Large model multi-agent-based leadless group interview system and method

Abstract

The invention discloses a leaderless group interview system and method based on large model multi-agent, and belongs to the technical field of artificial intelligence interview. The method comprises the following steps: generating interview questions by an enterprise human resource manager and a corresponding agent; scoring standard making, role description, role reference answer scheme generation and role cue word generation; after an interviewer receives the conference participation invitation and starts a leader-free group interview process, a preparation stage, an independent statement stage, a free discussion stage and a summary statement stage are performed in sequence, and a final interview record is generated; the enterprise human resource manager sends the interviewer name, the interview question, the final interview record and the scoring standard to the scoring agent, and checks and revises the interview score and evaluation output by the scoring agent, and finally the background management terminal automatically sends the interview score and evaluation to the interviewer. And completing leaderless group interview based on the large-model multi-agent.

Description

Technical Field

[0001] The present invention belongs to the technical field of artificial intelligence interview, and in particular to a leaderless group interview system and method based on large model multi-agent. Background Art

[0002] There are three problems in the implementation of the existing leaderless group interview:

[0003] 1. Efficiency: The human resource manager of an enterprise needs to spend a lot of time and energy to organize all the interviewees to a fixed time and a fixed space before the interview can proceed smoothly. Therefore, all interviewees need to meet the time and space matching. The matching threshold is relatively high. Once an interviewee is absent, the human resource manager of the enterprise needs to have strong organizational and coordination skills, which will cause consumption of multiple parties and may even affect the smooth progress of the entire recruitment plan.

[0004] 2. Accuracy: Existing leaderless interview questions have the attribute of non-unique answers, and require multiple interviewers to score, which makes it difficult to ensure the consistency of the scores; and through the subjective analysis of the interviewers, it is difficult to ensure the objectivity of the leaderless group interview.

[0005] 3. Record retention: The existing process data of leaderless interviews is difficult to preserve and record, and it is impossible to review and audit the interviewees. Summary of the invention

[0006] The purpose of the present invention is to solve the problems existing in the prior art and to provide a leaderless group interview system and method based on a large model multi-agent, which is used to improve the efficiency of leaderless group interviews, enhance the objectivity and accuracy of interviewers' scoring of leaderless group interviews, retain records of the entire interview process, and facilitate review and audit of interviewees.

[0007] In order to achieve the above-mentioned invention object, the present invention specifically adopts the following technical solutions:

[0008] In a first aspect, the present invention provides a leaderless group interview system based on a large model multi-agent, which comprises:

[0009] The question-generating module is used to receive the interview question reference source sent by the enterprise human resources manager, and send the interview question reference source to the question-generating agent, which generates the corresponding interview question according to the interview question reference template, and then sends the interview question generated by the question-generating agent to the scoring standard-setting agent, which generates the corresponding scoring standard according to the pre-built scoring standard-setting prompt words. When the role-analyzing agent receives the interview question sent by the enterprise human resources manager and the preset number of role agents, the role-analyzing agent outputs the role description and role reference answer scheme corresponding to each role agent, and finally inputs the interview question, role description and role reference answer scheme together into the prompt word-generating agent, which outputs the role prompt words corresponding to each role agent;

[0010] The interview configuration module is used to send the interview meeting invitation information to the interviewer terminal through the background management terminal. When the interviewer terminal receives the interview trigger instruction of the interviewer, the interview management terminal starts the leaderless group interview process;

[0011] The interview module is used in the preparation stage of the leaderless group interview process. The interview management terminal interface displays the precautions of the preparation stage and the interview questions to the interviewee. After receiving the instruction that the interviewee has finished reading within the specified time, the independent statement stage is entered; the interview management terminal interface displays the precautions of the independent statement stage to the interviewee. After receiving the instruction that the interviewee has finished reading, the interview management terminal records the views stated by the interviewee within the specified time, and enters the free discussion stage after the statement is completed; the interview management terminal interface displays the precautions of the free discussion stage to the interviewee. After receiving the instruction that the interviewee has finished reading, the interview management terminal switches to the free discussion interface and records the behavior and voice of the interviewee. After receiving the instruction that the interviewee believes that a consensus has been reached, the closing statement stage is entered; the interview management terminal interface displays the precautions of the closing statement stage to the interviewee. After receiving the instruction that the interviewee has finished reading, if the interviewee makes a closing statement within the specified time, the interview management terminal records the interviewee's closing statement and generates a final interview record;

[0012] The scoring module is used to receive the interviewee's name, interview topic, final interview record and scoring criteria sent by the company's human resources manager and send them to the scoring agent, which generates prompt words based on the score to output the interview score and evaluation. Upon receiving instructions from the company's human resources manager to complete the content review and revision of the interview score and evaluation output by the scoring agent, the background management terminal automatically sends the interview score and evaluation to the interviewee.

[0013] Based on the above scheme, each step can be implemented in the following preferred specific manner.

[0014] As a preferred method, the interview question reference sources are the interview question template reference source, the first type of job requirement reference source and the second type of job requirement reference source; wherein the interview question template reference source is the theme provided by the enterprise human resources manager; the first type of job requirement reference source is the job requirement information matched in the knowledge graph constructed by the personal information data and interview data of the employees who have passed the interview and successfully joined the company; the second type of job requirement reference source is the job description provided by the enterprise human resources manager;

[0015] The scoring criteria formulation prompts need to include both the scoring criteria template preset by the company's human resources manager and the thinking chain of the scoring criteria formulation agent when generating the scoring criteria. The scoring criteria template includes the types of abilities that need to be assessed during the interview, the scoring weights of each ability type, and a specific description of each ability that needs to be assessed.

[0016] As a preferred embodiment, the specific process of the backend management terminal sending the interview meeting invitation information to the interviewer terminal is as follows: when the backend management terminal receives the interview question book name and the interview questions generated by the question generation agent sent by the enterprise's human resources manager, the backend management terminal creates an interview question book entry and stores it in the database; when the backend management terminal receives the interview meeting name, interview time, interview question book and interviewer contact information sent by the enterprise's human resources manager, the backend management terminal creates an interview meeting entry and stores it in the database, and at the same time generates interview meeting invitation information and sends it to the interviewer terminal.

