Generation device and generation method

Abstract

To easily perform training matched with the level of a training object person.SOLUTION: The RAG system 20 includes a receiving unit 21 that receives question information including a question related to training and information related to a user from the user, a determining unit 22 that determines a training level of the user based on the question information, a determining unit 23 that determines character information of a pseudo answerer who answers the question based on the question information and the training level of the user, a generating unit 24 that generates a prompt for instructing generation of an answer of the pseudo answerer to the question based on the character information, and an inputting unit 25 that inputs the prompt generated by the generating unit 24 to a generation AI model 31 that generates an answer based on the prompt.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] One aspect of the present disclosure relates to a generating device and a generating method. [Background technology]

[0002] Patent Document 1 discloses an interactive interview training system that uses a patient simulation device. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2016-53606 Summary of the Invention [Problem to be solved by the invention]

[0004] In the above-mentioned training system for medical interviews and the like, it is difficult to provide training that is tailored to the level of the trainee in an easy manner.

[0005] One aspect of the present disclosure has been made in consideration of the above-mentioned situation, and aims to provide a generation device and generation method that can easily perform training that is tailored to the level of the person being trained. [Means for solving the problem]

[0006] A generation device according to one aspect of the present disclosure includes a reception unit that receives question information from a user, the question information including training-related questions and information related to the user; a determination unit that determines the training level of the user based on the question information; a determination unit that determines character information of a pseudo answerer who will answer the question based on the question information and the training level of the user; a generation unit that generates a prompt to instruct the pseudo answerer to generate an answer to the question based on the character information; and a control unit that controls a generation AI model that generates answers based on the prompt generated by the generation unit.

[0007] A generation device according to one aspect of the present disclosure receives question information including information related to a question and a user, and determines the user's training level from the question information. Then, character information of a mock answerer is determined from the question information and the training level, and a prompt is generated based on the character information to instruct the mock answerer to generate an answer to the question. With this configuration, character information of the mock answerer is determined based on the training level corresponding to the user, and a prompt is generated to instruct the mock answerer to generate an answer corresponding to the character information. By inputting the prompt into a generation AI model, an answer (answer to the user's question) by the mock answerer that matches the user's training level can be appropriately generated. In this way, the mock answerer's answer to the question matches the user's training level, allowing training tailored to the level of the user who is the trainee. Furthermore, with this method, simply receiving question information from the user automatically provides the answer, facilitating training such as interviews in which questions and answers are exchanged. As described above, the generation device according to one aspect of the present disclosure allows training tailored to the level of the trainee to be easily performed. [Effects of the Invention]

[0008] According to one aspect of the present disclosure, training can be easily carried out according to the level of the trainee. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram showing the device configuration of the answer generation system according to this embodiment. [Figure 2] FIG. 2 is a diagram illustrating an example of a prompt generated by the RAG system. [Figure 3] FIG. 3 is a flowchart showing the processing executed by the RAG system. [Figure 4] FIG. 4 is a diagram illustrating an example of a hardware configuration of the RAG system. DETAILED DESCRIPTION OF THE INVENTION

[0010] The present disclosure will be described with reference to the accompanying drawings. Whenever possible, the same parts are designated by the same reference numerals and redundant description will be omitted.

[0011] FIG. 1 is a diagram illustrating the device configuration of an answer generation system according to this embodiment. The answer generation system illustrated in FIG. 1 is a system used for training trainees (users). Specifically, the answer generation system is a system that trains trainees (users) in a dialogue between the trainees (users) and a pseudo-answerer by realizing the dialogue. As an example, the answer generation system is used for training doctors, who are trainees, in their medical interviews. Specifically, the answer generation system is used to improve the doctors' medical interview skills by repeatedly generating and outputting answers from a pseudo-patient to questions from the trainees (users), who are doctors. In the following, the answer generation system is described as being used for training doctors, who are trainees (users), in their medical interviews, but is not limited to this. That is, the answer generation system may also be used for training call center employees or repairers of electronic devices and electrical appliances such as PCs, smartphones, and home appliances (training to identify the causes of malfunctions in electronic devices, etc.), etc.

