Team dialogue generation and tutoring learning method and system based on large language model

Abstract

This invention discloses a team-based dialogue generation and tutoring learning method and system based on a large language model. First, a rule-based expert course knowledge base is constructed according to course needs and knowledge structure. Second, in an observational learning context, personalized knowledge dialogues are generated between different teachers and a fixed set of students based on a generative teacher prompt instruction set, and personalized knowledge dialogues are generated between a fixed teacher and different students based on a generative student prompt instruction set. Then, in an interactive learning context, teaching statements from expert teachers are generated based on a generative interactive prompt instruction set, and personalized feedback statements from expert teachers are generated based on student feedback and the generative interactive feedback prompt instruction set. Finally, based on student user experience feedback, knowledge learning effectiveness is analyzed, and precise team-based learning contexts are optimized to achieve personalized tutoring for students.

Description

Technical Field

[0001] This invention belongs to the field of large language model technology, specifically involving a team-based dialogue generation and personalized tutoring learning method and system based on large language model. The aim is to use the prompt word strategy of large language model to design generative teachers and generative students, optimize dialogue generation, initiation and feedback technology, help students to conduct intelligent learning, and promote the development of intelligent tutoring system towards personalization and intelligence. Background Technology

[0002] Intelligent Tutoring Systems (ITS) represent a new paradigm of personalized education. By simulating the role of a professional tutor and utilizing intelligent technology, they provide students with customized content and feedback, achieving a high degree of personalization in learning. While ITS offers significant advantages in enhancing learning outcomes, traditional ITS suffers from isolated learning, mechanical interaction, and a lack of variety in its models, making it difficult to meet the needs of intelligent teaching or support higher-order thinking activities. With the unleashing of enormous potential in language learning, mathematical problem-solving, and psychological counseling by large language models such as Sora, GPT-4o, and iFlytek Spark, GPT-based intelligent tutoring applications are gaining increasing attention. Therefore, how to leverage GPT to empower the future development of intelligent tutoring systems has become a topic of widespread interest.

[0003] GPT's anthropomorphic interactive advantages are expected to strengthen expert teachers and interactive dialogue components, enhance the naturalness of human-computer interaction in intelligent tutoring systems, and accelerate their integration into classroom teaching applications. Documents such as the "New Generation Artificial Intelligence Development Plan" propose to make full use of artificial intelligence and other technologies to build a new education system of intelligent learning and interactive learning. Currently, there are already large language model education applications such as Spark Language Partner, EmoGPT, and MathGPT. These follow the design methods of traditional intelligent tutoring systems, emphasizing the personalized generation of knowledge, which is easier for students to accept and use. However, these applications still face the following problems: (1) the dialogue quality is not high, and there is a lack of domain knowledge from expert teachers; (2) there is a lack of educational theory guidance, which may make it difficult to achieve ideal tutoring results; (3) the prompting technology is singular, and students are still in a one-on-one isolated learning situation.

[0004] In summary, while large language models enhance the practicality of intelligent tutoring systems, generative dialogues without prompts and validation may not necessarily improve student learning outcomes. Currently, it is necessary to explore teaching models adapted to intelligent tutoring to foster deeper interaction in the classroom. This requires reconstructing intelligent tutoring systems by combining design, theory, and practice to provide students with an effective learning experience. Therefore, this invention, based on the current state of intelligent tutoring system development, designs a team-based dialogue generation and personalized tutoring learning method and system based on large language models. This integrates student observation and interactive learning, providing technical support for students' self-directed and personalized learning. Summary of the Invention

[0005] This invention addresses the problems of rigid dialogue patterns, mechanical interactions, and limited prompting technology in current intelligent tutoring systems. Starting with a large language model, it generates personalized knowledge dialogues, teaching statements, and personalized feedback statements to achieve the teaching objectives of precise tutoring, continuous student support, and active guidance. This invention also provides a tutoring and learning system based on a large language model, providing technical support for precise tutoring of students' course learning.

[0006] This invention provides a team-based dialogue generation and tutoring learning method based on a large language model, comprising the following steps:

[0007] Step 1: Based on the course requirements and knowledge structure, construct a rule-based expert course knowledge base;

[0008] Step 2: Enter the observational learning context. Based on the generative teacher prompt instruction set, generate personalized knowledge dialogue A between different teachers and fixed students. Based on the generative student prompt instruction set, generate personalized knowledge dialogue B between fixed teachers and different students. The personalized knowledge dialogue A and personalized knowledge dialogue B are used to assist students in understanding knowledge and accompany students in learning knowledge, respectively.

[0009] Step 3: Enter the interactive learning context, initiate a set of prompts based on generative interaction, and generate teaching statements C from expert teachers. These teaching statements C are used to induce students to participate in knowledge interaction.