[0017] Preferably, during the preparation stage of the leaderless group interview process, the interview management terminal automatically plays a video of the preparation stage precautions described by the virtual interviewer and displays it on the interview management terminal interface for the interviewee to watch. After the preparation stage precautions video is played, the interview management terminal interface displays the interview questions to the interviewee and requires the interviewee to complete reading the interview questions within the specified time. If the interview management terminal interface receives an instruction from the interviewee to complete reading the interview questions in advance, it enters the independent statement stage.

[0018] Preferably, during the independent statement stage in the leaderless group interview process, the interview management terminal automatically plays a video of the precautions for the independent statement stage described by the virtual interviewer and displays it on the interview management terminal interface for the interviewee to watch. After the video of the precautions for the independent statement stage is played, the interview management terminal automatically turns on the camera and microphone to record the interviewee's behavior and voice, and requires the interviewee to complete the statement of views within the specified time. Upon receiving the instruction that the interviewee has completed the statement of views, the interview management terminal converts the audio of the interviewee's statement of views into text to obtain a text record of the independent statement. If the interview management terminal interface receives the instruction from the interviewee to complete the statement of views in advance, the free discussion stage is entered.

[0019] Preferably, during the free discussion stage in the leaderless group interview process, the interview management terminal automatically plays a video of the precautions for the free discussion stage described by the virtual interviewer and displays it on the interview management terminal interface for the interviewee to watch. After the video of the precautions for the free discussion stage is played, the interview management terminal automatically switches to the free discussion interface and automatically turns on the camera and microphone to record the interviewee's behavior and voice. The free discussion interface consists of a floating window and multiple avatar images. The floating window plays the real-time behavior of the interviewee collected by the interview management terminal through the camera. Each avatar image shows a role agent image configured by the company's human resources manager in the question-setting module. Each role agent plays a different interview role based on its corresponding role prompt word. When the interviewer or each role agent is speaking, the frame color of the floating window or the frame color of the avatar image is highlighted. Before the end of the free discussion, if the interview management terminal interface receives an instruction from the interviewee that he or she believes that a consensus has been reached, the closing statement stage is entered.

[0020] Preferably, in the free discussion stage of the leaderless group interview process, at the beginning of the free discussion stage, the interview management terminal asks the interviewee whether he needs to speak first through a pop-up window on the interface. If the interviewee's selection instruction is not received within the specified time, it is deemed that the interviewee automatically gives up the opportunity to speak first; during the free discussion stage, each role agent speaks in a preset fixed order. At the end of each role agent's speech, the interview management terminal asks the interviewee whether he wants to speak through a pop-up window on the interface. If the interviewer's selection instruction is not received within the specified time, it is deemed that the interviewee automatically gives up the opportunity to speak; during the free discussion stage, all role agents and interviewers each have a serial number to determine the order of each round of speeches. The interviewer's serial number is displayed below the floating window representing the interviewer, and the serial number of the role agent is displayed below the avatar picture representing the role agent. When the total number of role agents is m, the serial number of the role agent is randomly generated by the background management terminal, and the starting serial number of the role agent is 1, the ending serial number is m, and the serial number of the interviewer is m+1.

[0021] Preferably, during the free discussion stage in the leaderless group interview process, when the interview management terminal turns on the camera and microphone to record the interviewee's behavior and voice, the interview management terminal sends the interviewee's behavior video file and sound file to the background management terminal, which stores the two files in a database, and at the same time uses an audio-to-text model to convert the interviewee's voice file into a speech text, and stores the speech text and the interviewer's serial number in the database as part of the free discussion record; when the background management terminal sends the free discussion record to the corresponding role agent, the role agent outputs the speech text, and the background management terminal stores the role agent's speech text and the role agent's serial number in the database as another part of the free discussion record, and at the same time uses a text-to-audio model to convert the speech text generated by the role agent into speech audio and sends it to the interview management terminal, which automatically plays the speech audio after receiving it.

[0022] As a preferred method, during the concluding statement stage of the leaderless group interview process, the interview management terminal automatically plays a video of the precautions for the concluding statement stage described by the virtual interviewer and displays it on the interview management terminal interface for the interviewee to watch. After the video of the precautions for the concluding statement stage is played, the interview management terminal asks the interviewee whether to make a concluding statement through a pop-up window on the interface:

[0023] If the interviewee receives an instruction to make a closing statement within the specified time, the interview management terminal automatically turns on the camera and microphone to record the interviewee's behavior and voice and requires the interviewee to make a closing statement within the specified time. After receiving an instruction from the interviewee that the closing statement is complete, the interview management terminal converts the audio of the interviewee's closing statement into text to obtain a closing statement text record. If the interview management terminal receives an instruction from the interviewee to complete the closing statement in advance, the interview is terminated, and the free discussion record obtained in the free discussion stage, the independent statement text record obtained in the independent statement stage, and the closing statement text record obtained in the closing statement stage together constitute the final interview record;

[0024] If the interviewee's instruction to make a closing statement is not received within the specified time, it will be deemed that the interviewee automatically gives up the opportunity to make a closing statement. At this time, the free discussion record obtained in the free discussion stage and the independent statement text record obtained in the independent statement stage together constitute the final interview record. The free discussion record obtained in the free discussion stage is sent by the background management terminal to the closing statement intelligent agent, and the closing statement intelligent agent outputs the closing statement text. At the same time, the background management terminal stores the closing statement text in the database, and uses the text-to-audio model to convert the closing statement text generated by the closing statement intelligent agent into closing statement audio and send it to the interview management terminal. The interview management terminal automatically plays the closing statement audio after receiving it, and automatically ends the interview after playing.

