaction page associated with an intelligent agent; displaying a plurality of prompt word options in response to a trigger operation on a prompt word bank in the intelligent interaction page; in response to a trigger operation on a target cue word option in the plurality of cue word options, at least displaying a target cue word template corresponding to the target cue word option; in response to a content editing operation on the target cue word template, filling a target content corresponding to the content editing operation into the target cue word template to obtain a target cue word; and in response to a sending operation on the target cue word, displaying a target answer of the intelligent agent to the target cue word through the large model on the intelligent interaction page. The expression logic of the user for the intention can be optimized through the cue word, the content quality of questions is improved, then the accuracy of answers generated by the large model is improved, and the use experience of the user is improved.

Description

Technical Field

[0001] The present disclosure relates to the fields of large models, intelligent agents, and computer technology. Specifically, it relates to a method and device for A / B test question and answer based on a large model intelligent agent. Background Art

[0002] With the development of computer technology and large model technology, more and more platforms are connected to intelligent agents associated with large models. For example, in an A / B test platform, the intelligent agent can answer questions raised by users to assist users in solving problems encountered during the A / B test process.

[0003] In related technologies, users ask questions to the intelligent agent through input operations in the input box on the intelligent interaction page, and obtain answers from the intelligent agent through the large model to the questions. The expression ability of users will affect the quality of the questions, and further affect the accuracy of the large model's understanding of the user's intention and the accuracy of the answers generated by the large model. Summary of the Invention

[0004] This Summary of the Invention section is provided to introduce concepts in a brief form, which will be described in detail in the subsequent Detailed Description section. This Summary of the Invention section is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0005] In a first aspect, the present disclosure provides a method for A / B test question and answer based on a large model intelligent agent. The method for A / B test question and answer based on a large model intelligent agent includes: In response to a trigger operation on the intelligent agent identifier in the A / B test platform, display the intelligent interaction page associated with the intelligent agent, where the intelligent agent is used to answer the content input by the user in the intelligent interaction page through the large model; In response to a trigger operation on the prompt word library in the intelligent interaction page, display a plurality of prompt word options, where each prompt word option includes the name of the prompt word and usage description information, and each prompt word option is associated with a preset prompt word template; In response to a trigger operation on the target prompt word option among the plurality of prompt word options, at least display the target prompt word template corresponding to the target prompt word option; In response to a content editing operation on the target prompt word template, fill the target content corresponding to the content editing operation into the target prompt word template to obtain a target prompt word; In response to a sending operation on the target prompt word, display, in the intelligent interaction page, the target answer of the intelligent agent to the target prompt word through the large model.

[0006] Second aspect, the present disclosure provides an A / B test question-answering device based on a large model agent. The A / B test question-answering device based on a large model agent includes: A first display module, configured to display an intelligent interaction page associated with the agent in response to a trigger operation on the agent identifier in the A / B test platform. The agent is configured to answer the content input by the user in the intelligent interaction page through a large model. A second display module, configured to display a plurality of prompt word options in response to a trigger operation on the prompt word library in the intelligent interaction page. Each of the prompt word options includes the name of the prompt word and usage description information, and each prompt word option is associated with a preset prompt word template. A third display module, configured to at least display the target prompt word template corresponding to the target prompt word option in response to a trigger operation on the target prompt word option among the plurality of prompt word options. A filling module, configured to fill the target content corresponding to the content editing operation into the target prompt word template to obtain a target prompt word in response to a content editing operation on the target prompt word template. A fourth display module, configured to display the target answer of the agent to the target prompt word through the large model in the intelligent interaction page in response to a sending operation on the target prompt word.

[0007] Third aspect, the present disclosure provides a computer-readable medium, on which a computer program is stored. When the program is executed by a processing device, the steps of the method described in the first aspect are implemented.

[0008] Fourth aspect, the present disclosure provides an electronic device, including: A storage device, on which a computer program is stored; A processing device, configured to execute the computer program in the storage device to implement the steps of the method described in the first aspect.

[0009] Fifth aspect, the present disclosure provides a computer program product, including a computer program. When the computer program is executed by a processor, the steps of the method described in the first aspect are implemented.

[0010] Through the above technical solution, in the A / B test platform, users can display multiple prompt word options by triggering the prompt word library on the intelligent interaction page. Each prompt word option includes the name of the prompt word and usage description information, facilitating users to select the target prompt word template from the prompt word templates associated with the multiple prompt word options according to their needs. Then, by editing the content of the target prompt word template, users can fill in the target prompt word template according to their needs to obtain the target prompt word. Finally, by sending the target prompt word on the intelligent interaction page, the agent can obtain the target answer to the target prompt word through the large model. By adopting the above method, compared with the way of directly asking questions by users, the prompt words can optimize the expression logic of users' own intentions, improve the quality of the content of the questions, obtain target prompt words that are easy for the large model to understand, so that the large model can correctly understand the users' intentions, and further improve the accuracy of the answers generated by the large model, thereby improving the user experience.

[0011] Other features and advantages of the present disclosure will be described in detail in the subsequent specific implementation part. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Combined with the drawings and referring to the following specific implementation manners, the above and other features, advantages and aspects of the embodiments of the present disclosure will become more obvious. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic, and the original components and elements are not necessarily drawn to scale. In the drawings: Figure 1 is a flowchart of a method for A / B test question answering based on a large model agent according to an exemplary embodiment of the present disclosure; Figure 2 is a schematic diagram showing the display of a prompt word library according to an exemplary embodiment of the present disclosure; Figure 3 is a schematic diagram of a splash screen page according to an exemplary embodiment of the present disclosure; Figure 4 is a schematic diagram showing the display of a historical task record according to an exemplary embodiment of the present disclosure; Figure 5 is a schematic diagram showing the display of a historical interaction record according to an exemplary embodiment of the present disclosure; Figure 6 is a schematic diagram of the editing of a prompt word template according to an exemplary embodiment of the present disclosure; Figure 7 is another schematic diagram of the editing of a prompt word template according to an exemplary embodiment of the present disclosure; Figure 8 is a schematic diagram of the optimization of an input question according to an exemplary embodiment of the present disclosure; Figure 9It is a schematic diagram of a recommended prompt word template shown according to an exemplary embodiment of the present disclosure; Figure 10 It is a schematic diagram of a recommended test tool shown according to an exemplary embodiment of the present disclosure; Figure 11 It is a schematic diagram of a reference operation shown according to an exemplary embodiment of the present disclosure; Figure 12 It is a schematic diagram of a content retrieval process shown according to an exemplary embodiment of the present disclosure; Figure 13 It is a structural block diagram of an A / B test question-and-answer device based on a large model agent shown according to an exemplary embodiment of the present disclosure; Figure 14 It is a schematic structural diagram of an electronic device shown according to an exemplary embodiment of the present disclosure. Detailed implementation manners

[0013] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Instead, these embodiments are provided to more thoroughly and completely understand the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not used to limit the protection scope of the present disclosure.