[0012] 1, the answer generation system includes a terminal 10, a Retrieval-Augmented Generation (RAG) system 20, and a server device 30, which are configured to be able to communicate with each other via a network including a wireless communication network and a fixed communication network. The RAG system 20 constitutes a generation device that generates a prompt based on information received from the terminal 10. A prompt is information indicating an instruction or question input to an AI model in an interactive system such as a dialogue with an AI model or a command line interface (CLI).

[0013] The terminal 10 is a device used by a doctor who is a training subject (user) receiving training on medical interviews. The terminal 10 is, for example, a personal computer, a smartphone, a tablet terminal, a feature phone, a server device, a game console, or the like. Note that while only two terminals 10 are illustrated in FIG. 1, the answer generation system may include any number of terminals 10 greater than or equal to two.

[0014] The server device 30 stores a generative AI model 31 and enables the provision of answers (pseudo-patient answers) to a user using the generative AI model 31. The generative AI model is a model that, in response to a prompt including input information, generates content according to any one or a combination of the instructions, context, question, and output format indicated by the prompt and returns the content as response information. The prompt may also include input information, in which case the generative AI model 31 generates response information targeted at the input information. The generative AI model 31 may be, for example, an interactive AI that includes a large-scale language model (LLM) and a user interface (UI) for interacting with the user and enables text or voice chat with the user. Examples of such generative AI models include ChatGPT, GPT (registered trademark)-3.5, GPT-4V, and PaLM2. While the present embodiment describes an example in which the server device 30 provides questions and the like using one generative AI model 31, the server device 30 may also provide questions and the like using multiple generative AI models. 1 shows only one server device 30, the answer generation system may include multiple server devices 30. Although the above describes a large-scale language model as an example, other AI models may also be used. The generative AI model 31 generates a question or answer for the user based on the prompt.

[0015] The RAG system 20 is configured to include, as functional components, a reception unit 21, a determination unit 22, a determination unit 23, a generation unit 24, an input unit 25 (control unit), and a storage unit 26. The RAG system 20 inputs a prompt corresponding to input information from the terminal 10 to the server device 30, and relays response information from the server device 30 to the prompt to the terminal 10. The RAG system 20 also has a function to generate a prompt based on the input information from the terminal 10. The function of each functional unit of the RAG system 20 will be described in detail below.

[0016] The reception unit 21 receives question information from a doctor who is a training subject (user), including questions related to the training (medical interview in this case) and information related to the user. The reception unit 21 outputs the received information to the determination unit 22. The reception unit 21 receives question information from a user, for example, via the terminal 10. A question related to the medical interview is a question that a doctor asks when conducting a medical interview, such as, for example, "What symptoms did you experience today?" As described above, the answer generation system assumes that the user's questions and the pseudo answerer's answers are repeated. Therefore, questions related to the medical interview from the second time onwards (requests asked in response to the answer of the pseudo answerer) will be in line with the answer of the pseudo answerer. The information related to the user may be, for example, user attribute information (details will be described later) or simply information that uniquely identifies the user. The information related to the user may be information that the user directly inputs to the terminal 10, or information that the terminal 10 determines and generates in response to the user's question input operation.

[0017] The determination unit 22 determines the training level of the user based on the question information from the user received by the reception unit 21. The training level is information indicating the user's proficiency in the content of the training subject (here, medical interview). The determination unit 22 may determine the level in several stages, for example, level 3 (high), level 2 (normal), and level 1 (low). The determination unit 22 may also determine the level in more detail based on the score.

[0018] The determination unit 22 may determine the user's training level based on, for example, user-related information included in the question information. The determination unit 22 may determine the user's training level based on user attribute information included in the user-related information or estimated based on the user-related information. When the user-related information included in the question information includes user attribute information, the determination unit 22 determines the user's training level using the attribute information as is. Furthermore, when the user-related information includes, for example, information that uniquely identifies the user, the determination unit 22 estimates the user's attribute information based on the information that uniquely identifies the user and information stored in the storage unit 26, and determines the user's training level using the attribute information. In this case, the information stored in the storage unit 26 is information in which the information that uniquely identifies the user is linked to the user's attribute information.