[0025] In a second aspect, the present invention provides a leaderless group interview method using the leaderless group interview system described in the first aspect, and the specific steps are as follows:

[0026] The question-setting module generates interview questions, scoring criteria and role prompts corresponding to each role agent; the interview configuration module sends the interview meeting invitation information to the interviewee's terminal and starts the leaderless group interview process; the role agents play the corresponding interview roles according to the role prompts of each role agent, and the interview module generates the final interview record of the interviewee; the interview questions and scoring criteria generated by the question-setting module are input into the scoring module to obtain the interviewee's interview score and evaluation, thus completing the leaderless group interview based on large model multi-agent.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] 1. The present invention proposes a leaderless group interview system and method based on a large model and multiple agents, which utilizes the content generation capability of a large language model to solve the problem of fixed time and space required for leaderless group interviews, thereby improving the efficiency of leaderless interviews.

[0029] 2. The present invention proposes to use the theme provided by the enterprise human resources manager as the first input data, the job requirements matched in the knowledge graph constructed by the personal information data and interview data of employees who have passed the interview and successfully joined the company as the second input data, and the job description provided by the enterprise human resources manager as the third input data. The three reference sources are used as the basis for the generation of interview questions, and the corresponding scoring standards are formulated based on the generated interview questions, which improves the objectivity of the leaderless group interview scoring and improves the matching degree between the interviewee and the job requirements.

[0030] 3. The leaderless group video interview method provided by the present invention can record the entire process, thereby achieving traceability and facilitating the review and examination of the interviewee. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 is a system block diagram of the present invention;

[0032] Figure 2 is a flow chart of the steps of the method of the present invention;

[0033] Figure 3 is a functional block diagram of the method of the present invention;

[0034] Figure 4 A swimming lane diagram of the method of the present invention;

[0035] Figure 5 It is a schematic diagram of the process of the leaderless group interview in the present invention. DETAILED DESCRIPTION

[0036] In order to make the above-mentioned purpose, features and advantages of the present invention more obvious and easy to understand, the specific implementation mode of the present invention is described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below. The technical features in each embodiment of the present invention can be combined accordingly without conflicting with each other.

[0037] like Figure 1 As shown, in a preferred implementation of the present invention, the leaderless group interview system based on large model multi-agent includes:

[0038] The question-generating module is used to receive the interview question reference source sent by the enterprise human resources manager, and send the interview question reference source to the question-generating agent, which generates the corresponding interview question according to the interview question reference template, and then sends the interview question generated by the question-generating agent to the scoring standard-setting agent, which generates the corresponding scoring standard according to the pre-built scoring standard-setting prompt words. When the role-analyzing agent receives the interview question sent by the enterprise human resources manager and the preset number of role agents, the role-analyzing agent outputs the role description and role reference answer scheme corresponding to each role agent, and finally inputs the interview question, role description and role reference answer scheme together into the prompt word-generating agent, which outputs the role prompt words corresponding to each role agent;

[0039] The interview configuration module is used to send the interview meeting invitation information to the interviewer terminal through the background management terminal. When the interviewer terminal receives the interview trigger instruction of the interviewer, the interview management terminal starts the leaderless group interview process;

[0040] The interview module is used in the preparation stage of the leaderless group interview process. The interview management terminal interface displays the precautions of the preparation stage and the interview questions to the interviewee. After receiving the instruction that the interviewee has finished reading within the specified time, the independent statement stage is entered; the interview management terminal interface displays the precautions of the independent statement stage to the interviewee. After receiving the instruction that the interviewee has finished reading, the interview management terminal records the views stated by the interviewee within the specified time, and enters the free discussion stage after the statement is completed; the interview management terminal interface displays the precautions of the free discussion stage to the interviewee. After receiving the instruction that the interviewee has finished reading, the interview management terminal switches to the free discussion interface and records the behavior and voice of the interviewee. After receiving the instruction that the interviewee believes that a consensus has been reached, the closing statement stage is entered; the interview management terminal interface displays the precautions of the closing statement stage to the interviewee. After receiving the instruction that the interviewee has finished reading, if the interviewee makes a closing statement within the specified time, the interview management terminal records the interviewee's closing statement and generates a final interview record;

[0041] The scoring module is used to receive the interviewee's name, interview topic, final interview record and scoring criteria sent by the company's human resources manager and send them to the scoring agent, which generates prompt words based on the score to output the interview score and evaluation. Upon receiving instructions from the company's human resources manager to complete the content review and revision of the interview score and evaluation output by the scoring agent, the background management terminal automatically sends the interview score and evaluation to the interviewee.

[0042] It should be noted that the question setting module of the present invention includes 4 parts. The first is the interview question generation part, the second is the scoring standard formulation part, the third is the role description and role reference answer solution generation part, and the fourth is the role prompt word generation part. The role description and role reference answer solution generation part and the role prompt word generation part can be unified as the intelligent agent role configuration part.

[0043] In the generation of interview questions, it is necessary to rely on three interview question reference sources to generate interview questions that meet the requirements of relevant positions and have a high degree of matching. Specifically, the interview question reference sources are the interview question template reference source, the first type of position requirement reference source, and the second type of position requirement reference source; among them, the interview question template reference source is the theme provided by the company's human resources manager; the first type of position requirement reference source is the job requirement information matched in the knowledge graph constructed by the personal information data and interview data of employees who have passed the interview and successfully joined the company; the second type of position requirement reference source is the job description provided by the company's human resources manager.

[0044] In this embodiment, relying on the above three interview question reference sources and using them as the data input part of the question generation agent, relying on the text generation ability of the core large language model agent, finally generate interview questions that conform to the theme template structure and meet the corresponding job requirements.

[0045] In this embodiment, when generating interview questions, examples of prompt words used are as follows:

[0046] #Role

[0047] You are a leaderless group interview question maker. Your job is to create new leaderless group interview questions based on the interview question reference information provided.

[0048] #Skill

[0049] You need to create new leaderless group interview questions based on the interview question template, Job Requirements 1, and Job Requirements 2 in the reference source.