[0014] It should be understood that the various steps recorded in the method embodiments of the present disclosure can be executed in different orders and / or executed in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this regard.

[0015] As used herein, the term "including" and its variants are open-ended, that is, "including but not limited to". The term "based on" is "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0016] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependent relationships.

[0017] It should be noted that the modifications of "one" and "multiple" mentioned in the present disclosure are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly stated in the context, it should be understood as "one or more".

[0018] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are for illustrative purposes only and are not used to limit the scope of these messages or information.

[0019] It can be understood that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, usage scopes, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to users in an appropriate manner and the authorization of users should be obtained in accordance with relevant laws and regulations.

[0020] For example, when responding to receiving an active request from a user, a prompt message is sent to the user to clearly prompt the user that the operation requested by the user will require obtaining and using the user's personal information. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, an application program, a server, or a storage medium that performs the operations of the technical solutions of the present disclosure according to the prompt message.

[0021] As an optional but non-limiting implementation manner, the manner of sending a prompt message to the user in response to receiving an active request from the user may be, for example, a pop-up window manner, and the prompt message may be presented in text in the pop-up window. In addition, the pop-up window may also carry selection controls for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0022] It can be understood that the above process of notifying and obtaining user authorization is only illustrative and does not limit the implementation manners of the present disclosure. Other manners that meet relevant laws and regulations can also be applied to the implementation manners of the present disclosure.

[0023] At the same time, it can be understood that the data involved in the technical solutions of the present disclosure (including but not limited to the data itself, the acquisition or use of the data) should comply with the requirements of corresponding laws, regulations and related provisions.

[0024] An A / B testing platform is a tool or system for conducting controlled experiments, which can help users evaluate the effects of different design schemes or strategies, so as to optimize products, services or promotion activities.

[0025] In actual use, when users ask questions, they may have difficulty clearly describing their intentions. There is a certain gap between the model answers obtained through vague or broad expressions and the user's demands, which cannot meet the user's expectations, the use experience is not smooth enough, and thus the user experience is poor.

[0026] In addition, the A / B testing platform may involve complex business processes and a large amount of business data. When connecting intelligent agents, it is necessary to consider the combination with the original business logic of the platform, as well as issues such as multiple task executions and multi-threaded scheduling.

[0027] In view of this, the present disclosure provides an A / B test question-answering method and apparatus based on a large model agent to solve the above technical problems.

[0028] The following further explains the embodiments of the present disclosure with reference to the accompanying drawings.

[0029] Figure 1 is a flowchart of an A / B test question-answering method based on a large model agent shown according to an exemplary embodiment of the present disclosure. Referring to Figure 1 this, the A / B test question-answering method based on a large model agent may include the following steps: S101: In response to a trigger operation on the agent identifier in the A / B test platform, display the intelligent interaction page associated with the agent.

[0030] Wherein, the agent is used to answer the content input by the user in the intelligent interaction page through a large model.

[0031] Exemplarily, the agent identifier may be displayed on the platform page of the A / B test platform. When the agent identifier is triggered by an operation such as clicking, the intelligent interaction page associated with the agent may be displayed, where the agent is the XX intelligent assistant in Figure 2 . Of course, the intelligent interaction page associated with the agent may also be triggered and displayed in other ways, and the embodiments of the present disclosure do not impose any restrictions on this.

[0032] S102: In response to a trigger operation on the prompt word library in the intelligent interaction page, display multiple prompt word options.

[0033] Wherein, each prompt word option includes the name of the prompt word and usage description information, and each prompt word option is associated with a preset prompt word template.

[0034] Exemplarily, as Figure 2 shown, a control 21 for triggering the prompt word library is displayed in the intelligent interaction page. In response to the trigger operation on the control 21, multiple prompt word options are displayed, and each prompt word option includes the name of the prompt word and usage description information.

[0035] It should be noted that different prompt word templates can be set for different test scenarios for users to select. For example, different test scenarios can be set based on the experimental process of the A / B test, or different test scenarios can be set for the business scenario of the A / B test platform. The present disclosure does not limit this. As Figure 2 shown, a scenario switching control can be displayed in the display area of multiple prompt word options. By triggering different scenario switching controls, the prompt word options corresponding to the corresponding test scenario are switched and displayed. A search function can also be provided to facilitate users to quickly find the required prompt word template and improve the question-answering efficiency.

[0036] S103: In response to a triggering operation on a target prompt option among multiple prompt options, at least display a target prompt template corresponding to the target prompt option.

[0037] In a possible way, in response to a triggering operation on a target prompt option among multiple prompt options, at least display a target prompt template corresponding to the target prompt option, including: in response to a triggering operation on a target prompt option among multiple prompt options, display prompt details corresponding to the target prompt option; in response to an application operation on the target prompt option, at least display a target prompt template corresponding to the target prompt option.

[0038] Exemplarily, before displaying a target prompt template corresponding to a target prompt option, details information of the target prompt template can be shown first so that the user can further determine whether to apply the prompt template. Continuing to refer to Figure 2 , in response to a triggering operation on the prompt template 3 option, display prompt details corresponding to the prompt template 3 option, such as the name of the prompt, usage description information, and the prompt template, and then in response to Figure 2 the triggering operation of the application control in , indicating that the user determines to use the prompt template, the prompt template can be displayed for the user to edit.

[0039] S104: In response to a content editing operation on the target prompt template, fill the target content corresponding to the content editing operation into the target prompt template to obtain a target prompt.

[0040] Exemplarily, as Figure 2 shown, the underlined part in the prompt template represents the content that needs to be filled, such as test parameters required for A / B testing, etc. By performing content editing on the target prompt template, the final target prompt can be obtained.

[0041] S105: In response to a sending operation on the target prompt, display, on the intelligent interaction page, a target answer of the intelligent agent to the target prompt through a large model.