[0019] The above-mentioned user attribute information may include, for example, information indicating the user's experience period or level of proficiency with respect to the training content. That is, the user attribute information may include information indicating the years of experience or level of proficiency of the doctor conducting the medical interview training. In this case, the determination unit 22 may determine, for example, that if the years of experience are 10 years or more, the user is at level 3 (high); if the years of experience are 3 years or more but less than 10 years, the user is at level 2 (normal); and if the years of experience are less than 3 years, the user is at level 1 (low).

[0020] The user's attribute information may include information indicating the user's specialty for the training content. That is, the user's attribute information may include information indicating a medical department such as "internal medicine" or "anesthesia and surgery." The user's attribute information may also include information indicating other medical levels.

[0021] The determination unit 22 may determine the training level of the user based on, for example, a question related to a medical interview included in the question information. That is, the determination unit 22 may determine the training level of the user depending on the quality of the question from the user (whether the question is asked by an experienced doctor or a novice doctor, etc.). Furthermore, the determination unit 22 may determine the training level of the user by taking into consideration both the question related to a medical interview included in the question information and information related to the user (for example, the user's attributes).

[0022] The determination unit 23 determines character information of a pseudo-answerer (pseudo-patient) who will answer the question, based on the question information received by the reception unit 21 and the training level determined by the determination unit 22. The determination unit 23 may determine the character information, for example, by taking into consideration information about the user (e.g., user attribute information) included in the question information and the training level. The determination unit 23 may determine the character information, for example, by taking into consideration the content of the question included in the question information and the training level. The character information of the pseudo-patient is information that is taken into consideration when a prompt for instructing the generation of an answer is created by the generation unit 24 (described later), and is information that determines the content of the answer of the pseudo-patient. The character information is information that links information such as the patient's attributes (gender, age), symptoms (type of illness), personality (character), etc. to the patient. In other words, the character information can be information that defines the character. That is, the character information may be information indicating the patient's speaking style, the amount of words spoken (e.g., information indicating the number of characters or phrases), information indicating the patient's personality, etc.

[0023] The determination unit 23 may determine the character information so that the higher the training level, the less talkative the character, and the lower the training level, the more talkative the character. This makes it possible to conduct interview training with pseudo-patients whose interview difficulty is high for doctors with a high training level, and to conduct interview training with pseudo-patients whose interview difficulty is low for doctors with a low training level.

[0024] The determination unit 23 may determine the symptoms (type of illness) included in the character information of the pseudo-patient based on, for example, the user's attribute information (more specifically, the doctor's years of experience, level of proficiency, etc.), and may determine the illness to be more difficult to diagnose as the level of proficiency, etc. increases. This allows for an appropriate level of medical interview training to be performed according to the user's level of proficiency, etc.

[0025] The determination unit 23 may determine character information of the pseudo-patient based on, for example, the question information received by the reception unit 21, the training level determined by the determination unit 22, and the information stored in the storage unit 26. In this case, the information stored in the storage unit 26 is information in which the question information, the training level, and the character information are linked.

[0026] The generation unit 24 generates a prompt to instruct the pseudo-answerer (pseudo-patient) to generate an answer to the question based on the character information determined by the determination unit 23. The prompt expresses, in text, for example, the command to be executed by the generation AI model 31 (dialogue AI model), the task to be executed by the generation AI model 31, the background / context (e.g., role, condition) to be considered by the generation AI model 31, the question to be answered by the generation AI model 31, and the output format of the response information from the generation AI model 31. The prompt may also include input information to be used as the command / task to be executed by the generation AI model 31. Examples of such input information include data files with file names including a predetermined extension, such as text data, image data, application-related data, audio data, video data, and still image data. Application-related data is data such as document data, table data, and graph data that can be processed by a default application program.

[0027] FIG. 2 is a diagram showing an example of a prompt generated by the RAG system 20. The prompt shown in FIG. 2 specifies a role, a task, and a condition. The role is information that specifies the role that the generative AI model 31 will output. Here, the patient who will be interviewed by the user is specified, and as an example, "You are a diabetic patient." The task is information that outlines the command content for the generative AI model 31, and here, it is specified as "Please respond to the questions from the user."

[0028] The condition is information indicating detailed conditions of the task, and in this case, it is specified as "Please do not mention the name of the disease." In this way, the generation unit 24 may generate a prompt that specifies content that prohibits the generative AI model 31 from giving a clear answer, so that the medical interview training can be carried out smoothly.