[0050] #Reference source

[0051] ##Interview question template

[0052] {{Interview question template insertion location}}

[0053] ##Job Requirements 1

[0054] {{Job requirement - insert position}}

[0055] ##Job Requirements 2

[0056] {{Insert position for job requirement 2}}

[0057] #limit

[0058] - You cannot directly use the interview question template

[0059] -Don't mention information you took out of context

[0060] -Don't mention how you work

[0061] In the scoring standard formulation part, it is necessary to rely on specific interview questions to generate corresponding scoring standards. Specifically, the interview questions generated by the question generation agent are used as the data input part of the scoring standard formulation agent. The scoring standard formulation agent combines the scoring standard formulation prompt words and analyzes the types of abilities that need to be assessed during the interview, the specific description of each ability that needs to be assessed, and the scoring weight of each ability type contained in the scoring standard template to generate specific detailed items and finally obtain a complete scoring standard.

[0062] The scoring criteria formulation prompt of the present invention needs to include both the scoring criteria template preset by the enterprise human resources manager and the thinking chain of the scoring criteria formulation intelligent agent when generating the scoring criteria. The scoring criteria template includes the types of abilities that need to be assessed during the interview, the scoring weights of each ability type, and a specific description of each ability that needs to be assessed.

[0063] It should also be noted that in the scoring standard template of this embodiment, the sum of the scoring weights of all ability types is 1. That is, when the scoring standard formulation agent analyzes that there are n types of ability types for the interviewer in the interview question, then the scoring weights corresponding to these n ability types are α1,…,α n , and satisfies That is, the sum of the scoring weights of all ability types is 1, α i represents the scoring weight corresponding to the i-th ability type. Therefore, the formula corresponding to the total score is:

[0064]

[0065] In this embodiment, when formulating the scoring criteria, the scoring criteria formulation prompt words used are specifically as follows:

[0066] #Role

[0067] You are an expert in developing scoring standards for leaderless group interviews.

[0068] Be able to analyze the interview questions to assess the specific abilities of the interviewees based on the leaderless group interview questions and formulate corresponding scoring standards.

[0069] #Skill

[0070] You need to analyze the interview questions in the reference source to find out the specific abilities of the interviewees to be assessed by the interview questions, and formulate corresponding scoring criteria based on the scoring criteria template in the reference source.

[0071] #Reference source

[0072] Interview Questions

[0073] {{Interview question insert location}}

[0074] ##Grading criteria template

[0075] |Score Weight|Specific Abilities|Ability Description|

[0076] |----|--------|--------|

[0077] |0.25|Problem Solving Skills|Can the interviewee come up with creative and practical solutions to solve XXXX problems? |

[0078] |0.2|Communication skills|Whether the interviewee can clearly and effectively convey information and persuade others when dealing with XXXX. |

[0079] |0.1|Stress Management|Whether the interviewee can remain calm and make reasonable decisions when facing emergencies and pressure from multiple parties. |

[0080] |0.15|Crisis handling ability|Whether the interviewee can quickly identify a crisis and take effective measures to mitigate or avoid the crisis. |

[0081] |0.3|Long-term planning ability|Can the interviewee come up with strategies or suggestions to prevent similar problems in the future? |

[0082] Total score = (0.25 × problem-solving ability score) + (0.2 × communication skills score) + (0.1 × stress management score) + (0.15 × crisis handling ability score) + (0.3 × long-term planning ability score)

[0083] #limit

[0084] - You cannot directly use the scoring criteria template

[0085] -Don't mention information you took out of context

[0086] -Don't mention how you work

[0087] In the agent role configuration part, the interview questions generated by the interview question generation part are first relied on, and together with the preset number of role agents, they are used as the data input part of the role analysis agent. The role analysis agent generates role descriptions and role reference answer solutions that meet the preset number of role agents by analyzing the interview questions. Then, the generated multiple role descriptions and role reference answer solutions and interview questions are used as data inputs of the prompt word generation agent, and the role prompt words corresponding to each role agent are generated with reference to the default role prompt word template, which is used for the role agent to play the corresponding role.

[0088] In this embodiment, when generating the role description and the role reference answer scheme, the prompt words used are specifically as follows:

[0089] #Role

[0090] You are a leaderless group interview topic role analyst,

[0091] Your function is to analyze the interview questions provided and assess the specific abilities of the interviewees.

[0092] And according to the number of interviewees in the reference source, assign corresponding roles to each interviewee.

[0093] #Skill

[0094] You need to analyze the interview questions in the reference source to determine the specific abilities of the interviewees to be assessed by the interview questions, and assign corresponding roles to each interviewee from the perspective of answering questions based on the number of interviewees in the reference source.

[0095] - Describe the role in detail

[0096] -And provide reference solutions to answer questions from different roles' perspectives.

[0097] #Reference source

[0098] ##Number of interviewees

[0099] {{Insert the number of character agents}}

[0100] Interview Questions

[0101] {{Interview question insert location}}

[0102] #limit

[0103] -Don't mention information you took out of context

[0104] -Don't mention how you work

[0105] In this embodiment, when generating role prompt words, the prompt words used are specifically as follows:

[0106] #Role

[0107] You are a prompt editor for a leaderless group interview role.

[0108] Ability to write role prompts for large language model agents based on the role's description.

[0109] #Skill

[0110] Based on the interview questions and role descriptions in the reference source, you need to answer the interview questions from the perspective of the corresponding role as your goal, and write role prompts for the large language model agent to play.