[0042] Through the above technical solution, in the A / B test platform, users can display multiple prompt word options by triggering the prompt word library on the intelligent interaction page. Each prompt word option includes the name of the prompt word and usage description information, facilitating users to select the target prompt word template from the prompt word templates associated with the multiple prompt word options according to their needs. Then, by editing the content of the target prompt word template, users can fill in the target prompt word template according to their needs to obtain the target prompt word. Finally, the target prompt word is sent on the intelligent interaction page to obtain the target answer of the intelligent agent to the target prompt word through the large model. By adopting the above method, compared with the way of users directly asking questions, the prompt word can optimize the user's expression logic of their own intentions, improve the quality of the content of the question, obtain the target prompt word that is easy for the large model to understand, so that the large model can correctly understand the user's intentions, and then improve the accuracy of the answer generated by the large model, thereby improving the user's experience.

[0043] In a possible way, different intelligent interaction pages can be displayed for different Q&A scenarios. Displaying the intelligent interaction page associated with the intelligent agent includes: when the operating environment of the A / B test platform is the first environment, displaying the first recommended question for the A / B test platform on the intelligent interaction page; when the operating environment of the A / B test platform is the second environment, displaying the second recommended question for the A / B test platform and the first preset number of test scenarios on the intelligent interaction page, and in response to the triggering operation on the target scenario in the test scenarios, switching the second recommended question to the third recommended question in the target scenario; when the operating environment of the A / B test platform is the third environment, displaying the second preset number of test tools on the intelligent interaction page, and the test tools are used to analyze and process the A / B test in the A / B test platform; where the first environment is used to demonstrate the functions of the A / B test platform to users, the second environment is used for users to use the functions of the A / B test platform within the first preset duration, the third environment is used for users to use the functions of the A / B test platform within the second preset duration, and the first preset duration is less than the second preset duration.

[0044] In the embodiments of the present disclosure, an A / B test platform that can operate in different environments can be provided, which can not only meet the usage needs of different users, but also divert different users. For example, an A / B test platform operating in the first environment for demonstrating the functions of the A / B test platform to users, an A / B test platform operating in the second environment for users to use within the first preset duration, or an A / B test platform operating in the third environment for users to use within the second preset duration. Among them, the A / B test platform operating in the second environment can be used by trial users to use the platform functions during the trial period or by users to try out the new platform functions before the new functions of the platform are released in the formal environment. The third environment can be understood as the formal environment for formal users or paid users to use the platform functions.

[0045] It should be noted that in the related art, historical high-frequency questions are usually displayed on the splash screen page of the intelligent assistant. The displayed content is relatively fixed, lacking flexibility and not conforming to the real question scenarios of users.

[0046] In the embodiments of the present disclosure, different splash screen pages can be set for the A / B test platforms operating in different environments. The content is flexible and meets the usage requirements of users in different scenarios.

[0047] Exemplarily, as Figure 3 shown, the first recommended question for the A / B test platform can be displayed on the splash screen page of the A / B test platform operating in the first environment. The first recommended question can be a high-frequency question in the A / B test platform, and the present disclosure does not limit this. Thus, the basic question-and-answer ability of the A / B test platform can be demonstrated to users.

[0048] Exemplarily, as Figure 3 shown, the first preset number of test scenarios can be displayed on the splash screen page of the A / B test platform operating in the second environment. For example, different test scenarios can be set based on the experimental process of the A / B test or for the business scenarios of the A / B test platform. The present disclosure does not limit this. The recommended questions in the default scenario can be displayed by default on the splash screen page, and then in response to the test scenario switching operation, the recommended questions in the switched test scenario are displayed. The recommended questions in different test scenarios can be high-frequency questions in the corresponding test scenarios, and the present disclosure does not limit this. Thus, the basic question-and-answer ability of the A / B test platform and the guiding ability for different test scenarios can be demonstrated to users.

[0049] Exemplarily, as Figure 3 shown, the second preset number of test tools can be displayed on the splash screen page of the A / B test platform operating in the third environment. Thus, the test tools provided by the A / B test platform for the A / B test can be demonstrated to users.

[0050] It should be noted that the first preset number and the second preset number can be set according to requirements. The display content on the splash screen pages of the A / B test platforms operating in the above different environments can be set according to requirements. For example, the splash screen page corresponding to the second environment in Figure 3 can be displayed on the splash screen page of the A / B test platform operating in the third environment, and then the test tools are displayed when the user has performed the A / B test. Specifically, it can be set according to requirements, and the present disclosure does not limit this.

[0051] In a possible way, display a second preset number of test tools on the intelligent interaction page, including: in response to the A / B test in the A / B test platform being in the first test stage, display a second preset number of first test tools on the intelligent interaction page, where the first test tools are used to analyze and process the A / B test in the first test stage; in response to the A / B test in the A / B test platform being in the second test stage, display a second preset number of second test tools on the intelligent interaction page, where the second test tools are used to analyze and process the A / B test in the second test stage.

[0052] Exemplarily, corresponding test tools can be displayed according to the test stage of the A / B test in the A / B test platform. For example, during the progress of the A / B test, a traffic splitting detection tool can be displayed, such as Figure 3 the traffic splitting unevenness troubleshooting tool in, or after the A / B test is completed, a test analysis tool can be displayed, such as Figure 3 the experimental data query tool in, etc. It can be specifically set according to requirements, and the present disclosure does not limit this.

[0053] Through the above method, it can be combined with the business logic of the A / B test platform to display different splash pages and different test tools for different test stages, improving the business professionalism of the intelligent agent.

[0054] In a possible way, the A / B test Q&A method further includes: in response to there being unfinished tool tasks in the second target tool among the second preset number of test tools, display a processing task identifier around the second target tool; in response to a trigger operation on the processing task identifier, expand and display the tool tasks corresponding to the second target tool and their corresponding task statuses.

[0055] Exemplarily, when there are unfinished tool tasks in the test tool, such as unstarted or in-execution tool tasks, a processing task identifier can be displayed around the test tool, such as Figure 4 shown, the processing task identifier can represent the start countdown of the unstarted tool task, or the number of unfinished tool tasks. When including multiple tasks, different statuses of the tasks can also be displayed in different styles, and the present disclosure does not limit this.

[0056] Exemplarily, continue with the parameter Figure 4 . By triggering the processing task identifier of the second target tool, the tool tasks corresponding to the second target tool and their corresponding task statuses can be expanded and displayed, so that users can observe the execution situations of different tool tasks.

[0057] It should be understood that multiple test tools can execute different tool tasks in parallel through multi-threaded scheduling. Correspondingly, multiple tool tasks can also be executed in parallel under the same test tool, and the present disclosure does not limit this.

[0058] In a possible manner, the intelligent interaction page displays historical interaction records associated with historical question-and-answer behaviors and / or historical task records associated with historical tool tasks. The A / B test question-and-answer method further includes: in response to a trigger operation on the historical interaction record in the intelligent interaction page, displaying the historical input content and historical answer content corresponding to the historical question-and-answer behavior; in response to a trigger operation on the historical task record in the intelligent interaction page, displaying the task information of the historical tool task.

