Question answering tutoring method and related device

Abstract

The invention discloses a question answering tutoring method and a related device, and relates to the technical field of artificial intelligence, in the scheme, a first large model is pre-trained by using question stems, answers, analysis and question answering tutoring dialogue data corresponding to question samples, and after pre-training is performed, optimization is performed through supervised fine tuning training, so that the accuracy of question answering tutoring is improved. The training effect of the question answering tutoring model is greatly improved, the question answering tutoring model is connected with the multiple subject function plug-ins, each subject function plug-in corresponds to one function requirement of one subject, the question answering tutoring model is called to process the student question to obtain the answer corresponding to the student question, the accuracy and reliability of the answer can be improved, and the user experience is improved. And the effect of question answering tutoring is improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a question-answering tutoring method and related devices. Background Art

[0002] In the learning process, students often encounter various problems and need timely and effective Q&A guidance. Traditional Q&A guidance methods include manual guidance (mainly including face-to-face guidance by teachers, online guidance platforms, etc.) and traditional intelligent guidance methods (such as some simple intelligent guidance software, auxiliary learning tools based on large models, etc.). Traditional intelligent guidance methods mostly use dialogue systems to simulate human dialogue to provide students with Q&A guidance.

[0003] However, the traditional question-answering tutoring method has many limitations. Among them, in the manual tutoring method, teachers’ face-to-face tutoring resources are limited, and it is difficult to meet the needs of a large number of students anytime and anywhere; online tutoring platforms often rely on the online time of manual teachers, and there may be problems with uneven teacher quality. In the traditional intelligent tutoring method, although the intelligent tutoring software can provide a certain degree of automated question answering, it is difficult to give accurate and comprehensive answers to complex and diverse free dialogue questions, especially the logical reasoning and calculation processes involved in the questions. Auxiliary learning tools based on large models often cannot ensure that every step in the problem-solving process is accurate, especially when it comes to solving complex problems, the errors of the large model may accumulate and eventually affect the accuracy of the answer. Therefore, auxiliary learning tools based on large models also have certain problems.

[0004] Therefore, how to provide an intelligent question-answering and tutoring method to improve the accuracy and reliability of answers, and thereby improve the effect of question-answering and tutoring, has become a technical problem that needs to be urgently solved by technical personnel in this field. Summary of the invention

[0005] In view of the above problems, this application provides a question-answering tutoring method and related devices to achieve the purpose of improving the accuracy and reliability of answers, thereby improving the effect of question-answering tutoring. The specific scheme is as follows:

[0006] The first aspect of the present application provides a question-answering tutoring method, comprising:

[0007] Get student questions;

[0008] The question-answering tutoring model is called to process the student question to obtain the answer corresponding to the student question. The question-answering tutoring model is obtained by pre-training and supervised fine-tuning the first large model in sequence using the question stem, answer, analysis and question-answering tutoring dialogue data corresponding to the question sample. The question-answering tutoring model is connected to multiple subject function plug-ins, and each subject function plug-in corresponds to a functional requirement of a subject.

[0009] In a possible implementation, for each question sample, the question-answering tutoring dialogue data corresponding to the question sample is generated using the second largest model according to the question stem and analysis corresponding to the question sample.

[0010] In a possible implementation, the method of generating the question-answering tutoring dialogue data corresponding to the question sample includes:

[0011] Obtaining a preset question-answering and tutoring dialogue data generation prompt template, wherein the question-answering and tutoring dialogue data generation prompt template includes question-answering and tutoring dialogue data generation task description information, a question stem filling slot, and a resolution filling slot;

[0012] Fill the stem of the question sample into the stem filling slot, and fill the solution of the question sample into the solution filling slot, to obtain a question-answering tutoring dialogue data generation prompt;

[0013] The Prompt generated by the question-answering and tutoring dialogue data is input into the second large model to obtain the question-answering and tutoring dialogue data corresponding to the question sample generated by the second large model.

[0014] In a possible implementation, the training method of the question-answering tutoring model includes:

[0015] Pre-training the first large model using the question stem and the answer corresponding to the question sample to obtain the pre-trained first large model;

[0016] The pre-trained first large model is subjected to supervised fine-tuning training using the question stem, analysis, and question-answering tutoring dialogue data corresponding to the question sample to obtain the question-answering tutoring model.