[0111] #Reference source

[0112] Interview Questions

[0113] {{Interview question insert location}}

[0114] ##Character Description

[0115] {{insert role description location}}

[0116] ##Role reference answer plan

[0117] {{Insert position of character reference answer plan}}

[0118] ##Prompt word template

[0119] ```text

[0120] #Interview Role

[0121] Name of the character

[0122] #describe

[0123] Detailed description of the role

[0124] #Skill

[0125] Character Skills

[0126] #Specific scenarios and expressions

[0127] Specific scenarios and answer options for the characters to refer to, at least 5

[0128] #limit

[0129] -Don't mention information you took out of context

[0130] -Don't mention how you work

[0131] ```

[0132] #limit

[0133] -Don't mention information you took out of context

[0134] -Don't mention how you work

[0135] It should be noted that, in the interview configuration module of the present invention, the specific process of the background management terminal sending the interview meeting invitation information to the interviewer terminal is as follows: when the background management terminal receives the interview question book name and the interview questions generated by the question generation agent sent by the enterprise's human resources manager, the background management terminal creates an interview question book entry and stores it in the database; when the background management terminal receives the interview meeting name, interview time, interview question book and interviewer contact information sent by the enterprise's human resources manager, the background management terminal creates an interview meeting entry and stores it in the database, and at the same time generates interview meeting invitation information and sends it to the interviewer terminal.

[0136] In this embodiment, the interviewer terminal can be a mobile device with an interview function, or other forms, as long as the interviewer can obtain the interview meeting invitation information from it. The interviewer contact information specifically refers to the interviewer's contact number or email address. The interview meeting invitation information generated by the background management terminal can be an interview meeting invitation link, which includes an interview meeting invitation code, or other forms, as long as the interviewer can obtain the method of entering the leaderless group interview and the arrangement of the interview meeting, which is not limited in the present invention.

[0137] In this embodiment, the interviewee can open the interview management terminal and enter the interview invitation code to enter the leaderless group interview. The interview management terminal here can specifically be a WeChat applet, an Alipay applet, etc., as long as the device that allows the interviewee to participate in the interview can be used, so it is not limited in the present invention.

[0138] It should be noted that in the interview module of the present invention, the leaderless group interview process includes four stages: Figure 5 As shown, there are preparation stage, independent statement stage, free discussion stage and closing statement stage. The interviewee completes the interview in the order of the above four stages.

[0139] 1) During the preparation stage of the leaderless group interview process, the interview management terminal automatically plays a video of the preparation stage precautions described by the virtual interviewer and displays it on the interview management terminal interface for the interviewee to watch. After the preparation stage precautions video is played, the interview management terminal interface displays the interview questions to the interviewee and requires the interviewee to complete the reading of the interview questions within the specified time. If the interview management terminal interface receives an instruction from the interviewee to complete the reading of the interview questions in advance, it enters the independent statement stage.

[0140] 2) During the independent statement stage of the leaderless group interview process, the interview management terminal automatically plays a video of the precautions for the independent statement stage described by the virtual interviewer and displays it on the interview management terminal interface for the interviewee to watch. After the video of the precautions for the independent statement stage is played, the interview management terminal automatically turns on the camera and microphone to record the interviewee's behavior and voice, and requires the interviewee to complete the statement of views within the specified time. When receiving the interviewee's instruction to complete the statement of views, the interview management terminal converts the audio of the interviewee's statement of views into text to obtain an independent statement text record. If the interview management terminal interface receives the interviewee's instruction to complete the statement of views in advance, it enters the free discussion stage.

[0141] 3) During the free discussion stage of the leaderless group interview process, the interview management terminal automatically plays a video of the precautions for the free discussion stage described by the virtual interviewer and displays it on the interview management terminal interface for the interviewee to watch. After the video of the precautions for the free discussion stage is played, the interview management terminal automatically switches to the free discussion interface and automatically turns on the camera and microphone to record the interviewee's behavior and voice. The free discussion interface consists of a floating window and multiple avatar images. The floating window plays the real-time behavior of the interviewee collected by the interview management terminal through the camera. Each avatar image shows a role agent image configured by the company's human resources manager in the question-setting module. Each role agent plays a different interview role based on its corresponding role prompt word. When the interviewer or each role agent is speaking, the frame edge color of the floating window or the frame edge color of the avatar image is highlighted. Before the end of the free discussion, if the interview management terminal interface receives an instruction from the interviewee that he or she believes that a consensus has been reached, the closing statement stage is entered.

[0142] Furthermore, at the beginning of the free discussion stage, the interview management terminal asks the interviewee whether he needs to speak first through a pop-up window on the interface. If the interviewee's selection instruction is not received within the specified time, it is deemed that the interviewee automatically gives up the opportunity to speak first; during the free discussion stage, each role agent speaks in a preset fixed order. When each role agent finishes speaking, the interview management terminal asks the interviewee whether he wants to speak through a pop-up window on the interface. If the interviewer's selection instruction is not received within the specified time, it is deemed that the interviewee automatically gives up the opportunity to speak; during the free discussion stage, all role agents and interviewers each have a serial number to determine the order of each round of speeches, and this serial number is also used to distinguish different speaking objects. The interviewer's serial number is displayed below the floating window representing the interviewer, and the serial number of the role agent is displayed below the avatar image representing the role agent. These serial numbers are randomly generated by the background management terminal at the beginning of the free discussion stage based on the total number n of role agents. When the total number of role agents is m, the serial number of the role agent is randomly generated by the background management terminal, and the starting serial number of the role agent is 1, the ending serial number is m, and the serial number of the interviewer is m+1.

[0143] In this embodiment, the prompt word template used by the role agent is as follows:

[0144] #Role

[0145] You are an interviewee in a leaderless group interview. You need to participate in the discussion on the interview questions from the perspective of the interview role, analyze the remarks made by other interviewees, and express your own opinions.