[0059] In an embodiment of the present disclosure, an entry for the historical interaction record or historical task record can be displayed on the splash screen page, so that the user can continue to perform the interaction operation corresponding to the historical interaction record or the historical task corresponding to the historical task record.

[0060] Exemplarily, as Figure 5 shown, when there is a historical question-and-answer behavior, the historical interaction record associated with the historical question-and-answer behavior, such as the historical input content, can be displayed on the intelligent interaction page. In response to a trigger operation on the historical interaction record, the intelligent dialogue page is displayed, and the detailed historical input content and the corresponding historical answer content are displayed on the intelligent dialogue page.

[0061] It should be noted that the most recent historical interaction record can be displayed on the intelligent interaction page, and then more historical input content and historical answer content corresponding to the historical question-and-answer behavior can be displayed on the intelligent dialogue page through operations such as swiping up. The present disclosure does not limit this.

[0062] Exemplarily, as Figure 4 shown, when there is a historical tool task, the historical task record associated with the historical tool task can be displayed on the intelligent interaction page. In response to a trigger operation on the historical task record, the historical task page is displayed, and the detailed historical task information is displayed on the historical task page.

[0063] It should be noted that the most recent historical task record can be displayed on the intelligent interaction page, and then more historical task records can be displayed on the historical task page. The present disclosure does not limit this. In addition, the historical tool task can be an unfinished or in-execution task for the user to continue to execute the task, or a completed task for the user to view the corresponding task information. The present disclosure does not limit this.

[0064] In a possible way, in response to a triggering operation on a target prompt option among multiple prompt options, at least display a target prompt template corresponding to the target prompt option, including: in response to a first triggering operation on the target prompt option among multiple prompt options, display the target prompt template corresponding to the target prompt option in an input box on the intelligent interaction page, where the target prompt template includes missing content to be input. In response to a content editing operation on the target prompt template, fill the target content corresponding to the content editing operation into the target prompt template to obtain a target prompt, including: in response to an input operation on the missing content in the target prompt template in the input box, fill the target content corresponding to the input operation into the target prompt template to obtain a target prompt.

[0065] Exemplarily, the selected prompt template can be brought into the input box on the intelligent interaction page for the user to edit the content according to needs, such as filling in the test parameters required for A / B testing. The present disclosure does not limit this. As Figure 5 shown, the underlined part is the content to be filled. Fill the content edited by the user into the prompt template to obtain the final target prompt.

[0066] It should be noted that the part to be filled can be input manually, or the candidate content can be expanded by triggering the part to be filled for the user to select. For example, if the part to be filled is the test parameters required for A / B testing, then expand the candidate test parameters and candidate parameter values for the user to select, so as to improve the user's editing efficiency, optimize the logical expression of the user's question, and further improve the accuracy of the model's answer.

[0067] In a possible way, in response to a triggering operation on a target prompt option among multiple prompt options, at least display a target prompt template corresponding to the target prompt option, including: in response to a second triggering operation on the target prompt option among multiple prompt options, display the target prompt template corresponding to the target prompt option and the parameter configuration items of the parameters required in the target prompt template in the dialogue area on the intelligent interaction page, where the parameter configuration items include preset parameter values for the user to select. In response to a content editing operation on the target prompt template, fill the target content corresponding to the content editing operation into the target prompt template to obtain a target prompt, including: in response to a selection operation among the preset parameter values, fill the preset parameter value corresponding to the selection operation into the target prompt template to obtain a target prompt.

[0068] Exemplarily, as Figure 7As shown, after selecting the target prompt word option, the target prompt word template corresponding to the target prompt word option and the parameter configuration items of the parameters required in the target prompt word template can be displayed in the conversation area of the intelligent interaction page, so that the user can directly fill in the content to be filled in the target prompt word template through the parameter configuration items to obtain the final target prompt word. For example, select the test parameters and corresponding parameter values required for A / B testing. The present disclosure does not limit this, thereby improving the user's editing efficiency, optimizing the logical expression of the user's question, and further improving the accuracy of the model's answer.

[0069] It should be noted that for the above two editing methods, one of the methods can be selected according to the business requirements for the user to edit, or the selection items corresponding to the two editing methods can be displayed for the user to select by themselves. The present disclosure does not limit this.

[0070] In a possible way, the A / B test question and answer method further includes: in response to a first input operation in the input box on the intelligent interaction page, a first prompt word template is displayed in the input box. The first prompt word template is optimized by the intelligent agent through the large model based on the first intention recognition result of the first input operation corresponding to the first input content. The first intention recognition result is obtained by the intelligent agent through the large model for intention recognition of the first input content.

[0071] Exemplarily, as Figure 8 shown, in response to the user entering content in the input box on the intelligent interaction page, a prompt word template for optimizing the input content can be automatically displayed. Or an optimization control 81 can be displayed on the intelligent interaction page, and then in response to the user's triggering operation on the optimization control 81 after the user enters the content, a prompt word template for optimizing the input content is displayed. Or in response to the user's triggering operation on the sending control 82 after the user enters the content, intercept the sending of the user input content, and then display a prompt word template for optimizing the input content. It can also be the case that when the user's intention cannot be determined after the large model performs intention recognition on the user input content, a prompt word template for optimizing the input content is displayed, etc. The present disclosure does not limit this. Thereby, the user's expression can be quickly optimized, the quality of the question content can be improved, and a target prompt word that is easy for the large model to understand can be obtained, so that the large model can correctly understand the user's intention, and further improve the accuracy of the answer generated by the large model, thereby improving the user's experience.

[0072] It should be noted that the first prompt template is obtained by optimizing the user input content based on the result of intent recognition by the large model after the large model recognizes the intent of the user input content. The first prompt template can be a directly usable prompt template in business scenarios not involving parameter settings, and can also be a prompt template that needs to be filled in business scenarios involving parameter settings, which can be determined according to the actual business scenario, and the present disclosure does not limit this.

[0073] In a possible manner, the A / B test Q&A method further includes: in response to a sending operation of the second input content in the input box of the intelligent interaction page, displaying the second input content and the second prompt template in the conversation area of the intelligent interaction page; or, in response to generating a first answer for the second input content, displaying the first answer and the second prompt template in the conversation area; wherein, the second prompt template is obtained by the intelligent agent from the prompt library based on the second intent recognition result, and the second intent recognition result is obtained by the large model recognizing the intent of the second input content.