[0017] In a possible implementation, calling the question-answering tutoring model to process the student question and obtain an answer corresponding to the student question includes:

[0018] Performing semantic analysis on the student questions to extract subject features of the student questions;

[0019] Calling a target subject function plug-in to process the student question and obtain a plug-in processing result; the target subject function plug-in is a subject function plug-in that matches the subject characteristics of the student question among the multiple subject function plug-ins;

[0020] The question-answering tutoring model is called to generate the answer according to the plug-in processing result.

[0021] In a possible implementation, calling the target subject function plug-in to process the student question and obtaining the plug-in processing result includes:

[0022] Performing semantic analysis on the student questions to extract key knowledge points and computing requirement information of the student questions;

[0023] Passing the key knowledge points and computing requirement information of the student question to the target subject function plug-in, so that the target subject function plug-in processes the student question based on the key knowledge points and computing requirement information of the student question;

[0024] Obtain the plug-in processing result returned after the target subject function plug-in processes the student problem.

[0025] The second aspect of the present application provides a question-answering and tutoring device, comprising:

[0026] An acquisition unit, used to acquire student questions;

[0027] The answering unit is used to call the question-answering tutoring model to process the student question and obtain the answer corresponding to the student question. The question-answering tutoring model is obtained by pre-training and supervised fine-tuning the first large model in sequence using the question stem, answer, analysis and question-answering tutoring dialogue data corresponding to the question sample. The question-answering tutoring model is connected to multiple subject function plug-ins, and each subject function plug-in corresponds to a functional requirement of a subject.

[0028] In a possible implementation, for each question sample, the question-answering tutoring dialogue data corresponding to the question sample is generated using the second largest model according to the question stem and analysis corresponding to the question sample.

[0029] In a possible implementation, the device further includes: a question-answering and tutoring dialogue data generating unit;

[0030] The question-answering and coaching dialogue data generating unit is specifically used for:

[0031] Obtaining a preset question-answering and tutoring dialogue data generation prompt template, wherein the question-answering and tutoring dialogue data generation prompt template includes question-answering and tutoring dialogue data generation task description information, a question stem filling slot, and a resolution filling slot;

[0032] Fill the stem of the question sample into the stem filling slot, and fill the solution of the question sample into the solution filling slot, to obtain a question-answering tutoring dialogue data generation prompt;

[0033] The Prompt generated by the question-answering and tutoring dialogue data is input into the second large model to obtain the question-answering and tutoring dialogue data corresponding to the question sample generated by the second large model.

[0034] In a possible implementation, the device further includes: a question-answering tutoring model training unit;

[0035] The question-answering tutoring model training unit is specifically used for:

[0036] Pre-training the first large model using the question stem and the answer corresponding to the question sample to obtain the pre-trained first large model;

[0037] The pre-trained first large model is subjected to supervised fine-tuning training using the question stem, analysis, and question-answering tutoring dialogue data corresponding to the question sample to obtain the question-answering tutoring model.

[0038] In a possible implementation, the answering unit includes:

[0039] A semantic analysis unit, used for performing semantic analysis on the student question and extracting subject features of the student question;

[0040] A plug-in processing unit, used for calling a target subject function plug-in to process the student question and obtain a plug-in processing result; the target subject function plug-in is a subject function plug-in that matches the subject feature of the student question among the multiple subject function plug-ins;

[0041] The answer generation unit is used to call the question-answering tutoring model to generate the answer according to the plug-in processing result.

[0042] In a possible implementation, the plug-in processing unit is specifically configured to:

[0043] Performing semantic analysis on the student questions to extract key knowledge points and computing requirement information of the student questions;

[0044] Passing the key knowledge points and computing requirement information of the student question to the target subject function plug-in, so that the target subject function plug-in processes the student question based on the key knowledge points and computing requirement information of the student question;

[0045] Obtain the plug-in processing result returned after the target subject function plug-in processes the student problem.

[0046] The third aspect of the present application provides a computer program product, including computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements the question-answering and tutoring method of the first aspect or any implementation of the first aspect.

[0047] A fourth aspect of the present application provides an electronic device, comprising at least one processor and a memory connected to the processor, wherein:

[0048] The memory is used to store computer programs;

[0049] The processor is used to execute the computer program so that the electronic device can implement the question-answering and tutoring method of the above-mentioned first aspect or any implementation method of the first aspect.

[0050] A fifth aspect of the present application provides a computer-readable storage medium, which carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement the question-answering and tutoring method of the above-mentioned first aspect or any implementation of the first aspect.