[0146] #Reference source

[0147] Interview Questions

[0148] {{Interview question insert location}}

[0149] ##Interview Role

[0150] {{insert the role's name at position}}

[0151] ##describe

[0152] {{insert the detailed description of the role}}

[0153] ##Skill

[0154] {{The character's skill insert position}}

[0155] ##Specific scenarios and expressions

[0156] {{Specific scenes and expression insertion locations for character reference}}

[0157] #Free discussion record

[0158] {{Free discussion record insertion position}}

[0159] #limit

[0160] -Don't mention information you took out of context

[0161] -Don't mention how you work

[0162] Furthermore, in the free discussion stage of the leaderless group interview process, when the interview management terminal turns on the camera and microphone to record the interviewee's behavior and voice, the interview management terminal sends the interviewee's behavior video file and sound file to the background management terminal, and the background management terminal stores the two files in the database. At the same time, the audio-to-text model is used to convert the interviewee's sound file into a speech text, and the speech text and the interviewer's serial number are stored in the database as part of the free discussion record; when the background management terminal sends the free discussion record to the corresponding role agent, the role agent outputs the speech text, and the background management terminal stores the role agent's speech text and the role agent's serial number in the database as another part of the free discussion record. At the same time, the background management terminal uses the text-to-audio model to convert the speech text generated by the role agent into speech audio and sends it to the interview management terminal. The interview management terminal automatically plays the speech audio after receiving it.

[0163] 4) During the concluding statement stage of the leaderless group interview process, the interview management terminal automatically plays a video of the precautions for the concluding statement stage described by the virtual interviewer and displays it on the interview management terminal interface for the interviewee to watch. After the concluding statement stage precautions video is played, the interview management terminal asks the interviewee whether to make a concluding statement through a pop-up window on the interface:

[0164] If the interviewee receives an instruction to make a closing statement within the specified time, the interview management terminal automatically turns on the camera and microphone to record the interviewee's behavior and voice and requires the interviewee to make a closing statement within the specified time. After receiving an instruction from the interviewee that the closing statement is complete, the interview management terminal converts the audio of the interviewee's closing statement into text to obtain a closing statement text record. If the interview management terminal receives an instruction from the interviewee to complete the closing statement in advance, the interview is terminated, and the free discussion record obtained in the free discussion stage, the independent statement text record obtained in the independent statement stage, and the closing statement text record obtained in the closing statement stage together constitute the final interview record;

[0165] If the interviewee's instruction to make a closing statement is not received within the specified time, it will be deemed that the interviewee automatically gives up the opportunity to make a closing statement. At this time, the free discussion record obtained in the free discussion stage and the independent statement text record obtained in the independent statement stage together constitute the final interview record. The free discussion record obtained in the free discussion stage is sent by the background management terminal to the closing statement intelligent agent, and the closing statement intelligent agent outputs the closing statement text. At the same time, the background management terminal stores the closing statement text in the database, and uses the text-to-audio model to convert the closing statement text generated by the closing statement intelligent agent into closing statement audio and send it to the interview management terminal. The interview management terminal automatically plays the closing statement audio after receiving it, and automatically ends the interview after playing.

[0166] In this embodiment, the prompt words used by the concluding statement agent are as follows:

[0167] #Role

[0168] You are an interviewee giving a closing statement in a leaderless group interview.

[0169] You can combine the group interview process notes into a clear summary statement.

[0170] In order to show the team results to the interviewer.

[0171] #Skill

[0172] You need to record the free discussion in the reference source.

[0173] Integrate the group interview process into a clear summary statement.

[0174] In order to show the team results to the interviewer.

[0175] #Reference source

[0176] ##Free discussion record

[0177] {{Free discussion record insertion position}}

[0178] #limit

[0179] -Don't mention information you took out of context

[0180] -Don't mention how you work

[0181] It should be noted that, in the evaluation module of the present invention, the score generation prompt words need to include the format template for score and evaluation generation preset by the enterprise human resources manager.

[0182] In this embodiment, the scoring agent uses the following scoring prompts:

[0183] #Role

[0184] You are a scoring expert for leaderless group interviews and are able to score interviewees in leaderless group interviews based on their interview records and corresponding scoring formulas.

[0185] #Skill

[0186] You need to refer to the interviewer specified in the reference source and the interview record in the reference source.

[0187] Score the interview performance of the designated interviewer using the scoring criteria in the reference source.

[0188] #Reference source

[0189] Interviewer

[0190] {{Interviewer name insert location}}

[0191] Interview Questions

[0192] {{Interview question insert location}}

[0193] ##Interview Record

[0194] {{Interview record insert location}}

[0195] ## Scoring Criteria

[0196] {{Insert scoring criteria location}}

[0197] #Strictly output in the following format:

[0198] Your total score for the leaderless group interview: {total score calculated based on the scoring criteria}

[0199] Evaluation: {one sentence comment}

[0200] -{Specific Ability 1}:{Specific Ability 1 Scoring}:{Use dialectical thinking, carry the details of the interviewee's answer, and give examples to explain the reasons for scoring}

[0201] -{Specific Ability 2}:{Specific Ability 2 Scoring}:{Use dialectical thinking, carry the details of the interviewee's answer, and give examples to explain the reasons for scoring}

[0202] -{Specific Ability 3}:{Specific Ability 3 Scoring}:{Use dialectical thinking, carry the details of the interviewee's answer, and give examples to explain the reasons for scoring}

[0203] #limit

[0204] - The two symbols { and} do not need to be output

[0205] -Don't mention information you took out of context

[0206] -Don't mention how you work

[0207] Similarly, based on the same inventive concept, another preferred embodiment of the present invention also provides a leaderless group interview method using the leaderless group interview system provided by the above embodiment, such as Figure 2 As shown, the specific steps are as follows:

[0208] The question-setting module generates interview questions, scoring criteria and role prompts corresponding to each role agent; the interview configuration module sends the interview meeting invitation information to the interviewee's terminal and starts the leaderless group interview process; the role agents play the corresponding interview roles according to the role prompts of each role agent, and the interview module generates the final interview record of the interviewee; the interview questions and scoring criteria generated by the question-setting module are input into the scoring module to obtain the interviewee's interview score and evaluation, thus completing the leaderless group interview based on large model multi-agent.