[0010] Step 4: Enter the interactive learning context. Based on the students' feedback and the generative interactive feedback prompt instruction set, generate personalized feedback statements D from expert teachers to respond to the students' knowledge needs.

[0011] Step 5: Repeat steps 2, 3, and 4 to optimize personalized knowledge dialogue A, personalized knowledge dialogue B, teaching statements C, and personalized feedback statements D until the student no longer repeatedly enters steps 2, 3, and 4. At this point, it is considered that the student has mastered the course knowledge, thus achieving intelligent learning guidance for the student.

[0012] Furthermore, in step 1, constructing the rule-based expert course knowledge base includes:

[0013] Organize course knowledge concepts, key points, difficulties, and case studies according to the overall knowledge representation rules;

[0014] Based on the rule of progressively deepening knowledge from simple to complex, organize the connections between key and difficult knowledge points. These connections include sequential and complementary relationships.

[0015] The course knowledge concepts, key points and difficulties, case studies, and connections between key points and difficulties are compiled to obtain an expert course knowledge base.

[0016] Furthermore, the steps for generating personalized knowledge dialogue A and personalized knowledge dialogue B in step 2 are as follows:

[0017] (21) Call the large language model and configure the generative teacher prompt instruction set environment and the generative student prompt instruction set environment;

[0018] (22) Obtain the knowledge content that students need to supplement, and obtain the corresponding knowledge content in the expert course knowledge base;

[0019] (23) Determine the generative teacher prompt instruction set and the generative student prompt instruction set;

[0020] (24) Input the generative teacher prompting instruction set environment and the generative student prompting instruction set environment, knowledge content, generative teacher prompting instruction set and generative student prompting instruction set into the large language model for natural language processing to generate personalized knowledge dialogue A and personalized knowledge dialogue B.

[0021] Furthermore, the generative teacher prompting instruction set includes:

[0022] (a1) Construct generative teachers T = {t1, t2, t3} and random generative students s, where t1, t2, and t3 are active teachers, constructive teachers, and interactive teachers, respectively;

[0023] (a2) The prompting instructions of the proactive teacher t1 are to directly tell the randomly generated students s what knowledge point K is, what the principle of the knowledge point is, and how to apply the knowledge point, so as to stimulate the feedback of the randomly generated students s; the prompting instructions of the constructive teacher t2 are to tell the randomly generated students s knowledge points, principles and applications through analogy and metaphor, so as to stimulate the randomly generated students s to connect new information with existing knowledge; the prompting instructions of the interactive teacher t3 are to guide the randomly generated students s through cases and questions, and encourage the randomly generated students s to construct their understanding of knowledge points and participate in classroom activities;

[0024] (a3) Input the constructed generative teacher T into the large language model to obtain the classroom speech of the generative teacher T;

[0025] (a4) Determine the prompting instructions P(s) of the randomly generated student s, where P(s) is the response of the generated teacher T to the generated teacher T in the role of an ordinary student, based on the classroom speech of the generated teacher T.

[0026] (a5) The prompting instructions of the constructed generative teacher T = {t1, t2, t3} and the randomly generated student s are the generative teacher prompting instruction set.

[0027] Furthermore, the generative student prompt instruction set includes:

[0028] (b1) Construct a randomized teacher t and a generative student S = {s1, s2}, where s1 and s2 are expert students and novice students, respectively;

[0029] (b2) Determine the prompting instructions P(t) of the randomly generated teacher t, where P(t) is the classroom dialogue conducted by the expert teacher around the knowledge point "pointer";

[0030] (b3) Input the constructed randomized generative teacher t into the large language model to obtain the classroom speech of generative teacher T;

[0031] (b4) Determine the prompting instructions for the generative student S. The prompting instructions for the expert student S1 are to correctly answer and supplement the knowledge points taught by the random generative teacher t in the role of an expert student, and to ask questions to the random generative teacher t and apply the knowledge in other ways, based on the classroom language of the random generative teacher t. The prompting instructions for the novice student S2 are to incorrectly answer or refuse to answer the knowledge points taught by the random generative teacher t in the role of a novice student, and to ask questions to the random generative teacher t and seek help.

[0032] (b5) The prompting instructions of the constructed random generative teacher t and generative student S = {s1, s2} are the generative student prompting instruction set.

[0033] Furthermore, the steps for generating the teaching statement C in step 3 are as follows:

[0034] (31) Call the large language model and configure the generative interactive prompt instruction set environment;

[0035] (32) Obtain the knowledge content that students need to supplement, and obtain the corresponding knowledge content in the expert course knowledge base;

[0036] (33) Determine the generative interaction initiation prompt instruction set, the prompt instruction set including {P1(T Q ), P2(T Q)};P1(T Q P2(T) is when students first enter an interactive learning environment, the expert teacher asks the students if they need learning assistance; Q When students are not in an interactive learning context for the first time, an expert teacher will pose application questions and set up cases based on the knowledge content that students choose to supplement, the corresponding knowledge content in the expert course knowledge base, the previous teaching statement C, and the students' responses. The questions will be in the form of closed-ended questions and open-ended questions, with the difficulty of the questions ranging from simple to difficult.