[0051] By means of the above technical scheme, the present application provides a method and related device for question-answering tutoring. In this scheme, the first large model is first pre-trained using the question stem, answer, analysis and question-answering tutoring dialogue data corresponding to the question sample. After the pre-training, it is optimized through supervised fine-tuning training, which greatly improves the training effect of the question-answering tutoring model. In addition, the question-answering tutoring model is connected to multiple subject function plug-ins, each subject function plug-in corresponds to a functional requirement of a subject, and the question-answering tutoring model is called to process student questions and obtain the answers corresponding to the student questions, which can improve the accuracy and reliability of the answers, thereby improving the effect of question-answering tutoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0053] Figure 1 A flowchart of a question-answering tutoring method provided in an embodiment of the present application;

[0054] Figure 2 A flowchart of a method for generating question-answering and tutoring dialogue data corresponding to a question sample provided in an embodiment of the present application;

[0055] Figure 3 A flowchart of a training method for a question-answering tutoring model provided in an embodiment of the present application;

[0056] Figure 4 A flowchart of a method for invoking a question-answering tutoring model to process a student's question and obtain an answer to the student's question provided in an embodiment of the present application;

[0057] Figure 5 A schematic diagram of the structure of a question-answering and tutoring device provided in an embodiment of the present application;

[0058] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0059] The following describes the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. The terms used in the implementation method section of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application.

[0060] The embodiments of the present application are described below in conjunction with the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0061] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, which is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0062] In the learning process, students often encounter various problems and need timely and effective Q&A guidance. Traditional Q&A guidance methods include manual guidance (mainly including face-to-face guidance by teachers, online guidance platforms, etc.) and traditional intelligent guidance methods (such as some simple intelligent guidance software, auxiliary learning tools based on large models, etc.). Traditional intelligent guidance methods mostly use dialogue systems to simulate human dialogue to provide students with Q&A guidance.

[0063] A dialogue system is a computer system based on artificial intelligence and natural language processing (NLP) technology, designed to interact with users through natural language. It can understand user input and generate corresponding responses, simulating human conversation.

[0064] As an important research direction in the field of artificial intelligence, dialogue systems have a wide range of applications in real life and technology. First, dialogue systems can provide a natural and intuitive way of human-computer interaction. In real life, the main way for people to communicate is through natural language dialogue; secondly, dialogue systems have important applications in the fields of personal assistants, intelligent customer service, and personalized education. In addition, dialogue systems can enrich the way users acquire knowledge and retrieve information, and provide users with accurate and real-time answers and solutions; finally, the development of dialogue systems has also promoted the advancement of artificial intelligence technology.

[0065] However, the traditional question-and-answer tutoring method has many limitations.

[0066] Among them, in the manual tutoring method, teachers' face-to-face tutoring resources are limited, which makes it difficult to meet the needs of a large number of students anytime and anywhere; online tutoring platforms often rely on the online time of manual teachers, and there may be problems with uneven teacher quality.

[0067] In traditional intelligent tutoring methods, although intelligent tutoring software can provide a certain degree of automated answering, it is difficult to give accurate and comprehensive answers to complex and diverse free dialogue questions, especially those involving logical reasoning and calculation processes.

[0068] Intelligent tutoring software usually uses rule-based expert systems or simple machine learning models. Rule-based expert systems require a large number of rules to be manually written to deal with different types of problems, such as writing specific problem-solving algorithm rules for mathematical problems and specific application rules for physical formulas. The advantage of this method is that it can give relatively accurate answers to specific types of problems and is more explanatory. However, its disadvantages are also very obvious. The workload of writing rules is huge and it is difficult to cover all possible problem scenarios. It cannot respond flexibly to new problems or expressions, and has poor scalability. Simple machine learning models collect a certain amount of answered question data as a training set, train the model to predict the answers to new questions. However, due to the limited amount of training data and insufficient data feature extraction, this model often has a low accuracy rate when facing complex problems, especially those involving multi-step reasoning and calculation.

[0069] In recent years, with the rapid development of artificial intelligence technology, the application of big models in the field of education has gradually become a research hotspot. The emergence of big models has brought new opportunities for question-answering and tutoring. At present, learning tools based on big models have been widely used in educational scenarios of different disciplines, such as mathematics and physics problem-solving assistants based on big models. Although big models have powerful language understanding and generation capabilities and can handle various problems in natural language form, in the application scenario of question-answering and tutoring, big models often cannot ensure that every step in the problem-solving process is accurate. Especially when it comes to solving complex problems, the errors of big models may accumulate and eventually affect the accuracy of the answers.