[0029] As described above, the generation unit 24 generates a prompt that defines the role of the pseudo respondent (pseudo patient) based on the character information. The generation unit 24 may also generate a prompt that defines the patient's attributes (gender, age), symptoms (type of illness), personality (character), number of words (response), etc. based on the character information.

[0030] The generation unit 24 may generate a prompt that specifies a time limit and a number of interactions according to a training level. That is, the generation unit 24 may generate a prompt that specifies that the higher the training level, the shorter the time limit and the fewer the number of interactions.

[0031] The generation unit 24 may generate prompts that specify example questions and answers for each training level. For example, the generation unit 24 may specify example answers such that the higher the training level, the more difficult the medical interview becomes, such as "Level 1 (low): How does your stomach hurt? It hurts like a needle in this part," "Level 2 (normal): How does your stomach hurt? It hurts in this part," and "Level 3 (high): How does your stomach hurt? It hurts, hurts, hurts." Such example information may be generated based on information stored in the storage unit 26. The information stored in the storage unit 26 is information that associates training levels with questions and answers.

[0032] The storage unit 26 stores information used for processing by each functional unit. The storage unit 26 may store information in which information that uniquely identifies a user is linked to the user's attribute information, as information used by the determination unit 22 when estimating the user's attribute information. The storage unit 26 may store information in which question information and a training level are linked to character information, as information used by the determination unit 23 when determining character information. The storage unit 26 may store information in which a training level is linked to a question and an answer, as information used by the generation unit 24 when generating example data for a prompt.

[0033] The input unit 25 controls the generative AI model 31 based on the prompt generated by the generation unit 24. The input unit 25 inputs the prompt to the generative AI model 31. The generative AI model 31 generates and outputs an answer of a pseudo answerer (pseudo patient) in response to the input prompt.

[0034] Next, the processing executed by the RAG system 20 will be described with reference to Fig. 3. Fig. 3 is a flowchart showing the processing executed by the RAG system.

[0035] As shown in FIG. 3, first, in the RAG system 20, question information including a question about training and information about the user is accepted from the user via the terminal 10 (step S1).

[0036] Next, the RAG system 20 determines the training level of the user based on the question information (step S2).

[0037] Next, in the RAG system 20, character information of a pseudo-answerer (pseudo-patient) who will answer the question is determined based on the question information and the training level of the user (step S3).

[0038] Next, the RAG system 20 generates a prompt for instructing the pseudo-answerer (pseudo-patient) to generate an answer to the question based on the character information (step S4).

[0039] Finally, in the RAG system 20, the generated prompt is input to the generative AI model 31 (step S5).

[0040] Next, the effects of the RAG system 20 according to this embodiment will be described.

[0041] The RAG system 20 of this embodiment includes a reception unit 21 that receives question information from a user, including questions related to training and information related to the user; a judgment unit 22 that determines the training level of the user based on the question information; a determination unit 23 that determines character information of a pseudo answerer who will answer the question based on the question information and the user's training level; a generation unit 24 that generates a prompt to instruct the pseudo answerer to generate an answer to the question based on the character information; and an input unit 25 that inputs the prompt generated by the generation unit 24 to a generation AI model 31 that generates an answer based on the prompt.

[0042] In the RAG system 20 according to this embodiment, question information including a question and information related to the user is received, and the user's training level is determined from the question information. Then, character information of a mock respondent is determined from the question information and the training level, and a prompt is generated based on the character information to instruct the mock respondent to generate an answer to the question. With this configuration, character information of the mock respondent is determined based on the training level corresponding to the user, and a prompt is generated to instruct the mock respondent to generate an answer corresponding to the character information. By inputting the prompt into the generation AI model 31, an answer (answer to the user's question) by the mock respondent that matches the user's training level can be appropriately generated. In this way, since the mock respondent's answer to the question matches the user's training level, training tailored to the level of the user (the trainee) can be performed. Furthermore, with this method, simply receiving question information from the user automatically obtains the above-described answer, making it easy to perform training such as interviews in which questions and answers are exchanged. As described above, the RAG system 20 according to this embodiment makes it easy to perform training tailored to the level of the trainee.