[0074] In the embodiments of the present disclosure, after the user sends an input question, the large model can be used to recognize the intent of the user's question, and a recommended prompt template can be selected from the prompt library based on the intent recognition result, or the prompt word optimized for the user's question can be directly displayed, and the present disclosure does not limit this. Thereby, it can prompt the user to optimize the expression logic of their own intent through the prompt word, improve the quality of the question content, obtain a target prompt word that is easy for the large model to understand, so that the large model can correctly understand the user's intent, and further improve the accuracy of the answer generated by the large model, thereby improving the user experience.

[0075] Exemplarily, as Figure 9 shown, after the user sends an input question, the question and the recommended prompt template can be displayed in the conversation area of the intelligent interaction page, or after the large model generates an answer to the question, the answer and the recommended prompt template can be displayed in the conversation area of the intelligent interaction page. It is also possible to use the large model to recognize the intent of the user input question, and when the large model cannot accurately judge the user's intent, then display the recommended prompt template, etc., and the present disclosure does not limit this.

[0076] It should be noted that after the user confirms to apply the recommended prompt template, the prompt template can be brought into the input box, or the parameter configuration items required for the prompt template can be further displayed, and the present disclosure does not limit this.

[0077] In a possible way, the A / B test Q&A method further includes: identifying the intention of the third input content entered by the user on the intelligent interaction page through a large model; when the result of the intention recognition represents that the third input content matches the first target tool in the test tools provided by the A / B test platform, displaying the call entry and / or tool usage instructions of the first target tool, and the test tool is used to analyze and process the A / B test in the A / B test platform; in response to the trigger operation on the call entry, displaying the parameter configuration items for the user to configure the first target tool.

[0078] In the embodiments of the present disclosure, as Figure 10 shown, when the input content of the user is recognized by the large model as being related to the capabilities of the test tools provided by the A / B test platform, the call entry and / or tool usage instructions of the tool can be displayed in the dialogue area so that the user can use the test tool to solve the problems raised. Of course, the test tool control displayed on the intelligent interaction page can also be directly triggered to call the test tool.

[0079] It should be noted that different business scenarios can be determined according to information such as the input content of the user and the A / B test being performed, and then the test tools corresponding to the corresponding scenarios can be matched for the user to use. The tool usage instructions can include information such as tool capabilities and tool usage steps.

[0080] Exemplarily, as Figure 10 shown, after the user determines to call the test tool, a specific type of test tool can also be automatically called. For example, when checking for uneven grouping, the uneven grouping checking tool can be called, and the pre-operations for using the test tool can also be automatically performed, such as obtaining the data required for using the test tool, etc., to improve the efficiency of using the test tool. Then, the parameter configuration items for configuring the test tool are displayed for the user to configure according to their needs. The present disclosure does not limit this. Among them, the execution process of the pre-operations and the specific tool call process can be displayed on the intelligent interaction page, such as Figure 10 the intelligent interaction page shown in the middle, or the specific execution process or tool call process may not be displayed. The present disclosure does not limit this.

[0081] By adopting the above method, on the basis of optimizing the user's question, an intention correction and fault tolerance mechanism can be provided to help the large model correctly understand the user's intention, or on the basis of providing an answer, optimization suggestions can be provided to help the user better express their intention. In addition, instant feedback can be provided according to the user's operation, the status and process during the operation can be displayed, and subsequent operation guidance can be provided.

[0082] In a possible way, the A / B test Q&A method further includes: in response to a reference operation on historical interaction content in the intelligent interaction page, displaying the target interaction content corresponding to the reference operation in the input box of the intelligent interaction page; in response to a second input operation in the input box, obtaining fourth input content based on the input content corresponding to the second input operation and the target interaction content; in response to a send operation on the fourth input content, displaying a second answer of the intelligent agent to the fourth input content through a large model in the intelligent interaction page.

[0083] In the embodiments of the present disclosure, operations such as re - editing, referencing, copying, and sharing historical interaction content in the intelligent interaction page can be performed.

[0084] Exemplarily, as Figure 11 shown, by triggering the re - editing control 111 of the historical interaction content, the corresponding historical interaction content can be brought into the input box for re - editing, so that the user can dynamically adjust the question content according to the model answer in the feedback, improving the Q&A quality.

[0085] Exemplarily, as Figure 11 shown, by triggering the reference control 112 of the historical interaction content, the corresponding historical interaction content can be displayed in the input box in a referenced form, so that the user can supplement new input content on this basis. The large model combines the historical interaction content and the new input content to obtain a new answer, improving the Q&A quality by enhancing the context relevance. One or more historical interaction contents can be referenced, and the present disclosure does not limit this.

[0086] Exemplarily, continuing to refer to Figure 11 , the historical interaction content can be copied by triggering the copy control 113 of the historical interaction content, or the historical interaction content can be shared by triggering the share control 114 of the historical interaction content. Both the copy operation and the share operation can be performed on one or more historical interaction contents, and the present disclosure does not limit this.

[0087] In a possible way, as Figure 12 shown, in the conversation area of the intelligent interaction page, during the process of waiting for the large model to output an answer, the retrieval content of the intelligent agent can be dynamically displayed. For example, the source of the retrieval content can be displayed according to the retrieval progress to clarify the information source channel. When the answer is output, the retrieval content can be folded, and then the retrieval content can be expanded to view, and it can also be jumped to the corresponding source for viewing. For example, if content B is retrieved from website A, then it can be jumped to website A to view content B, and the present disclosure does not limit this. By providing a visual information flow when outputting the answer, the information source can be traced, the model answer has a basis, and the answer credibility is improved.

[0088] Through the above method, for the Q&A scenario of the A / B test platform, it can cover the A / B test platform in different operating environments, carry the test tools related to the platform, divert users, and display different splash pages according to the historical interaction behaviors of users, such as displaying recommended questions, scenario-based guidance, etc. By providing a prompt word library, optimizing prompt words, and recommending prompt word templates, the quality of users' questions can be improved, and it can assist users in obtaining higher-quality answers. And it can give users immediate feedback, understand users' intentions through large models, and provide quick tool call entrances to assist users in connecting the platform business paths and helping users complete intelligent interactions more easily.