[0209] like Figure 3As shown, it is a functional block diagram corresponding to the method of the present invention, which covers the application layer, the agent layer, the model layer and the resource layer. Among them, the application layer includes a question-setting module, an interview configuration module, an interview module and a scoring module. The question-setting module is used to generate interview questions, scoring criteria, role descriptions, role reference answer schemes and role prompts corresponding to each role agent; the interview configuration module is used to send the interview meeting invitation information to the interviewer terminal through the background management terminal. When the interviewer terminal receives the interview trigger instruction of the interviewer, the interview management terminal starts the leaderless group interview process; the interview module is used to implement the entire process of the leaderless group interview, which specifically includes the preparation stage, the independent statement stage, the free discussion stage and the concluding statement stage, and finally generates an interview record; the scoring module is used to output the interview score and evaluation, and feedback the interview score and evaluation to the interviewer. In the agent layer, the specific process covered in the question-setting module is completed by the question generation agent, the scoring standard formulation agent, the role analysis agent and the prompt word generation agent; the specific process covered in the interview module is completed by the role agent and the concluding statement agent; the specific process covered in the scoring module is completed by the scoring agent. Considering that all the above-mentioned intelligent agents are designed based on the large language model, the model layer is an AI large language model. At the resource layer, it is specifically the job requirement knowledge map, interview question bank and interview record constructed by the present invention. Figure 3 Correspondingly, the lane diagram of the method of the present invention is as follows Figure 4 shown.

[0210] The above-described embodiment is only a preferred solution of the present invention, but it is not intended to limit the present invention. A person skilled in the relevant technical field may make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, any technical solution obtained by equivalent replacement or equivalent transformation falls within the protection scope of the present invention.

Claims

1. A leaderless group interview system based on large model multi-agent, characterized by: include: The question-generating module is used to receive the interview question reference source sent by the enterprise human resources manager, and send the interview question reference source to the question-generating agent, which generates the corresponding interview question according to the interview question reference template, and then sends the interview question generated by the question-generating agent to the scoring standard-setting agent, which generates the corresponding scoring standard according to the pre-built scoring standard-setting prompt words. When the role-analyzing agent receives the interview question sent by the enterprise human resources manager and the preset number of role agents, the role-analyzing agent outputs the role description and role reference answer scheme corresponding to each role agent, and finally inputs the interview question, role description and role reference answer scheme together into the prompt word-generating agent, which outputs the role prompt words corresponding to each role agent; The interview configuration module is used to send the interview meeting invitation information to the interviewer terminal through the background management terminal. When the interviewer terminal receives the interview trigger instruction of the interviewer, the interview management terminal starts the leaderless group interview process; The interview module is used in the preparation stage of the leaderless group interview process. The interview management terminal interface displays the precautions and interview questions in the preparation stage to the interviewee. After receiving the instruction that the interviewee has finished reading within the specified time, the independent statement stage begins. The interview management terminal interface displays the notes of the independent statement stage to the interviewee. After receiving the instruction that the interviewee has finished reading, the interview management terminal records the interviewee's views within the specified time. After the statement is completed, the free discussion stage begins. The interview management terminal interface displays the notes of the free discussion stage to the interviewee. When receiving the instruction from the interviewee that the interviewee has finished reading, the interview management terminal switches to the free discussion interface and records the interviewee's behavior and voice. When receiving the instruction from the interviewee that the interviewee believes that a consensus has been reached, the interview management terminal enters the closing statement stage. The interview management terminal interface displays the notes of the closing statement stage to the interviewee. After receiving the instruction from the interviewee that the interviewee has finished reading, if the interviewee makes a closing statement within the specified time, the interview management terminal records the interviewee's closing statement and generates the final interview record. The scoring module is used to receive the interviewee's name, interview topic, final interview record and scoring criteria sent by the company's human resources manager and send them to the scoring agent, which generates prompt words based on the score to output the interview score and evaluation. Upon receiving instructions from the company's human resources manager to complete the content review and revision of the interview score and evaluation output by the scoring agent, the background management terminal automatically sends the interview score and evaluation to the interviewee.

2. A leaderless group interview system based on large model multi-agent as claimed in claim 1, characterized in that: The reference sources of interview questions are interview question template reference sources, first-category job requirement reference sources, and second-category job requirement reference sources; among them, the interview question template reference sources are the themes provided by the enterprise human resources manager; the first-category job requirement reference sources are the job requirement information matched in the knowledge graph constructed by the personal information data and interview data of employees who have passed the interview and successfully joined the company; the second-category job requirement reference sources are the job descriptions provided by the enterprise human resources manager; The scoring criteria formulation prompts need to include both the scoring criteria template preset by the company's human resources manager and the thinking chain of the scoring criteria formulation agent when generating the scoring criteria. The scoring criteria template includes the types of abilities that need to be assessed during the interview, the scoring weights of each ability type, and a specific description of each ability that needs to be assessed.

3. A leaderless group interview system based on large model multi-agent as claimed in claim 1, characterized in that: The specific process of the backend management terminal sending the interview meeting invitation information to the interviewer terminal is as follows: when the backend management terminal receives the interview question book name and the interview questions generated by the question generation agent sent by the company's human resources manager, the backend management terminal creates an interview question book entry and stores it in the database; when the backend management terminal receives the interview meeting name, interview time, interview question book and interviewer contact information sent by the company's human resources manager, the backend management terminal creates an interview meeting entry and stores it in the database, and at the same time generates the interview meeting invitation information and sends it to the interviewer terminal.

4. A leaderless group interview system based on large model multi-agent as claimed in claim 1, characterized in that: During the preparation stage of the leaderless group interview process, the interview management terminal automatically plays a video of the preparation stage precautions described by the virtual interviewer and displays it on the interview management terminal interface for the interviewee to watch. After the preparation stage precautions video is played, the interview management terminal interface displays the interview questions to the interviewee and requires the interviewee to complete reading the interview questions within the specified time. If the interview management terminal interface receives an instruction from the interviewee to complete reading the interview questions in advance, it enters the independent statement stage.