[0070] In order to improve the performance of auxiliary learning tools based on large models, some auxiliary learning tools based on large models currently enhance the computing power of the model by integrating specific plug-ins, especially in mathematical problems, by calculating complex mathematical formulas through plug-ins. However, these plug-ins are highly dependent and poorly integrated with the overall model, which easily leads to inconsistencies between the calculation results and the model reasoning.

[0071] In addition, due to the lack of a large number of real intelligent tutoring data sets that do not require manual annotation, many large models currently still rely on manual annotation in the process of generating training data sets. Therefore, many auxiliary learning tools based on large models currently rely on manually annotated training data sets for training, which may not only lead to deviations in the training data sets, but also make the model training process costly and inefficient.

[0072] Therefore, there are still some problems with the accuracy and reliability of auxiliary learning tools based on large models.

[0073] In order to solve the above problems, the embodiment of the present application provides a question-answering and coaching method. The question-answering and coaching method of the embodiment of the present application is described in detail below in conjunction with the accompanying drawings.

[0074] Reference Figure 1 , Figure 1 A flowchart of a question-answering tutoring method provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, a question-answering tutoring method provided in an embodiment of the present application may include the following steps, which are described in detail below.

[0075] S101: Get student questions;

[0076] The student problem can be a problem of any subject. In one achievable manner, the student problem can be a problem of a science subject (eg, mathematics, physics, chemistry, etc.).

[0077] S102: Calling the question-answering tutoring model to process the student question and obtain the answer corresponding to the student question. The question-answering tutoring model is obtained by pre-training and supervised fine-tuning the first large model in sequence using the question stem, answer, analysis and question-answering tutoring dialogue data corresponding to the question sample. The question-answering tutoring model is connected to multiple subject function plug-ins, and each subject function plug-in corresponds to a functional requirement of a subject.

[0078] In the present application, a large number of practice questions, test questions, etc. can be collected in advance from the question bank as question samples. The question types can be any type, including but not limited to multiple-choice questions, fill-in-the-blank questions, short-answer questions, etc. The subjects covered can be any subject (such as mathematics, physics, chemistry, etc.), and the question stems, answers, and analyses corresponding to the question samples can also be obtained from the preset question bank. The question-answering tutoring dialogue data corresponding to the question samples can be manually constructed based on the question stems, answers, and analyses corresponding to the question samples, or can be automatically constructed. This application does not impose any restrictions on this.

[0079] In the present application, the first large model may be any existing large model, and the present application does not impose any limitation on this.

[0080] In this application, the question-answering and tutoring dialogue data includes n rounds of dialogues between the roles of "teacher" and "student", where n is an integer greater than or equal to 1. For example:

[0081] Student Question 1: xxx;

[0082] Teacher reply 1: xxx;

[0083] Student Question 2: xxx;

[0084] Teacher reply 2: xxx;

[0085] ……;

[0086] Student question n: xxx;

[0087] The teacher replied n: xxx.

[0088] In order to improve the performance of auxiliary learning tools based on large models, some auxiliary learning tools based on large models enhance the computing power of the model by integrating specific plug-ins, especially in mathematical problems, by calculating complex mathematical formulas through plug-ins. However, these plug-ins are highly dependent and poorly integrated with the overall model, which easily leads to inconsistencies between the calculation results and the model reasoning.

[0089] In view of this, in this application, corresponding subject function plug-ins can be developed for different subjects and different functional requirements of the subjects. For example, in the subject of mathematics, a calculation plug-in is developed, which can parse mathematical expressions and perform precise numerical calculations, symbolic operations, and formula derivations. In the subject of physics, a physical quantity unit conversion plug-in, a physical formula calculation plug-in, etc. are developed to handle various calculation and unit conversion requirements in physical problems.

[0090] In this application, the subject function plug-in can adopt a modular design concept, and each plug-in independently encapsulates a specific function, which has good scalability and maintainability. And a special plug-in interface can be designed to connect each subject function plug-in with the question-answering tutoring model. In addition, an efficient communication protocol needs to be configured between each subject function plug-in and the question-answering tutoring model to ensure accurate data transmission.

[0091] The present embodiment provides a question-answering tutoring method. In this scheme, the first large model is pre-trained using the question stem, answer, analysis and question-answering tutoring dialogue data corresponding to the question sample. After the pre-training, it is optimized through supervised fine-tuning training, which greatly improves the training effect of the question-answering tutoring model. In addition, the question-answering tutoring model is connected to multiple subject function plug-ins, each subject function plug-in corresponds to a functional requirement of a subject. The question-answering tutoring model is called to process student questions and obtain answers corresponding to student questions, which can improve the accuracy and reliability of the answers, thereby improving the effect of question-answering tutoring.