[0089] Based on the same concept, the embodiments of the present disclosure also provide an A / B test Q&A device based on a large model agent, as Figure 13 shown. The A / B test Q&A device 130 based on a large model agent may include: A first display module 131, configured to display an intelligent interaction page associated with the agent in response to a trigger operation on the agent identifier in the A / B test platform, where the agent is used to answer the content input by the user in the intelligent interaction page through a large model; A second display module 132, configured to display a plurality of prompt word options in response to a trigger operation on the prompt word library in the intelligent interaction page, where each of the prompt word options includes the name of the prompt word and usage description information, and each of the prompt word options is associated with a preset prompt word template; A third display module 133, configured to at least display the target prompt word template corresponding to the target prompt word option in response to a trigger operation on the target prompt word option among the plurality of prompt word options; A filling module 134, configured to fill the target content corresponding to the content editing operation into the target prompt word template in response to a content editing operation on the target prompt word template to obtain a target prompt word; A fourth display module 135, configured to display the target answer of the agent to the target prompt word through the large model in the intelligent interaction page in response to a sending operation on the target prompt word.

[0090] Optionally, the third display module 133 is configured to: In response to a first trigger operation on the target prompt word option among the plurality of prompt word options, display the target prompt word template corresponding to the target prompt word option in the input box of the intelligent interaction page, where the target prompt word template includes the missing content to be input; The filling module 134 is configured to: In response to an input operation on the missing content in the target prompt template in the input box, fill the target content corresponding to the input operation into the target prompt template to obtain a target prompt.

[0091] Optionally, the third display module 133 is configured to: In response to a second trigger operation on a target prompt option among the multiple prompt options, display the target prompt template corresponding to the target prompt option and parameter configuration items for the required parameters in the target prompt template, where the parameter configuration items include preset parameter values for the user to select; The filling module 134 is configured to: In response to a selection operation among the preset parameter values, fill the preset parameter value corresponding to the selection operation into the target prompt template to obtain a target prompt.

[0092] Optionally, the third display module 133 is configured to: In response to a trigger operation on a target prompt option among the multiple prompt options, display the prompt details corresponding to the target prompt option; In response to an application operation on the target prompt option, at least display the target prompt template corresponding to the target prompt option.

[0093] Optionally, the A / B test question and answer device 130 based on the large model agent further includes: A first input model, configured to, in response to a first input operation in the input box on the intelligent interaction page, display a first prompt template in the input box, where the first prompt template is obtained by the agent optimizing the first input content corresponding to the first input operation through the large model based on a first intent recognition result, and the first intent recognition result is obtained by the agent performing intent recognition on the first input content through the large model.

[0094] Optionally, the A / B test question and answer device 130 based on the large model agent further includes a second input module, and the second input module is configured to: In response to a sending operation on second input content in the input box on the intelligent interaction page, display the second input content and a second prompt template in the conversation area of the intelligent interaction page; or, In response to generating a first answer for the second input content, display the first answer and the second prompt template in the conversation area; Among them, the second prompt template is obtained by the agent from the prompt library through the large model based on the second intention recognition result, and the second intention recognition result is obtained by the large model performing intention recognition on the second input content.

[0095] Optionally, the A / B test question-answering device 130 based on the large model agent further includes a third input module, and the second input module is configured to: Perform intention recognition on the third input content input by the user on the intelligent interaction page through the large model; When the result of the intention recognition indicates that the third input content matches the first target tool in the test tools provided by the A / B test platform, display the call entry of the first target tool and / or the tool usage instructions, and the test tool is used to analyze and process the A / B test in the A / B test platform; In response to a trigger operation on the call entry, display parameter configuration items for the user to configure the first target tool.

[0096] Optionally, the A / B test question-answering device 130 based on the large model agent further includes a reference module, and the reference module is configured to: In response to a reference operation on the historical interaction content on the intelligent interaction page, display the target interaction content corresponding to the reference operation in the input box of the intelligent interaction page; In response to a second input operation in the input box, obtain a fourth input content based on the input content corresponding to the second input operation and the target interaction content; In response to a send operation on the fourth input content, display a second answer of the agent to the fourth input content through the large model on the intelligent interaction page.

[0097] Optionally, the first display module 131 is configured to: When the operating environment of the A / B test platform is the first environment, display a first recommended question for the A / B test platform on the intelligent interaction page; When the operating environment of the A / B test platform is the second environment, display a second recommended question for the A / B test platform and a first preset number of test scenarios on the intelligent interaction page. In response to a trigger operation on the target scenario in the test scenarios, switch the second recommended question to a third recommended question under the target scenario; When the operating environment of the A / B test platform is the third environment, display a second preset number of test tools on the intelligent interaction page, and the test tools are used to analyze and process the A / B test in the A / B test platform; Among them, the first environment is used to demonstrate the functions of the A / B test platform to the user, the second environment is used for the user to use the functions of the A / B test platform within a first preset duration, the third environment is used for the user to use the functions of the A / B test platform within a second preset duration, and the first preset duration is less than the second preset duration.

[0098] Optionally, the first display module 131 is configured to: In response to the A / B test in the A / B test platform being in the first test stage, display a second preset number of first test tools on the intelligent interaction page, where the first test tools are used to analyze and process the A / B test in the first test stage; In response to the A / B test in the A / B test platform being in the second test stage, display a second preset number of second test tools on the intelligent interaction page, where the second test tools are used to analyze and process the A / B test in the second test stage.

[0099] Optionally, the A / B test question-answering device 130 based on the large model agent further includes a fifth display module, and the fifth display module is configured to: In response to an unfinished tool task existing in a second target tool among the second preset number of test tools, display a processing task identifier around the second target tool; In response to a trigger operation on the processing task identifier, expand and display the tool task corresponding to the second target tool and its corresponding task status.

[0100] Optionally, the intelligent interaction page displays historical interaction records associated with historical question-answering behaviors and / or historical task records associated with historical tool tasks. The A / B test question-answering device 130 based on the large model agent further includes a sixth display module, and the sixth display module is configured to: In response to a trigger operation on the historical interaction record on the intelligent interaction page, display the historical input content and historical answer content corresponding to the historical question-answering behavior; In response to a trigger operation on the historical task record on the intelligent interaction page, display the task information of the historical tool task.

[0101] Based on the same concept, an embodiment of the present disclosure also provides a computer-readable medium, on which a computer program is stored, and when the program is executed by a processing device, the steps of any of the above A / B test question-answering methods based on the large model agent are implemented.

[0102] Based on the same concept, an embodiment of the present disclosure also provides an electronic device, which may include: A storage device on which a computer program is stored; A processing device for executing the computer program in the storage device to implement the steps of any of the above A / B test question-answering methods based on large model agents.

[0103] Based on the same concept, an embodiment of the present disclosure also provides a computer program product, including a computer program which, when executed by a processor, implements the steps of any of the above A / B test question-answering methods based on large model agents.