[0037] (34) Input the generative interaction initiation prompt instruction set environment, knowledge content, and generative interaction initiation prompt instruction set into the large language model for natural language processing to generate teaching statement C.

[0038] Furthermore, the steps for generating the personalized feedback statement D in step 4 are as follows:

[0039] (41) Call the large language model and configure the generative interactive feedback prompt instruction set environment;

[0040] (42) Obtain the knowledge content that students choose to supplement, the teaching statements C, and the students' responses Sr;

[0041] (43) Determine the expert Q&A teacher T QA A generative interactive feedback prompt instruction set, the generative interactive feedback prompt instruction set including {P1(T A ), P2(T A ), P3(T A ), P4(T A )},P1(T A When a student's response Sr is non-empty and related to the knowledge content the student selected that requires assistance, the expert teacher provides feedback to the student based on their response Sr. This feedback can take the form of judging the correctness of the answer, offering encouragement or praise, or providing detailed explanations and hints. P2(T) A When a student's response Sr is not empty and is unrelated to the knowledge content the student selected that requires assistance, the system prompts the student to answer based on the selected knowledge content or to modify the selected knowledge content; P3(T) A P4 (T) is used when a student does not respond within a short period of time, and the expert teacher inquires whether the student needs help; A This means that if a student does not respond for an extended period of time, they will exit the interactive learning environment.

[0042] (44) Input the generative interactive feedback prompt instruction set environment, knowledge content, and generative interactive feedback prompt instruction set into the large language model for natural language processing to generate personalized feedback statement D;

[0043] (45) When the generative interactive feedback prompt instruction set P2(T) A When activated, repeat steps (2) to (4) until the student exits the interactive learning context.

[0044] Furthermore, the large language model includes Chatglm-4 and Chatglm-3-turbo.

[0045] This invention also provides a team-based dialogue generation and tutoring learning system based on a large language model, comprising the following modules:

[0046] The login and registration module is used to record student information so that the system can track students' learning progress and needs;

[0047] The expert course knowledge base module is used to store knowledge points, teaching resources and learning materials of various disciplines, so that students can access them at any time and build them according to the needs of the course and the knowledge structure of the course.

[0048] The self-observation learning module is used to generate personalized knowledge dialogues A and B based on generative teacher prompts and student prompts, helping students understand and master knowledge through observation.

[0049] The self-directed interactive learning module provides teaching statements C based on the generative interactive prompt instruction set, inducing students to participate in knowledge interaction; based on student feedback and the generative interactive feedback prompt instruction set, it generates personalized feedback statements D from expert teachers to respond to students' knowledge needs. Through interactive learning, students deepen their understanding and application of knowledge.

[0050] Compared with existing research and technology, this invention has the following advantages:

[0051] 1. This invention proposes a course knowledge base construction technology based on a large language model. By establishing connections between course knowledge points through knowledge representation rules and progressive in-depth rules, the error rate of knowledge generation during intelligent tutoring is reduced, and the understanding and retrieval of course knowledge by the large language model is improved.

[0052] 2. This invention combines educational and psychological theories with large language model technology to establish a personalized knowledge dialogue driven by a large language model, realizing the entire process of personalized knowledge tutoring. It meets students' observation and learning needs for course knowledge through a team-based learning environment, laying the foundation for students' efficient and seamless learning.

[0053] 3. This invention designs an interactive dialogue mode that integrates teaching statements and personalized feedback statements, adaptively adjusting different generation modes to support students' personalized intelligent learning needs. Attached Figure Description

[0054] Figure 1 A flowchart for a team-based dialogue generation and personalized tutoring learning method based on a large language model;

[0055] Figure 2 A diagram for generating and guiding learning in personalized knowledge dialogue A;

[0056] Figure 3 A diagram for generating and guiding learning in personalized knowledge dialogues (B);

[0057] Figure 4 A diagram illustrating the generation and guidance of instructional statement C and personalized feedback statement D;

[0058] Figure 5 A diagram illustrating the learning outcomes of personalized tutoring based on a large language model for students.

[0059] Figure 6 This is a feedback graph for personalized tutoring learning based on a large language model.

[0060] Figure 7 This is a diagram of a personalized tutoring learning system based on a large language model. Detailed Implementation

[0061] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings.