[0092] In real-life Q&A tutoring scenarios, due to the lack of a large number of real, intelligent tutoring data sets that do not require manual annotation, many large models currently still rely on manual annotation in the generation of training data sets. Therefore, many auxiliary learning tools based on large models currently rely on manually annotated training data sets for training, which may not only lead to deviations in the training data sets, but also make the model training process costly and inefficient. In this application, the language generation capabilities of the large model itself can be used to generate a large amount of Q&A tutoring dialogue data.

[0093] Then in a possible implementation, for each question sample, the question-answering tutoring dialogue data corresponding to the question sample is generated using the second largest model according to the question stem and analysis corresponding to the question sample.

[0094] In the present application, the second large model may be any existing large model, and the first large model and the second large model may be the same large model or different large models, and the present application does not impose any limitation on this.

[0095] In another embodiment of the present application, a method for generating question-answering and tutoring dialogue data corresponding to the question sample is described in detail.

[0096] Reference Figure 2 , Figure 2 A flowchart of a method for generating question-answering tutoring dialogue data corresponding to a question sample provided in an embodiment of the present application, the method may include the following steps:

[0097] S201: Obtaining a preset question-answering and tutoring dialogue data generation prompt template, wherein the question-answering and tutoring dialogue data generation prompt template includes question-answering and tutoring dialogue data generation task description information, question stem filling slots, and analysis filling slots;

[0098] In the present application, a question-answering and coaching dialogue data generation Prompt template may be preset, wherein the question-answering and coaching dialogue data generation Prompt template includes question-answering and coaching dialogue data generation task description information, question stem filling slots, and analysis filling slots.

[0099] For ease of understanding, the present application embodiment provides an example of a prompt template for generating question-answering and tutoring dialogue data, as follows:

[0100] "Please follow the question stem and the corresponding analysis, focus on the knowledge points related to the question, and use different question types and different difficulty levels as the dialogue goal, with the teacher answering the student's question, to simulate the two dialogue roles of "teacher" and "student" to generate dialogues, and output N groups of dialogues.

[0101] The stem of the question is [xxx], and the corresponding solution is [xxx]. ".

[0102] Preferably, the question-answering and coaching dialogue data generating Prompt template may also include a question-answering and coaching dialogue data example filling slot.

[0103] For ease of understanding, the present application embodiment provides another example of a prompt template for generating question-answering and tutoring dialogue data, as follows:

[0104] "Please follow the question stem and the corresponding analysis, focus on the knowledge points related to the question, and use different question types and different difficulty levels as the dialogue goal, with the teacher answering the student's question, to simulate the two dialogue roles of "teacher" and "student" to generate dialogues, and output N groups of dialogues.

[0105] The stem of the question is [xxx], the corresponding solution is [xxx], and the dialogue example is [xxx]. "

[0106] S202: Fill the stem of the question sample into the stem filling slot, and fill the solution of the question sample into the solution filling slot, to obtain question-answering tutoring dialogue data generation prompt;

[0107] S203: Input the Prompt generated by the question-answering and tutoring dialogue data into the second large model to obtain the question-answering and tutoring dialogue data corresponding to the question sample generated by the second large model.

[0108] In a possible implementation, the format of the question-answering and tutoring dialogue data generated by the large model may specifically be a dialogue of n rounds with the roles of "teacher" and "student", where n is an integer greater than or equal to 1. For example:

[0109] Student Question 1: xxx;

[0110] Teacher reply 1: xxx;

[0111] Student Question 2: xxx;

[0112] Teacher reply 2: xxx;

[0113] ……;

[0114] Student question n: xxx;

[0115] The teacher replied n: xxx.

[0116] For example, in the field of mathematics, the question-answering and tutoring dialogue data generated by the big model can be:

[0117] Student question: Please provide an example of an applied problem regarding [a mathematical knowledge point, such as function derivative] and give a detailed solution process.

[0118] The teacher replied: xxxx.

[0119] For another example, for a thought-based question, the Q&A coaching dialogue training data generated by the big model can be:

[0120] Student question: I have no idea how to solve this problem. Can you help me explain it?

[0121] The teacher replied: Student, your ideas are very good. Let’s take a look at the complete solution to this problem: xxxx (describe the solution to this problem).

[0122] For example, for detailed explanation questions, the Q&A tutoring dialogue training data generated by the big model can be:

[0123] Student question: What does xxxx (concept, formula, meaning, etc.) mean?