[0104] Next, refer to Figure 14 , which shows a schematic structural diagram of an electronic device (140 suitable for implementing the embodiments of the present disclosure. The terminal device in the embodiments of the present disclosure may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Tablet Computers), PMPs (Portable Multimedia Players), in-vehicle terminals (such as in-vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 14 The electronic device shown is only an example and should not impose any limitations on the functions and usage scope of the embodiments of the present disclosure.

[0105] As Figure 14 shown, the electronic device 140 may include a processing device (such as a central processing unit, a graphics processing unit, etc.) 141, which may perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 142 or the program loaded from the storage device 148 into the random access memory (RAM) 143. In the RAM 143, various programs and data required for the operation of the electronic device 140 are also stored. The processing device 141, the ROM 142, and the RAM 143 are connected to each other through a bus 144. The input / output (I / O) interface 145 is also connected to the bus 144.

[0106] Generally, the following devices may be connected to the I / O interface 145: an input device 146 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 147 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 148 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 149. The communication device 149 may allow the electronic device 140 to communicate with other devices wirelessly or wiredly to exchange data. Although Figure 14 shows the electronic device 140 having various devices, it should be understood that it is not required to implement or have all the shown devices. More or fewer devices may be alternatively implemented or had.

[0107] In particular, according to an embodiment of the present disclosure, the processes described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product that includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes program code for performing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 149, or installed from the storage device 148, or installed from the ROM 142. When the computer program is executed by the processing device 141, the above-mentioned functions defined in the method of the embodiment of the present disclosure are performed.

[0108] It should be noted that the above-mentioned computer-readable medium in the present disclosure can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, the computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present disclosure, a computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable medium other than the computer-readable storage medium, and the computer-readable signal medium can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted by any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0109] In some embodiments, any currently known or future-developed network protocol such as HTTP (HyperText Transfer Protocol) can be used for communication, and it can be interconnected with digital data communication in any form or medium (e.g., communication network). Examples of communication networks include local area networks (“LAN”), wide area networks (“WAN”), the Internet (e.g., the Internet), and end-to-end networks (e.g., ad hoc end-to-end networks), as well as any currently known or future-developed networks.

[0110] The above computer-readable medium can be included in the above electronic device; or it can exist separately without being assembled into the electronic device.

[0111] The above computer-readable medium carries one or more programs. When the above one or more programs are executed by the electronic device, the electronic device is caused to: in response to a trigger operation on an agent identifier in an A / B test platform, display an intelligent interaction page associated with the agent, where the agent is used to answer the content input by the user in the intelligent interaction page through a large model; in response to a trigger operation on a prompt word library in the intelligent interaction page, display a plurality of prompt word options, where each of the prompt word options includes the name of the prompt word and usage description information, and each of the prompt word options is associated with a preset prompt word template; in response to a trigger operation on a target prompt word option among the plurality of prompt word options, at least display the target prompt word template corresponding to the target prompt word option; in response to a content editing operation on the target prompt word template, fill the target content corresponding to the content editing operation into the target prompt word template to obtain a target prompt word; in response to a sending operation on the target prompt word, display, in the intelligent interaction page, the target answer of the agent to the target prompt word through the large model.

[0112] Computer program code for performing the operations of the present disclosure can be written in one or more programming languages or combinations thereof. The above programming languages include, but are not limited to, object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the “C” language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (e.g., by using an Internet service provider to connect through the Internet).

[0113] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a part of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that, in some alternative implementations, the functions noted in the blocks may occur in a different order than noted in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0114] The modules described in the embodiments of the present disclosure can be implemented in software or in hardware. In some cases, the name of a module does not constitute a limitation on the module itself.

[0115] The functions described above herein can be performed, at least in part, by one or more hardware logic components. By way of example, and without limitation, the types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system on a chip (SOCs), complex programmable logic devices (CPLDs), and the like.

[0116] In the context of the present disclosure, a machine-readable medium may be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0117] The above description is only a preferred embodiment of the present disclosure and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosure concept. For example, the technical solutions formed by mutually replacing the above features with the technical features (but not limited to) having similar functions disclosed in the present disclosure.

[0118] In addition, although the operations are depicted in a particular order, this should not be construed as requiring that the operations be performed in the particular order shown or in sequential order. In certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the present disclosure. Certain features described in the context of separate embodiments may also be implemented in combination in a single embodiment. Conversely, the various features described in the context of a single embodiment may also be implemented separately or in any suitable sub-combination in multiple embodiments.

[0119] Although the subject matter has been described in language specific to structural features and / or methodological logical acts, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. On the contrary, the specific features and acts described above are merely example forms for implementing the claims. Regarding the apparatus in the above embodiments, the specific manner in which each module performs operations has been described in detail in the embodiments related to the method, and will not be elaborated herein.

Claims

1. An A / B test Q&A method based on large model agents, characterized in that, The A / B test question-answering method based on a large model agent includes: In response to a trigger operation on the agent identifier in the A / B test platform, display the intelligent interaction page associated with the agent, where the agent is used to answer the content input by the user in the intelligent interaction page through a large model; In response to a trigger operation on the prompt word library in the intelligent interaction page, display multiple prompt word options, where each prompt word option includes the name of the prompt word and usage description information, and each prompt word option is associated with a preset prompt word template; In response to a trigger operation on the target prompt word option among the multiple prompt word options, at least display the target prompt word template corresponding to the target prompt word option; In response to a content editing operation on the target prompt word template, fill the target content corresponding to the content editing operation into the target prompt word template to obtain a target prompt word; In response to a sending operation on the target prompt word, display the target answer of the agent to the target prompt word through the large model in the intelligent interaction page.

2. The A / B test Q&A method based on large model agents according to claim 1, wherein, The step of, in response to a trigger operation on the target prompt word option among the multiple prompt word options, at least displaying the target prompt word template corresponding to the target prompt word option includes: In response to a first trigger operation on the target prompt word option among the multiple prompt word options, display the target prompt word template corresponding to the target prompt word option in the input box of the intelligent interaction page, where the target prompt word template includes blank content to be input; The step of, in response to a content editing operation on the target prompt word template, filling the target content corresponding to the content editing operation into the target prompt word template to obtain a target prompt word includes: In response to an input operation on the blank content in the target prompt word template in the input box, fill the target content corresponding to the input operation into the target prompt word template to obtain a target prompt word.