[0043] The determination unit 22 may determine the training level of the user based on user attribute information that is included in the information about the user or that is estimated based on the information about the user. By taking the user attribute information into consideration, the training level of the user can be determined with higher accuracy.

[0044] The user's attribute information may include information indicating the user's length of experience or level of proficiency with respect to the training content. By taking the user's length of experience or level of proficiency into consideration, the user's training level can be determined with higher accuracy.

[0045] The user's attribute information may include information indicating the user's specialty with respect to the training content. By taking the information indicating the user's specialty into consideration, the user's training level can be determined with higher accuracy.

[0046] The determination unit 23 may determine the character information so that the higher the training level, the less talkative the character, and the lower the training level, the more talkative the character. This allows a user with a high training level to undergo training in questioning patients who are difficult to answer (less talkative), and a user with a low training level to undergo training in questioning patients who are easy to answer (more talkative), so that training suited to the user's training level can be performed.

[0047] The generation unit 24 may generate a prompt that defines the role of the pseudo respondent based on the character information. This allows the generation unit 24 to appropriately generate an answer that is in line with the character information, taking into consideration the type of patient the pseudo respondent is, etc.

[0048] The generator 24 may generate prompts that define examples of questions and answers for each training level. By defining such examples, answers that correspond to the training level can be easily and appropriately generated.

[0049] The generating device and generating method of the present disclosure have the following configuration.

[0050] [1] a reception unit that receives question information from a user, the question information including a question about training and information about the user; a determination unit that determines a training level of the user based on information related to the user included in the question information; a determination unit that determines character information of a pseudo answerer for the question based on information related to the user and a training level of the user; a generating unit that generates a prompt for instructing the pseudo answerer to generate an answer based on the character information; a control unit that controls a generative AI model that generates an answer based on the prompt generated by the generation unit; A generating device comprising:

[0051] [2] The generation device described in [1], wherein the judgment unit judges the training level of the user based on attribute information of the user that is included in information related to the user or that is estimated based on information related to the user.

[0052] [3] The generation device according to [2], wherein the user's attribute information includes information indicating the user's length of experience or level of proficiency with the training content.

[0053] [4] The generation device according to [2], wherein the user's attribute information includes information indicating the user's area of ​​expertise with respect to the training content.

[0054] [5] The generation device according to any one of [1] to [4], wherein the determination unit determines the character information so that the higher the training level, the less talkative the character becomes, and the lower the training level, the more talkative the character becomes.

[0055] [6] The generation device according to any one of [1] to [5], wherein the generation unit generates the prompt that defines the role of the pseudo respondent based on the character information.

[0056] [7] The generation device according to any one of [1] to [6], wherein the generation unit generates the prompt that defines examples of questions and answers for each of the training levels.

[0057] [8] The generating device according to any one of [1] to [7], which is used for training doctors in relation to medical interviews.

[0058] [9] 1. A method for generating a prompt by a generating device, comprising: receiving question information from a user, the question information including a question about training and information about the user; determining a training level of the user based on information about the user included in the question information; determining character information of a pseudo answerer for the question based on information about the user and a training level of the user; generating a prompt for instructing the pseudo answerer to generate an answer based on the character information; inputting the generated prompts into a generative AI model that generates answers based on the prompts; A generation method including:

[0059] The block diagrams used to explain the above embodiments show functional blocks. These functional blocks (components) are realized by any combination of hardware and / or software. Furthermore, the method for realizing each functional block is not particularly limited. That is, each functional block may be realized using a single device that is physically or logically coupled, or may be realized using two or more physically or logically separated devices that are connected directly or indirectly (for example, by wire, wirelessly, etc.) and these multiple devices. The functional block may also be realized by combining the single device or multiple devices with software.

[0060] Functions include, but are not limited to, judgment, determination, judgment, calculation, computation, processing, derivation, investigation, search, confirmation, reception, transmission, output, access, resolution, selection, election, establishment, comparison, assumption, expectation, consideration, broadcasting, notifying, communicating, forwarding, configuring, reconfiguring, allocation, mapping, and assignment. For example, a functional block (component) that performs transmission is called a transmitting unit or transmitter. As mentioned above, there are no particular limitations on how these functions are implemented.