[0124] The teacher replied: xxxx is xxxx (further explanation of concepts, formulas, meanings, etc.).

[0125] In this application, although the Q&A tutoring dialogue data is generated by a large model, it is based on a specific Prompt template and is centered around knowledge points, so it has high authenticity and practicality. Moreover, by using the Q&A tutoring dialogue data to generate the Prompt template, it is possible to construct hierarchical and typed Q&A tutoring dialogue data, thus obtaining comprehensive and rich unlabeled Q&A tutoring dialogue data.

[0126] In another embodiment of the present application, the training method of the question-answering tutoring model is described in detail.

[0127] Reference Figure 3 , Figure 3 A flowchart of a training method for a question-answering tutoring model provided in an embodiment of the present application, the method may include the following steps:

[0128] S301: Pre-training the first large model using the question stem and the answer corresponding to the question sample to obtain the pre-trained first large model;

[0129] During the pre-training process, the first large model learns how to understand the intention of the question, extract key information, and try to give a reasonable answer. For example, for the pre-training task of math multiple-choice questions, the first large model needs to learn to analyze the differences between options and make inferences based on the conditions of the question; for the pre-training task of physics short-answer questions, the first large model needs to learn how to organize physical principles and formulas to answer questions clearly.

[0130] The first model is pre-trained using the questions and answers corresponding to the question samples, that is, using a multi-question-answering task learning method to allow the first model to simultaneously learn the answering skills and knowledge application methods of different subjects and question types. In this way, the first model initially has the foundation of knowledge understanding and answering ability, laying a good foundation for subsequent supervised fine-tuning (SFT) training.

[0131] S302: Using the question stem, analysis and question-answering tutoring dialogue data corresponding to the question sample, supervised fine-tuning training is performed on the pre-trained first large model to obtain the question-answering tutoring model.

[0132] During the supervised fine-tuning training process, the question stem, explanation and student questions are used as input, and the teacher's answer is used as the expected output. A loss function calculation method similar to traditional supervised learning, such as the cross-entropy loss function, is used to adjust the parameters of the first large model after pre-training. Through multiple iterative training, the first large model after pre-training gradually adapts to the task requirements of free dialogue-style question-answering tutoring, and improves its accuracy and logic in handling student questions. For example, when dealing with students' free questions about the explanation of physical experimental phenomena, the model can combine the physical knowledge learned in the pre-training stage and the optimization of natural language expression in supervised fine-tuning training to give detailed, accurate and easy-to-understand answers.

[0133] In this application, by combining this pre-training with supervised fine-tuning training, the comprehensive performance of the model in question-answering tutoring can be gradually improved, including the accuracy, logic, completeness of the answers, and interactivity with students.

[0134] Based on the above content, another embodiment of the present application describes a specific implementation method for calling the question-answering tutoring model to process the student question and obtain the answer corresponding to the student question.

[0135] Reference Figure 4 , Figure 4 A flowchart of a method for invoking a question-answering tutoring model to process a student's question and obtain an answer to the student's question provided in an embodiment of the present application, the method may include the following steps:

[0136] S401: Performing semantic analysis on the student's question to extract subject features of the student's question;

[0137] In the present application, the question-answering tutoring model may be called to perform semantic analysis on the student questions and extract the subject features of the student questions.

[0138] S402: calling a target subject function plug-in to process the student question and obtaining a plug-in processing result; the target subject function plug-in is a subject function plug-in that matches the subject feature of the student question among the multiple subject function plug-ins;

[0139] In this application, the corresponding plug-in can be automatically called for processing according to the subject characteristics of the student's question. For example, when it is identified that the question is about rounding in mathematics, the mathematical calculation plug-in is called, the number to be rounded is substituted into the selected plug-in, the solution is performed, and the result is returned to the question-answering tutoring model.

[0140] In a possible implementation, the calling of the target subject function plug-in to process the student problem and obtain the plug-in processing result includes: performing semantic analysis on the student problem to extract key knowledge points and computing requirement information of the student problem; passing the key knowledge points and computing requirement information of the student problem to the target subject function plug-in so that the target subject function plug-in processes the student problem based on the key knowledge points and computing requirement information of the student problem; and obtaining the plug-in processing result returned by the target subject function plug-in after processing the student problem.

[0141] In the present application, the question-answering tutoring model may be called to perform semantic analysis on the student questions to extract key knowledge points and computing requirement information of the student questions.

[0142] In the present application, the question-answering tutoring model can be called to pass the key knowledge points and computing requirement information of the student question to the target subject function plug-in in a specific format, so that the target subject function plug-in can process the student question based on the key knowledge points and computing requirement information of the student question.