3. The A / B test question-answering method based on large model agents according to claim 1, wherein, The step of, in response to a trigger operation on the target prompt word option among the multiple prompt word options, at least displaying the target prompt word template corresponding to the target prompt word option includes: In response to a second trigger operation on the target prompt word option among the multiple prompt word options, display the target prompt word template corresponding to the target prompt word option and the parameter configuration items of the required parameters in the target prompt word template in the dialogue area of the intelligent interaction page, where the parameter configuration items include preset parameter values for the user to select; The step of, in response to a content editing operation on the target prompt word template, filling the target content corresponding to the content editing operation into the target prompt word template to obtain a target prompt word includes: In response to a selection operation among the preset parameter values, fill the preset parameter value corresponding to the selection operation into the target prompt word template to obtain a target prompt word.

4. The A / B test Q&A method based on large model agents according to any one of claims 1-3, characterized in that The step of, in response to a trigger operation on the target prompt word option among the multiple prompt word options, at least displaying the target prompt word template corresponding to the target prompt word option includes: In response to a triggering operation on a target prompt option among the multiple prompt options, display the prompt details corresponding to the target prompt option; In response to an application operation on the target prompt option, at least display the target prompt template corresponding to the target prompt option.

5. The A / B test question-answering method based on a large model agent according to any one of claims 1-3, characterized in that The A / B test Q&A method further includes: In response to a first input operation in the input box on the intelligent interaction page, display a first prompt template in the input box, where the first prompt template is optimized by the intelligent agent through the large model based on a first intent recognition result for the first input content corresponding to the first input operation, and the first intent recognition result is obtained by the intelligent agent through the large model for intent recognition of the first input content.

6. The A / B test question-answering method based on a large model agent according to any one of claims 1-3, characterized in that, The A / B test Q&A method further includes: In response to a sending operation on a second input content in the input box on the intelligent interaction page, display the second input content and a second prompt template in the conversation area of the intelligent interaction page; or, In response to generating a first answer for the second input content, display the first answer and the second prompt template in the conversation area; wherein the second prompt template is obtained by the intelligent agent through the large model based on a second intent recognition result from the prompt library, and the second intent recognition result is obtained by the large model for intent recognition of the second input content.

7. The A / B test question-answering method based on large model agents according to any one of claims 1-3, characterized in that, The A / B test Q&A method further includes: Perform intent recognition on a third input content input by the user on the intelligent interaction page through the large model; When the result of the intent recognition indicates that the third input content matches a first target tool in the test tools provided by the A / B test platform, display a call entry and / or tool usage instructions for the first target tool, where the test tool is used to analyze and process the A / B test in the A / B test platform; In response to a triggering operation on the call entry, display parameter configuration items for the user to configure the first target tool.

8. The A / B test Q&A method based on the large model agent according to any one of claims 1-3, characterized in that The A / B test Q&A method further includes: In response to a reference operation on historical interaction content on the intelligent interaction page, display the target interaction content corresponding to the reference operation in the input box on the intelligent interaction page; In response to a second input operation in the input box, obtain a fourth input content based on the input content corresponding to the second input operation and the target interaction content; In response to a sending operation on the fourth input content, display a second answer for the fourth input content by the intelligent agent through the large model on the intelligent interaction page.

9. The A / B test Q&A method based on large model agents according to any one of claims 1-3, characterized in that, The display of the intelligent interaction page associated with the intelligent agent includes: When the operating environment of the A / B test platform is a first environment, display a first recommended question for the A / B test platform on the intelligent interaction page; When the operating environment of the A / B test platform is the second environment, a second recommended question for the A / B test platform and a first preset number of test scenarios are displayed on the intelligent interaction page. In response to a trigger operation on a target scenario among the test scenarios, the second recommended question is switched to a third recommended question in the target scenario; When the operating environment of the A / B test platform is the third environment, a second preset number of test tools are displayed on the intelligent interaction page, and the test tools are used to analyze and process the A / B test in the A / B test platform; Among them, the first environment is used to demonstrate the functions of the A / B test platform to the user, the second environment is used for the user to use the functions of the A / B test platform within a first preset duration, the third environment is used for the user to use the functions of the A / B test platform within a second preset duration, and the first preset duration is less than the second preset duration.

10. The A / B test question-answering method based on large model agents according to claim 9, wherein, The step of displaying a second preset number of test tools on the intelligent interaction page includes: In response to the A / B test in the A / B test platform being in the first test stage, a second preset number of first test tools are displayed on the intelligent interaction page, and the first test tools are used to analyze and process the A / B test in the first test stage; In response to the A / B test in the A / B test platform being in the second test stage, a second preset number of second test tools are displayed on the intelligent interaction page, and the second test tools are used to analyze and process the A / B test in the second test stage.

11. The A / B test question-answering method based on large model agents according to claim 9, wherein The A / B test question-and-answer method further includes: In response to an unfinished tool task existing in a second target tool among the second preset number of test tools, a processing task identifier is displayed around the second target tool; In response to a trigger operation on the processing task identifier, the tool task corresponding to the second target tool and its corresponding task status are displayed in an expanded manner.

12. The A / B test Q&A method based on the large model agent according to any one of claims 1-3, characterized in that, The intelligent interaction page displays historical interaction records associated with historical question-and-answer behaviors and / or historical task records associated with historical tool tasks. The A / B test question-and-answer method further includes: In response to a trigger operation on the historical interaction record on the intelligent interaction page, the historical input content and historical answer content corresponding to the historical question-and-answer behavior are displayed; In response to a trigger operation on the historical task record on the intelligent interaction page, the task information of the historical tool task is displayed.

13. An A / B test question-answering device based on a large model agent, characterized in that, The A / B test question-and-answer device based on the large model agent includes: A first display module, configured to display an intelligent interaction page associated with the agent in response to a trigger operation on the agent identifier in the A / B test platform, and the agent is used to answer the content input by the user on the intelligent interaction page through the large model; The second display module is configured to display a plurality of prompt word options in response to a triggering operation on the prompt word library in the intelligent interaction page, wherein each of the prompt word options includes the name of the prompt word and usage description information, and each of the prompt word options is associated with a preset prompt word template; The third display module is configured to at least display the target prompt word template corresponding to the target prompt word option in response to a triggering operation on the target prompt word option among the plurality of prompt word options; The filling module is configured to fill the target content corresponding to the content editing operation into the target prompt word template in response to a content editing operation on the target prompt word template, so as to obtain a target prompt word; The fourth display module is configured to display the target answer of the intelligent agent to the target prompt word through the large model in the intelligent interaction page in response to a sending operation on the target prompt word.

14. A computer-readable medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processing device, it implements the steps of the method according to any one of claims 1-12.

15. An electronic device, characterized in that, Comprising: A storage device having a computer program stored thereon; A processing device configured to execute the computer program in the storage device to implement the steps of the method according to any one of claims 1-12.

16. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-12.

Images