[0062] This invention provides a method and system for team-based dialogue generation and personalized tutoring learning based on a large language model. Figure 1 This demonstrates the problems solved and the core ideas behind these steps. First, the large language model considered in this embodiment of the invention will be configured as an observational learning context and an interactive learning context, and constructed as a generative teacher T = {t1, t2, t3}, a randomly generated student s, a randomly generated teacher t, generative students S = {s1, s2}, and an expert tutor T. QA The system plays multiple roles, aiming to provide students with personalized knowledge dialogues (A), personalized knowledge dialogues (B), instructional statements (C), and personalized feedback statements (D) that align with the teaching content and basic teaching principles, prompting it to pay attention to the details of learning guidance during output. Specifically, it includes the following steps:

[0063] Step 1: Based on the course requirements and knowledge structure, construct a rule-based expert course knowledge base;

[0064] Step 2: Enter the observational learning context. Based on the generative teacher prompt instruction set, generate personalized knowledge dialogue A between different teachers and fixed students. Based on the generative student prompt instruction set, generate personalized knowledge dialogue B between fixed teachers and different students. The personalized knowledge dialogue A and personalized knowledge dialogue B are used to assist students in understanding knowledge and accompany students in learning knowledge, respectively.

[0065] Step 3: Enter the interactive learning context, initiate a set of prompts based on generative interaction, and generate teaching statements C from expert teachers. These teaching statements C are used to induce students to participate in knowledge interaction.

[0066] Step 4: Enter the interactive learning context. Based on the students' feedback and the generative interactive feedback prompt instruction set, generate personalized feedback statements D from expert teachers to respond to the students' knowledge needs.

[0067] Step 5: Repeat steps 2, 3, and 4 to optimize personalized knowledge dialogue A, personalized knowledge dialogue B, teaching statements C, and personalized feedback statements D until the student no longer repeatedly enters steps 2, 3, and 4. At this point, it is considered that the student has mastered the course knowledge, thus achieving intelligent learning guidance for the student.

[0068] In addition to the methods described above, we will test the students on the knowledge points they have learned. Once students believe they have mastered the knowledge, steps 2, 3, and 4 will not be repeated. Therefore, the way to judge whether a student has mastered the knowledge is to observe whether the student repeats steps 2, 3, and 4. If they repeat, they have not mastered the knowledge and are still learning; if they do not repeat, they have mastered the knowledge and will stop learning.

[0069] To achieve the above objectives, taking the university course "C Programming Language" as an example, the specific steps for constructing an expert course knowledge base include:

[0070] (1) Based on the understanding of the course knowledge structure, and taking Tan Haoqiang's "C Programming (Fifth Edition)" as the standard, the C language knowledge is divided into nine major units: introduction to algorithms, sequential structure, selection structure, loop structure, array, function, pointer, structure and file. The key points and cases of knowledge in each major unit are determined in turn to realize the overall integration of course knowledge. Table 1 is the overall integration representation of the knowledge point of "pointer".

[0071] Table 1 shows the overall understanding of the "Pointer" knowledge points.

[0072]

[0073]

[0074] (2) Within each major unit, the connection relationships of key knowledge points are established according to the rule of progressing from simple to complex. For example, the key knowledge points of "pointers" include: the basic concept of pointers, pointer declaration, pointer operations (address-of operator, dereference operator, pointer arithmetic operations), pointers and arrays, and pointers and functions. The connection relationships include sequential relationships and complementary relationships. In the example of "pointers", the sequential relationship is that the key knowledge points are from left to right, and the learning difficulty increases sequentially. The complementary relationship exists between pointer operations and pointers and arrays, because pointer arithmetic operations can be used to traverse arrays, and the concepts of pointers and arrays are closely related and can be alternated during learning.

[0075] Furthermore, taking the concept of "pointers" from the university course "C Programming Language" as an example, a personalized knowledge dialogue A is generated by entering an observational learning context, such as... Figure 2 As shown, the specific steps are as follows:

[0076] (1) Call large language models (such as Chatglm-4, Chatglm-3-turbo, etc.), configure the generative teacher prompt instruction set environment, and set the system information to "You are a setup assistant in an intelligent tutoring system, and your task is to provide intelligent learning tutoring for users";

[0077] (2) Obtain the knowledge content that students need to supplement, and obtain the corresponding knowledge content in the expert course knowledge base;

[0078] (3) Determine the generative teacher prompting instruction set;

[0079] (4) Input the generative teacher prompt instruction set environment, the knowledge content, and the generative teacher prompt instruction set into the large language model for natural language processing to generate personalized knowledge dialogue A;

[0080] (5) If the content does not meet the students’ learning needs, repeat step (4) until a satisfactory personalized knowledge dialogue result is generated.