[0143] In the present application, after the target subject function plug-in processes the student question based on the key knowledge points and computing requirement information of the student question to obtain the plug-in processing result, the plug-in processing result can be returned to the question-answering tutoring model in the above-mentioned specific format.

[0144] S403: Calling the question-answering tutoring model to generate the answer according to the plug-in processing result.

[0145] In this application, the question-answering tutoring model can integrate the plug-in processing results into natural language answers and generate answers to feed back to users.

[0146] In this application, the question-answering tutoring model combines multiple subject function plug-ins to realize question-answering tutoring. The use of multiple subject function plug-ins is dynamic. The question-answering tutoring model will automatically select the corresponding plug-in according to the type of question. The question-answering tutoring model plays a core role in question understanding, context association and natural language generation, while the subject function plug-in focuses on professional calculation and processing. The two work together to further improve the accuracy and reliability of the answers, thereby further improving the effect of question-answering tutoring.

[0147] A question-answering and coaching method provided by an embodiment of the present application is introduced above, and a device for executing the above-mentioned question-answering and coaching method will be introduced below.

[0148] See also Figure 5 , Figure 5 This is a schematic diagram of the structure of a question-answering and tutoring device provided in an embodiment of the present application. Figure 5 As shown, the question-answering and tutoring device comprises:

[0149] An acquisition unit 11, used for acquiring student questions;

[0150] The answering unit 12 is used to call the question-answering tutoring model to process the student question and obtain the answer corresponding to the student question. The question-answering tutoring model is obtained by pre-training and supervised fine-tuning the first large model in sequence using the question stem, answer, analysis and question-answering tutoring dialogue data corresponding to the question sample. The question-answering tutoring model is connected to multiple subject function plug-ins, and each subject function plug-in corresponds to a functional requirement of a subject.

[0151] In a possible implementation, for each question sample, the question-answering tutoring dialogue data corresponding to the question sample is generated using the second largest model according to the question stem and analysis corresponding to the question sample.

[0152] In a possible implementation, the device further includes: a question-answering and tutoring dialogue data generating unit;

[0153] The question-answering and coaching dialogue data generating unit is specifically used for:

[0154] Obtaining a preset question-answering and tutoring dialogue data generation prompt template, wherein the question-answering and tutoring dialogue data generation prompt template includes question-answering and tutoring dialogue data generation task description information, a question stem filling slot, and a resolution filling slot;

[0155] Fill the stem of the question sample into the stem filling slot, and fill the solution of the question sample into the solution filling slot, to obtain a question-answering tutoring dialogue data generation prompt;

[0156] The Prompt generated by the question-answering and tutoring dialogue data is input into the second large model to obtain the question-answering and tutoring dialogue data corresponding to the question sample generated by the second large model.

[0157] In a possible implementation, the device further includes: a question-answering tutoring model training unit;

[0158] The question-answering tutoring model training unit is specifically used for:

[0159] Pre-training the first large model using the question stem and the answer corresponding to the question sample to obtain the pre-trained first large model;

[0160] The pre-trained first large model is subjected to supervised fine-tuning training using the question stem, analysis, and question-answering tutoring dialogue data corresponding to the question sample to obtain the question-answering tutoring model.

[0161] In a possible implementation, the answering unit includes:

[0162] A semantic analysis unit, used for performing semantic analysis on the student question and extracting subject features of the student question;

[0163] A plug-in processing unit, used for calling a target subject function plug-in to process the student question and obtain a plug-in processing result; the target subject function plug-in is a subject function plug-in that matches the subject feature of the student question among the multiple subject function plug-ins;

[0164] The answer generation unit is used to call the question-answering tutoring model to generate the answer according to the plug-in processing result.

[0165] In a possible implementation, the plug-in processing unit is specifically configured to:

[0166] Performing semantic analysis on the student questions to extract key knowledge points and computing requirement information of the student questions;

[0167] Passing the key knowledge points and computing requirement information of the student question to the target subject function plug-in, so that the target subject function plug-in processes the student question based on the key knowledge points and computing requirement information of the student question;

[0168] Obtain the plug-in processing result returned after the target subject function plug-in processes the student problem.