[0061] For example, the RAG system 20 constituting the answer generation system according to an embodiment of the present disclosure may function as a computer that performs processing of the control method of the present disclosure. FIG. 4 is a diagram illustrating an example of the hardware configuration of the RAG system 20 according to this embodiment. The RAG system 20 described above may be physically configured as a computer device including a processor 1001, a memory 1002, a storage device 1003, a communication device 1004, an input device 1005, an output device 1006, a bus 1007, and the like. Note that the RAG system 20 may be configured as a computer device including at least one processor such as a CPU or GPU, or may be configured as a computer device including multiple processors or may be configured to include multiple computer devices. The terminal 10 and the server device 30 may also have a similar hardware configuration.

[0062] In the following description, the term "apparatus" can be interpreted as a circuit, a device, a unit, etc. The hardware configuration of the RAG system 20 may be configured to include one or more of the apparatuses shown in the figure, or may be configured to exclude some of the apparatuses.

[0063] Each function in the RAG system 20 is realized by loading specified software (programs) onto hardware such as the processor 1001 and memory 1002, causing the processor 1001 to perform calculations, control communication via the communication device 1004, and control at least one of reading and writing data in the memory 1002 and storage 1003.

[0064] The processor 1001 controls the entire computer by running, for example, an operating system. The processor 1001 may be configured by a central processing unit (CPU) including an interface with peripheral devices, a control device, an arithmetic unit, a register, etc. For example, the above-mentioned reception unit 21, judgment unit 22, decision unit 23, generation unit 24, input unit 25, etc. may be realized by the processor 1001.

[0065] The processor 1001 also loads programs (program codes), software modules, data, etc. from at least one of the storage 1003 and the communication device 1004 into the memory 1002 and executes various processes in accordance with the programs. The programs used are those that cause a computer to execute at least some of the operations described in the above-described embodiments. For example, the reception unit 21, the determination unit 22, the decision unit 23, the generation unit 24, and the input unit 25 may be implemented by a control program stored in the memory 1002 and running on the processor 1001, and similar implementations may be made for other functional blocks. While the above-described various processes have been described as being executed by one processor 1001, they may also be executed simultaneously or sequentially by two or more processors 1001. The processor 1001 may be implemented by one or more chips. The programs may also be transmitted from a network via a telecommunications line.

[0066] The memory 1002 is a computer-readable recording medium and may be configured, for example, by at least one of a read-only memory (ROM), an erasable programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a random access memory (RAM), etc. The memory 1002 may also be called a register, a cache, a main memory (primary storage device), etc. The memory 1002 can store executable programs (program codes), software modules, etc. for implementing a control method according to an embodiment of the present disclosure.

[0067] Storage 1003 is a computer-readable recording medium, and may be, for example, at least one of an optical disk such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (e.g., a compact disk, a digital versatile disk, a Blu-ray disc), a smart card, a flash memory (e.g., a card, a stick, a key drive), a floppy disk, a magnetic strip, etc. Storage 1003 may also be referred to as an auxiliary storage device. The above-mentioned storage medium may be, for example, a database, a server, or other appropriate medium including at least one of memory 1002 and storage 1003.

[0068] The communication device 1004 is hardware (transmission / reception device) for communicating between computers via at least one of a wired network and a wireless network, and is also referred to as, for example, a network device, a network controller, a network card, or a communication module. The communication device 1004 may be configured to include a high-frequency switch, a duplexer, a filter, a frequency synthesizer, etc. to realize at least one of frequency division duplex (FDD) and time division duplex (TDD). For example, the above-mentioned reception unit 21, input unit 25, etc. may be realized by the communication device 1004.

[0069] The input device 1005 is an input device (for example, a keyboard, a mouse, a microphone, a switch, a button, a sensor, etc.) that receives input from the outside. The output device 1006 is an output device (for example, a display, a speaker, an LED lamp, etc.) that outputs to the outside. The input device 1005 and the output device 1006 may be integrated into one device (for example, a touch panel).

[0070] Furthermore, each device, such as the processor 1001 and the memory 1002, is connected by a bus 1007 for communicating information. The bus 1007 may be configured using a single bus, or may be configured using different buses between each device.