[0081] Furthermore, the generative teacher prompting instruction set includes:

[0082] (1) Construct generative teachers T = {t1, t2, t3} and random generative students s, where t1, t2, and t3 are active teachers, constructive teachers, and interactive teachers, respectively;

[0083] (2) The prompting instructions of the proactive teacher t1 are to directly tell the randomly generated students s what knowledge point K is, what the principle of the knowledge point "pointer" is, and how to apply the knowledge point "pointer", so as to stimulate the feedback of the randomly generated students s; the prompting instructions of the constructivist teacher t2 are to tell the randomly generated students s the knowledge point "pointer", principle and application through analogy and metaphor, so as to stimulate the randomly generated students s to connect the new information with the original knowledge in a meaningful way; the prompting instructions of the interactive teacher t3 are to guide the randomly generated students s through cases and questions, and encourage the randomly generated students s to construct their own understanding of the knowledge point "pointer" and participate in classroom activities;

[0084] (3) Input the constructed generative teacher T into the large language model to obtain the classroom speech of the generative teacher T;

[0085] (4) Determine the prompting instructions P(s) of the randomly generated student s. P(s) is the response of the generated teacher T to the generated teacher T in the role of an ordinary student, based on the classroom language of the generated teacher T.

[0086] (5) The prompting instructions of the constructed generative teacher T = {t1, t2, t3} and the randomly generated student s are the generative teacher prompting instruction set.

[0087] Furthermore, taking the concept of "pointers" from the university course "C Programming Language" as an example, we enter an observational learning context and generate personalized knowledge dialogues (B), such as... Figure 3 As shown, the specific steps are as follows:

[0088] (1) Call large language models (such as Chatglm-4, Chatglm-3-turbo, etc.), configure the generative student prompt instruction set environment, and set the system information to "You are a setup assistant in an intelligent tutoring system, and your task is to provide intelligent learning tutoring for users";

[0089] (2) Obtain the knowledge content that students need to supplement, and obtain the corresponding knowledge content in the expert course knowledge base;

[0090] (3) Determine the generative student prompt instruction set;

[0091] (4) Input the generative student prompt instruction set environment, the knowledge content, and the generative student prompt instruction set into the large language model for natural language processing to generate personalized knowledge dialogue B;

[0092] (5) If the content does not meet the students’ learning needs, repeat step (4) until a satisfactory personalized knowledge dialogue result is generated.

[0093] Furthermore, the generative student prompt instruction set includes:

[0094] (1) Construct a randomized teacher t and a generative student S = {s1, s2}, where s1 and s2 are expert students and novice students, respectively;

[0095] (2) Determine the prompting instructions P(t) of the randomly generated teacher t. P(t) is the classroom dialogue carried out by the expert teacher around the knowledge point "pointer".

[0096] (3) Input the constructed randomized teacher t into the large language model to obtain the classroom speech of the generative teacher T;

[0097] (4) Determine the prompting instructions for the generative student S. The prompting instructions for the expert student s1 are to correctly answer and supplement the knowledge point "pointers" taught by the random generative teacher t based on the classroom language of the random generative teacher t, and to ask questions to the random generative teacher t and apply the knowledge in a general way. The prompting instructions for the novice student s2 are to incorrectly answer or refuse to answer the knowledge point "pointers" taught by the random generative teacher t based on the classroom language of the random generative teacher t, and to like to ask questions to the random generative teacher t and seek help.

[0098] (5) The prompting instructions of the constructed random generating teacher t and generating student S = {s1, s2} are the prompting instruction set of generating student.

[0099] Furthermore, taking the concept of "pointers" from the university course "C Programming Language" as an example, an interactive learning environment is created to generate teaching statements in C, such as... Figure 4 As shown, the specific steps are as follows:

[0100] (1) Call a large language model (such as Chatglm-4, Chatglm-3-turbo, etc.), configure a generative interactive prompt instruction set environment, and set the system information to "You are a setup assistant in an intelligent tutoring system. Your task is to provide interactive tutoring for users. Now you need to construct an expert Q&A teacher T around the knowledge point 'pointer'". QA "Ask students questions or show them knowledge examples to promote their in-depth understanding";

[0101] (2) Obtain the knowledge content that students need to supplement, and obtain the corresponding knowledge content in the expert course knowledge base;

[0102] (3) Determine the generative interaction initiation prompt instruction set, wherein the prompt instruction set includes {P1(T Q ), P2(T Q )};P1(T Q When students first enter an interactive learning environment, the expert teacher should ask them if they need learning assistance; P2(T)Q When students are not in an interactive learning context for the first time, the expert teacher needs to pose application questions and set up cases based on the knowledge content that the students choose to supplement, the corresponding knowledge content in the expert course knowledge base, the previous teaching statement C, and the students' responses. The questions are in the form of closed-ended questions and open-ended questions, with the difficulty of the questions ranging from simple to difficult.

[0103] (4) Input the generative interactive prompt instruction set environment, the knowledge content, and the generative interactive prompt instruction set into the large language model for natural language processing to generate teaching statement C;

[0104] (5) If the content does not meet the students’ learning needs, repeat step (4) until a satisfactory teaching statement result is generated.