[0169] The present application also provides an electronic device in an embodiment. Figure 6As shown, it shows a schematic diagram of the structure of an electronic device suitable for implementing the embodiment of the present application. The electronic device in the embodiment of the present application may include but is not limited to fixed terminals such as mobile phones, laptops, PDAs (personal digital assistants), PADs (tablet computers), desktop computers, etc. Figure 6 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0170] like Figure 6 As shown, the electronic device may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 to a random access memory (RAM) 603. When the electronic device is powered on, various programs and data required for the operation of the electronic device are also stored in the RAM 603. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0171] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a memory card, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although Figure 6 An electronic device having various devices is shown, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0172] An embodiment of the present application also provides a computer program product including computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements any one of the question-answering and tutoring methods provided in the embodiments of the present application.

[0173] A computer-readable storage medium is also provided in an embodiment of the present application. The storage medium carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement any question-answering and tutoring method provided in the embodiment of the present application.

[0174] It should also be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.

[0175] Through the description of the above implementation mode, the technicians in the field can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present application, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a ROM, a RAM, a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a training device, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0176] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0177] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a training device, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, training device, or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

Claims

1. A question-answering tutoring method, characterized in that: include: Get student questions; The question-answering tutoring model is called to process the student question to obtain the answer corresponding to the student question. The question-answering tutoring model is obtained by pre-training and supervised fine-tuning the first large model in sequence using the question stem, answer, analysis and question-answering tutoring dialogue data corresponding to the question sample. The question-answering tutoring model is connected to multiple subject function plug-ins, and each subject function plug-in corresponds to a functional requirement of a subject.

2. The method according to claim 1, characterized in that: For each question sample, the question-answering tutoring dialogue data corresponding to the question sample is generated using the second largest model based on the question stem and analysis corresponding to the question sample.

3. The method according to claim 2, characterized in that The method for generating the question-answering tutoring dialogue data corresponding to the question sample includes: Obtaining a preset question-answering and tutoring dialogue data generation prompt template, wherein the question-answering and tutoring dialogue data generation prompt template includes question-answering and tutoring dialogue data generation task description information, a question stem filling slot, and a resolution filling slot; Fill the stem of the question sample into the stem filling slot, and fill the solution of the question sample into the solution filling slot, to obtain a question-answering tutoring dialogue data generation prompt; The Prompt generated by the question-answering and tutoring dialogue data is input into the second large model to obtain the question-answering and tutoring dialogue data corresponding to the question sample generated by the second large model.

4. The method according to claim 1, characterized in that: The training method of the question-answering tutoring model includes: Pre-training the first large model using the question stem and the answer corresponding to the question sample to obtain the pre-trained first large model; The pre-trained first large model is subjected to supervised fine-tuning training using the question stem, analysis, and question-answering tutoring dialogue data corresponding to the question sample to obtain the question-answering tutoring model.

5. The method according to claim 1, characterized in that The calling of the question-answering tutoring model to process the student's question and obtain an answer corresponding to the student's question includes: Performing semantic analysis on the student questions to extract subject features of the student questions; Calling a target subject function plug-in to process the student question and obtain a plug-in processing result; the target subject function plug-in is a subject function plug-in that matches the subject characteristics of the student question among the multiple subject function plug-ins; The question-answering tutoring model is called to generate the answer according to the plug-in processing result.

6. The method according to claim 5, characterized in that The calling of the target subject function plug-in to process the student problem and obtain the plug-in processing result includes: Performing semantic analysis on the student questions to extract key knowledge points and computing requirement information of the student questions; Passing the key knowledge points and computing requirement information of the student question to the target subject function plug-in, so that the target subject function plug-in processes the student question based on the key knowledge points and computing requirement information of the student question; Obtain the plug-in processing result returned after the target subject function plug-in processes the student problem.

7. A question-answering and tutoring device, characterized in that: include: An acquisition unit, used to acquire student questions; The answering unit is used to call the question-answering tutoring model to process the student question and obtain the answer corresponding to the student question. The question-answering tutoring model is obtained by pre-training and supervised fine-tuning the first large model in sequence using the question stem, answer, analysis and question-answering tutoring dialogue data corresponding to the question sample. The question-answering tutoring model is connected to multiple subject function plug-ins, and each subject function plug-in corresponds to a functional requirement of a subject.

8. A computer program product, characterized in that It comprises computer-readable instructions, and when the computer-readable instructions are executed on an electronic device, the electronic device implements the question-answering and tutoring method as claimed in any one of claims 1 to 6.

9. An electronic device, characterized in that: The method comprises at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program so that the electronic device can implement the question-answering and tutoring method as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that: The storage medium carries one or more computer programs, and when the one or more computer programs are executed by an electronic device, the electronic device can implement the question-answering and tutoring method as described in any one of claims 1 to 6.

Images