[0071] Furthermore, RAG system 20 may be configured to include hardware such as a microprocessor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a programmable logic device (PLD), or a field programmable gate array (FPGA), and some or all of the functional blocks may be realized by the hardware. For example, processor 1001 may be implemented using at least one of these pieces of hardware.

[0072] The notification of information is not limited to the aspects / embodiments described in the present disclosure and may be performed using other methods. For example, the notification of information may be performed by physical layer signaling (e.g., Downlink Control Information (DCI), Uplink Control Information (UCI)), higher layer signaling (e.g., Radio Resource Control (RRC) signaling, Medium Access Control (MAC) signaling, broadcast information (Master Information Block (MIB), System Information Block (SIB))), other signals, or a combination thereof. Furthermore, the RRC signaling may be referred to as an RRC message, and may be, for example, an RRC Connection Setup message, an RRC Connection Reconfiguration message, or the like.

[0073] The order of the procedures, sequences, flowcharts, etc. of each aspect / embodiment described in this disclosure may be changed unless it is consistent. For example, the methods described in this disclosure present elements of various steps using an example order, and are not limited to the particular order presented.

[0074] Input and output information may be stored in a specific location (for example, memory) or may be managed using a management table. Input and output information may be overwritten, updated, or added to. Output information may be deleted. Input information may be sent to another device.

[0075] The determination may be made based on a value represented by one bit (0 or 1), a Boolean value (true or false), or a numerical comparison (e.g., comparison with a predetermined value).

[0076] Each aspect / embodiment described in this disclosure may be used alone, in combination, or switched depending on the implementation. Furthermore, notification of predetermined information (e.g., notification that "X is true") is not limited to being done explicitly, but may be done implicitly (e.g., by not notifying the predetermined information).

[0077] Although the present disclosure has been described in detail above, it is clear to those skilled in the art that the present disclosure is not limited to the embodiments described herein. The present disclosure can be implemented in modified and altered forms without departing from the spirit and scope of the present disclosure as defined by the claims. Therefore, the description of the present disclosure is intended to be illustrative and does not have any limiting meaning on the present disclosure.

[0078] Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, threads of execution, procedures, functions, etc., whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise.

[0079] Software, instructions, information, etc. may also be transmitted or received over a transmission medium. For example, if software is transmitted from a website, server, or other remote source using wired technologies (such as coaxial cable, fiber optic cable, twisted pair, Digital Subscriber Line (DSL)), and / or wireless technologies (such as infrared, microwave), these wired and / or wireless technologies are included within the definition of transmission media.

[0080] The information, signals, etc. described in this disclosure may be represented using any of a variety of different technologies. For example, data, instructions, commands, information, signals, bits, symbols, chips, etc. that may be referred to throughout the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or magnetic particles, optical fields or photons, or any combination thereof.

[0081] Note that terms explained in this disclosure and terms necessary for understanding this disclosure may be replaced with terms having the same or similar meanings. For example, at least one of a channel and a symbol may be a signal (signaling). Furthermore, a signal may be a message. Furthermore, a component carrier (CC) may be called a carrier frequency, a cell, a frequency carrier, etc.

[0082] Furthermore, the information, parameters, etc. described in the present disclosure may be expressed using absolute values, relative values ​​from a predetermined value, or other corresponding information. For example, a radio resource may be indicated by an index.

[0083] The names used for the above-described parameters are not intended to be limiting in any way. Furthermore, the mathematical expressions using these parameters may differ from those explicitly disclosed in this disclosure. The various channels (e.g., PUCCH, PDCCH, etc.) and information elements may be identified by any suitable names, and therefore the various names assigned to these various channels and information elements are not intended to be limiting in any way.

[0084] In this disclosure, the terms "Mobile Station (MS)," "user terminal," "User Equipment (UE)," "terminal," etc. may be used interchangeably.

[0085] A mobile station may also be referred to by those skilled in the art as a subscriber station, mobile unit, subscriber unit, wireless unit, remote unit, mobile device, wireless device, wireless communication device, remote device, mobile subscriber station, access terminal, mobile terminal, wireless terminal, remote terminal, handset, user agent, mobile client, client, or some other suitable terminology.