5. A leaderless group interview system based on large model multi-agent as claimed in claim 1, characterized in that: During the independent statement stage in the leaderless group interview process, the interview management terminal automatically plays a video of the precautions for the independent statement stage described by the virtual interviewer and displays it on the interview management terminal interface for the interviewee to watch. After the video of the precautions for the independent statement stage is played, the interview management terminal automatically turns on the camera and microphone to record the interviewee's behavior and voice, and requires the interviewee to complete the statement of views within the specified time. When receiving the interviewee's instruction to complete the statement of views, the interview management terminal converts the audio of the interviewee's statement of views into text to obtain an independent statement text record. If the interview management terminal interface receives the interviewee's instruction to complete the statement of views in advance, it enters the free discussion stage.

6. A leaderless group interview system based on large model multi-agent as claimed in claim 5, characterized in that: During the free discussion stage of the leaderless group interview process, the interview management terminal automatically plays a video of the precautions for the free discussion stage described by the virtual interviewer and displays it on the interview management terminal interface for the interviewee to watch. After the video of the precautions for the free discussion stage is played, the interview management terminal automatically switches to the free discussion interface and automatically turns on the camera and microphone to record the interviewee's behavior and voice. The free discussion interface consists of a floating window and multiple avatar images. The floating window plays the real-time behavior of the interviewee collected by the interview management terminal through the camera. Each avatar image shows a role agent image configured by the company's human resources manager in the question-setting module. Each role agent plays a different interview role based on its corresponding role prompt word. When the interviewer or each role agent is speaking, the frame edge color of the floating window or the frame edge color of the avatar image is highlighted. Before the end of the free discussion, if the interview management terminal interface receives an instruction from the interviewee that he or she believes that a consensus has been reached, the closing statement stage is entered.

7. A leaderless group interview system based on large model multi-agent as claimed in claim 6, characterized in that: In the free discussion stage of the leaderless group interview process, at the beginning of the free discussion stage, the interview management terminal will ask the interviewee whether he needs to speak first through a pop-up window on the interface. If the interviewee's selection instruction is not received within the specified time, it is deemed that the interviewee automatically gives up the opportunity to speak first; during the free discussion stage, each role agent speaks in a preset fixed order. At the end of each role agent's speech, the interview management terminal will ask the interviewee whether he wants to speak through a pop-up window on the interface. If the interviewer's selection instruction is not received within the specified time, it is deemed that the interviewee automatically gives up the opportunity to speak; during the free discussion stage, all role agents and interviewers each have a serial number to determine the order of each round of speeches. The interviewer's serial number is displayed below the floating window representing the interviewer, and the serial number of the role agent is displayed below the avatar picture representing the role agent. When the total number of role agents is m, the serial number of the role agent is randomly generated by the background management terminal, and the starting serial number of the role agent is 1, the ending serial number is m, and the serial number of the interviewer is m+1.

8. A leaderless group interview system based on large model multi-agent as claimed in claim 6, characterized in that: During the free discussion phase of the leaderless group interview process, when the interview management terminal turns on the camera and microphone to record the interviewee's behavior and voice, the interview management terminal sends the interviewee's behavior video file and voice file to the background management terminal, which stores the two files in the database, and uses the audio-to-text model to convert the interviewee's voice file into speech text, and stores the speech text and the interviewee's serial number as part of the free discussion record in the database; After the background management terminal sends the free discussion record to the corresponding role intelligent agent, the role intelligent agent outputs the speech text. The background management terminal stores the speech text of the role intelligent agent and the serial number of the role intelligent agent in the database as another part of the free discussion record. At the same time, the background management terminal uses the text-to-audio model to convert the speech text generated by the role intelligent agent into speech audio and sends it to the interview management terminal. The interview management terminal automatically plays the speech audio after receiving it.

9. A leaderless group interview system based on large model multi-agent as claimed in claim 8, characterized in that: During the concluding statement stage of the leaderless group interview process, the interview management terminal automatically plays a video of the precautions for the concluding statement stage described by the virtual interviewer and displays it on the interview management terminal interface for the interviewee to watch. After the concluding statement stage precautions video is played, the interview management terminal asks the interviewee whether to make a concluding statement through a pop-up window on the interface: If the interviewee receives an instruction to make a closing statement within the specified time, the interview management terminal automatically turns on the camera and microphone to record the interviewee's behavior and voice and requires the interviewee to make a closing statement within the specified time. After receiving an instruction from the interviewee that the closing statement is complete, the interview management terminal converts the audio of the interviewee's closing statement into text to obtain a closing statement text record. If the interview management terminal receives an instruction from the interviewee to complete the closing statement in advance, the interview is terminated, and the free discussion record obtained in the free discussion stage, the independent statement text record obtained in the independent statement stage, and the closing statement text record obtained in the closing statement stage together constitute the final interview record; If the interviewee's instruction to make a closing statement is not received within the specified time, it will be deemed that the interviewee automatically gives up the opportunity to make a closing statement. At this time, the free discussion record obtained in the free discussion stage and the independent statement text record obtained in the independent statement stage together constitute the final interview record. The free discussion record obtained in the free discussion stage is sent by the background management terminal to the closing statement intelligent agent, and the closing statement intelligent agent outputs the closing statement text. At the same time, the background management terminal stores the closing statement text in the database, and uses the text-to-audio model to convert the closing statement text generated by the closing statement intelligent agent into closing statement audio and send it to the interview management terminal. The interview management terminal automatically plays the closing statement audio after receiving it, and automatically ends the interview after playing.

10. A leaderless group interview method using the leaderless group interview system according to any one of claims 1 to 9, characterized in that: The specific steps are as follows: The question-generating module generates interview questions, scoring criteria, and role prompts for each role agent; The interview configuration module sends the interview meeting invitation information to the interviewer's terminal and starts the leaderless group interview process; According to the role prompt words of each role agent, the role agent plays the corresponding interview role, and the interview module generates the final interview record of the interviewee; The interview questions and scoring criteria generated by the question-setting module are input into the scoring module to obtain the interview scores and evaluations of the interviewees, thus completing the leaderless group interview based on large-model multi-agents.

Images