[0105] Furthermore, taking the concept of "pointers" from the university course "C Programming Language" as an example, an interactive learning environment is created to generate personalized feedback statements D, such as... Figure 4 As shown, the specific steps are as follows:

[0106] (1) Call a large language model (such as Chatglm-4, Chatglm-3-turbo, etc.), configure a generative interactive feedback prompt instruction set environment, and set the system information to "You are a setup assistant in an intelligent tutoring system. Your task is to provide interactive tutoring for users. Now you need to construct an expert Q&A teacher T around the knowledge point 'pointer'". QA "To answer students' questions and promote their deeper understanding";

[0107] (2) Obtain the knowledge content that students need assistance with, the teaching statements C, and the students' responses Sr;

[0108] (3) Determine the expert Q&A teacher T QA A generative interactive feedback prompt instruction set, wherein the generative interactive feedback prompt instruction set includes {P1(T A ), P2(T A ), P3(T A ), P4(T A )},P1(T A When a student's response Sr is non-empty and related to the knowledge content the student selected that requires assistance, the expert teacher should provide feedback to the student regarding their response Sr. This feedback may include judging the correctness of the answer, offering encouragement or praise, and providing detailed explanations and hints. P2(T) AWhen a student's response Sr is not empty and is unrelated to the knowledge content requiring assistance selected by the student, the teacher prompts the student to answer based on the selected knowledge content or to modify the selected knowledge content; P3(T) A When a student does not respond within a short period of time (e.g., the time spent is >= 5 minutes), it is necessary to ask the student, in the role of an expert teacher, if they need help; P4(T) A This means that if a student does not respond for an extended period of time (e.g., stay time >= 10 minutes), they exit the interactive learning environment.

[0109] (4) Input the generative interactive feedback prompt instruction set environment, the knowledge content, and the generative interactive feedback prompt instruction set into the large language model for natural language processing to generate personalized feedback statement D;

[0110] (5) When P2(T) in the generative interactive feedback prompt instruction set is... A When activated, repeat steps (2) to (4) until the student exits the interactive learning context.

[0111] Furthermore, to verify the effectiveness of the method of this invention, it was applied to university students tutoring the "C Programming Language" course. Their learning performance and interview results were collected, and the results are as follows: Figure 5 As shown, the horizontal axis represents knowledge points. At the instructor's request, sequential and selection structures were combined into a single test titled "Data Types," while loop structures were tested under the "Programs and Flow" category. The vertical axis represents the answer rate and the students' average score. For ease of presentation, the average score is shown as a percentage. The results showed that students achieved satisfactory results using the method of this invention for intelligent learning on different knowledge points. The more complex the knowledge points (e.g., functions, pointers), the higher the students' learning motivation and the better the learning outcomes. Interview details are as follows: Figure 6 As shown, the larger the font size, the higher the frequency of the keyword. Most students believe that this method is helpful for learning and is suitable for learning knowledge points. This type of tool provides them with a new way of guiding learning, and it is easy to use and understand, and has the value of continuous use and promotion.

[0112] This invention also provides a team-based dialogue generation and tutoring learning system based on a large language model, such as... Figure 7 As shown, it includes the following modules:

[0113] The login and registration module is used to record student information so that the system can track students' learning progress and needs;

[0114] The expert course knowledge base module is used to store knowledge points, teaching resources and learning materials of various disciplines, so that students can access them at any time and build them according to the needs of the course and the knowledge structure of the course.

[0115] The self-observation learning module is used to generate personalized knowledge dialogues A and B based on generative teacher prompts and student prompts, helping students understand and master knowledge through observation.

[0116] The self-directed interactive learning module provides teaching statements C based on the generative interactive prompt instruction set, inducing students to participate in knowledge interaction; based on student feedback and the generative interactive feedback prompt instruction set, it generates personalized feedback statements D from expert teachers to respond to students' knowledge needs. Through interactive learning, students deepen their understanding and application of knowledge.

[0117] The specific implementation methods of the expert course knowledge base module, the self-observation learning module, and the self-interactive learning module are the same as those in steps 1-4, and will not be described in this invention.

[0118] The specific embodiments described herein are merely illustrative of the spirit of the invention. Those skilled in the art to which this invention pertains may make various modifications or additions to the described specific embodiments or use similar methods to substitute them, without departing from the spirit of the invention or exceeding the scope defined by the appended claims.