[0086] As used in this disclosure, the terms "determining" and "determining" may encompass a wide variety of actions. "Determining" and "determining" may include, for example, judging, calculating, computing, processing, deriving, investigating, looking up, searching, inquiring (e.g., searching in a table, database, or other data structure), ascertaining, and the like. "Determining" and "determining" may also include receiving (e.g., receiving information), transmitting (e.g., sending information), input, output, accessing (e.g., accessing data in memory), and the like. Furthermore, "judgment" and "decision" can include regarding resolving, selecting, choosing, establishing, comparing, etc. as having been "judged" or "decided." In other words, "judgment" and "decision" can include regarding some action as having been "judged" or "decided." Furthermore, "judgment (decision)" can be interpreted as "assuming," "expecting," "considering," etc.

[0087] The terms "connected," "coupled," or any variation thereof, refer to any direct or indirect connection or coupling between two or more elements, and may include the presence of one or more intermediate elements between two elements that are "connected" or "coupled" to each other. The coupling or connection between elements may be physical, logical, or a combination thereof. For example, "connected" may be read as "access." As used in this disclosure, two elements may be considered to be "connected" or "coupled" to each other using one or more wires, cables, and / or printed electrical connections, as well as electromagnetic energy having wavelengths in the radio frequency range, microwave range, and optical (both visible and invisible) range, as some non-limiting and non-exhaustive examples.

[0088] As used in this disclosure, the phrase "based on" does not mean "based only on," unless expressly stated otherwise. In other words, the phrase "based on" means both "based only on" and "based at least on."

[0089] Any reference to an element using a designation such as "first," "second," etc., used in this disclosure does not generally limit the quantity or order of those elements. These designations may be used in this disclosure as a convenient way to distinguish between two or more elements. Thus, a reference to a first and a second element does not imply that only two elements may be employed or that the first element must in some way precede the second element.

[0090] When used in this disclosure, the terms "include," "including," and variations thereof are intended to be inclusive, similar to the term "comprising." Furthermore, when used in this disclosure, the term "or" is not intended to be an exclusive or.

[0091] In this disclosure, where articles are added by translation, such as a, an, and the in English, the disclosure may include that the nouns following these articles are in the plural form.

[0092] In the present disclosure, the term "A and B are different" may mean "A and B are different from each other." The term may also mean "A and B are each different from C." Terms such as "separate" and "coupled" may also be interpreted in the same way as "different." [Explanation of symbols]

[0093] 20...RAG system (generation device), 21...reception unit, 22...judgment unit, 23...decision unit, 24...generation unit, 25...input unit, 31...generative AI model.

Claims

1. a reception unit that receives question information from a user, the question information including a question about training and information about the user; a determination unit that determines a training level of the user based on information related to the user included in the question information; a determination unit that determines character information of a pseudo answerer for the question based on information related to the user and a training level of the user; a generating unit that generates a prompt for instructing the pseudo answerer to generate an answer based on the character information; A control unit controls a generative AI model that generates an answer based on the prompt generated by the generation unit; A generating device comprising:

2. The generation device according to claim 1 , wherein the determination unit determines the training level of the user based on attribute information of the user that is included in the information about the user or that is estimated based on the information about the user.

3. The generating device according to claim 2 , wherein the user attribute information includes information indicating the user's length of experience or level of proficiency with respect to the training content.

4. The generating device according to claim 2 , wherein the user attribute information includes information indicating the user's specialty field with respect to the training content.

5. The generating device according to claim 1 , wherein the determining unit determines the character information such that the higher the training level, the less talkative the character is, and the lower the training level, the more talkative the character is.

6. The generating device according to claim 1 , wherein the generating unit generates the prompt that defines a role of the pseudo respondent based on the character information.

7. The generating device according to claim 1 , wherein the generating unit generates the prompts that define examples of questions and answers for each of the training levels.

8. The generating device according to any one of claims 1 to 7, which is used for training doctors in relation to medical interviews.

9. 1. A method for generating a prompt by a generating device, comprising: receiving question information from a user, the question information including a question about training and information about the user; determining a training level of the user based on information about the user included in the question information; determining character information of a pseudo answerer for the question based on information about the user and a training level of the user; generating a prompt for instructing the pseudo answerer to generate an answer based on the character information; inputting the generated prompts into a generative AI model that generates answers based on the prompts; A generation method including:

Images