Claims

1. A team-based dialogue generation and tutoring learning method based on a large language model, characterized in that, Includes the following steps: Step 1: Based on the course requirements and knowledge structure, construct a rule-based expert course knowledge base; Step 2: Enter the observational learning context. Based on the generative teacher prompt instruction set, generate personalized knowledge dialogue A between different teachers and fixed students. Based on the generative student prompt instruction set, generate personalized knowledge dialogue B between fixed teachers and different students. The personalized knowledge dialogue A and personalized knowledge dialogue B are used to assist students in understanding knowledge and accompany students in learning knowledge, respectively. Step 3: Enter the interactive learning context, initiate a set of prompts based on generative interaction, and generate teaching statements C from expert teachers. These teaching statements C are used to induce students to participate in knowledge interaction. Step 4: Enter the interactive learning context. Based on the students' feedback and the generative interactive feedback prompt instruction set, generate personalized feedback statements D from expert teachers to respond to the students' knowledge needs. The steps for generating personalized feedback statement D in step 4 are as follows: (41) Call the large language model and configure the generative interactive feedback prompt instruction set environment; (42) Obtain the knowledge content that students choose to supplement, the teaching statement C, and the students' responses Sr; (43) Determine the expert Q&A teacher T QA A generative interactive feedback prompt instruction set, the generative interactive feedback prompt instruction set including {P1(T A ), P2(T A ), P3 (T A ), P4 (T A )}, P1(T A When a student's response Sr is non-empty and related to the knowledge content the student selected that requires assistance, the expert teacher provides feedback to the student based on their response Sr. This feedback can take the form of judging the correctness of the answer, offering encouragement or praise, or providing detailed explanations and hints. P2(T) A ) is when a student's reply Sr is not empty and is unrelated to the knowledge content the student selected that needs assistance, prompting the student to answer around the knowledge content the student selected that needs assistance, or to modify the selected knowledge content that needs assistance; P3(T A P4 (T) is used when a student does not respond within a short period of time, and the expert teacher asks the student if they need help; A This means that if a student does not respond for an extended period of time, they will exit the interactive learning environment. (44) Input the generative interactive feedback prompt instruction set environment, knowledge content, and generative interactive feedback prompt instruction set into the large language model for natural language processing to generate personalized feedback statement D; (45) When the generative interactive feedback prompt instruction set P2(T) A When ) is activated, repeat steps (42) to (44) until the student exits the interactive learning context; Step 5: Repeat steps 2, 3, and 4 to optimize personalized knowledge dialogue A, personalized knowledge dialogue B, teaching statements C, and personalized feedback statements D until the student no longer repeatedly enters steps 2, 3, and 4. At this point, it is considered that the student has mastered the course knowledge, thus achieving intelligent learning guidance for the student.

2. The team-based dialogue generation and tutoring learning method based on a large language model as described in claim 1, characterized in that: Step 1, constructing a rule-based expert course knowledge base, includes: Organize course knowledge concepts, key points, difficulties, and case studies according to the overall knowledge representation rules; Based on the rule of progressively deepening knowledge from simple to complex, organize the connections between key and difficult knowledge points. These connections include sequential and complementary relationships. The course knowledge concepts, key points and difficulties, case studies, and connections between key points and difficulties are compiled to obtain an expert course knowledge base.

3. The team-based dialogue generation and tutoring learning method based on a large language model as described in claim 1, characterized in that: The steps for generating personalized knowledge dialogue A and personalized knowledge dialogue B in step 2 are as follows: (21) Call the large language model and configure the generative teacher prompt instruction set environment and the generative student prompt instruction set environment; (22) Obtain the knowledge content that students choose to supplement, and obtain the corresponding knowledge content in the expert course knowledge base; (23) Determine the generative teacher prompt instruction set and the generative student prompt instruction set; (24) Input the generative teacher prompting instruction set environment and the generative student prompting instruction set environment, knowledge content, generative teacher prompting instruction set and generative student prompting instruction set into the large language model for natural language processing to generate personalized knowledge dialogue A and personalized knowledge dialogue B.

4. The team-based dialogue generation and tutoring learning method based on a large language model as described in claim 3, characterized in that: The generative teacher prompting instruction set includes: (a1) Construct generative teachers T={t1, t2, t3} and randomly generated students s, where t1, t2, and t3 are active teachers, constructive teachers, and interactive teachers, respectively; (a2) The prompting instructions of the proactive teacher t1 are to directly tell the randomly generated students s what knowledge point K is, what the principle of the knowledge point is, and how to apply the knowledge point, so as to stimulate the feedback of the randomly generated students s; the prompting instructions of the constructive teacher t2 are to tell the randomly generated students s knowledge points, principles and applications through analogy and metaphor, so as to stimulate the randomly generated students s to connect new information with existing knowledge; the prompting instructions of the interactive teacher t3 are to guide the randomly generated students s through cases and questions, and encourage the randomly generated students s to construct their understanding of knowledge points and participate in classroom activities. (a3) Input the constructed generative teacher T into the large language model to obtain the classroom speech of the generative teacher T; (a4) Determine the prompting instructions P(s) of the randomly generated student s, where P(s) is the response of the generated teacher T to the generated teacher T in the role of an ordinary student, based on the classroom speech of the generated teacher T. (a5) The prompting instructions of the constructed generative teacher T={t1, t2, t3} and the randomly generated student s are the generative teacher prompting instruction set.

5. The team-based dialogue generation and tutoring learning method based on a large language model as described in claim 3, characterized in that: The generative student prompt instruction set includes: (b1) Construct a randomized teacher t and a generative student S={s1, s2}, where s1 and s2 are expert students and novice students, respectively; (b2) Determine the prompting instructions P(t) of the randomly generated teacher t, where P(t) is the classroom dialogue conducted by the expert teacher around the knowledge point "pointer"; (b3) Input the constructed randomized teacher t into the large language model to obtain the classroom speech of the generative teacher T; (b4) Determine the prompting instructions for the generative student S. The prompting instructions for the expert student S1 are to correctly answer and supplement the knowledge points taught by the random generative teacher t based on the classroom language of the random generative teacher t, and to ask questions to the random generative teacher t and apply the knowledge in other ways. The prompting instructions for the novice student S2 are to incorrectly answer or refuse to answer the knowledge points taught by the random generative teacher t based on the classroom language of the random generative teacher t, and to ask questions to the random generative teacher t and seek help. (b5) The prompting instructions of the constructed random generating teacher t and generating student S={s1, s2} are the prompting instruction set of generating student.

6. The team-based dialogue generation and tutoring learning method based on a large language model as described in claim 1, characterized in that: The steps for generating the teaching statement C in step 3 are as follows: (31) Call the large language model and configure the generative interactive prompt instruction set environment; (32) Obtain the knowledge content that students choose to supplement, and obtain the corresponding knowledge content in the expert course knowledge base; (33) Determine the generative interaction initiation prompt instruction set, wherein the prompt instruction set includes {P1(T Q ), P2(T Q )};P1(T Q P2 (T) is when students first enter an interactive learning environment, the expert teacher asks the students if they need learning assistance; Q When students are not in an interactive learning context for the first time, the expert teacher, in the role of an expert teacher, poses application questions and sets up cases for students based on the knowledge content that students choose to supplement, the corresponding knowledge content in the expert course knowledge base, the previous teaching statement C, and the students' responses. The questions are in the form of closed-ended questions and open-ended questions, with the difficulty of the questions ranging from simple to difficult. (34) Input the generative interaction initiation prompt instruction set environment, knowledge content, and generative interaction initiation prompt instruction set into the large language model for natural language processing to generate teaching statement C.

7. A team-based dialogue generation and tutoring learning method based on a large language model as described in any one of claims 3-6, characterized in that: The large language models include Chatglm-4 and Chatglm-3-turbo.

8. A team-based dialogue generation and tutoring learning system based on a large language model, characterized in that, Includes the following modules: The login and registration module is used to record student information so that the system can track students' learning progress and needs; The expert course knowledge base module is used to store knowledge points, teaching resources and learning materials of various disciplines, so that students can access them at any time and build them according to the needs of the course and the knowledge structure of the course. The self-observation learning module is used to generate personalized knowledge dialogues A and B based on generative teacher prompts and student prompts, helping students understand and master knowledge through observation. The self-directed interactive learning module provides teaching statements C based on the generative interactive prompt instruction set to induce students to participate in knowledge interaction; based on student feedback and the generative interactive feedback prompt instruction set, it generates personalized feedback statements D from expert teachers to respond to students' knowledge needs. Through interactive learning, students deepen their understanding and application of knowledge. The steps for generating personalized feedback statements (D) are as follows: (41) Call the large language model and configure the generative interactive feedback prompt instruction set environment; (42) Obtain the knowledge content that students choose to supplement, the teaching statement C, and the students' responses Sr; (43) Determine the expert Q&A teacher T QA A generative interactive feedback prompt instruction set, the generative interactive feedback prompt instruction set including {P1(T A ), P2(T A ), P3 (T A ), P4 (T A )}, P1(T A When a student's response Sr is non-empty and related to the knowledge content the student selected that requires assistance, the expert teacher provides feedback to the student based on their response Sr. This feedback can take the form of judging the correctness of the answer, offering encouragement or praise, or providing detailed explanations and hints. P2(T) A ) is when a student's reply Sr is not empty and is unrelated to the knowledge content the student selected that needs assistance, prompting the student to answer around the knowledge content the student selected that needs assistance, or to modify the selected knowledge content that needs assistance; P3(T A P4 (T) is used when a student does not respond within a short period of time, and the expert teacher asks the student if they need help; A This means that if a student does not respond for an extended period of time, they will exit the interactive learning environment. (44) Input the generative interactive feedback prompt instruction set environment, knowledge content, and generative interactive feedback prompt instruction set into the large language model for natural language processing to generate personalized feedback statement D; (45) When the generative interactive feedback prompt instruction set P2(T) A When ) is activated, repeat steps (42) to (44) until the student exits the interactive learning context.

Images