Method, device, medium, and program product for guided questioning

Abstract

Embodiments of the present disclosure provide a method, device, storage medium and computer program product for guided questioning. The method comprises: generating, using a language model, a first question for a topic title; obtaining a first reply of a user reply to the first question; determining, using the language model, a key point for the topic title according to the first question and an answer corresponding to the first question; in response to the first reply not including a target answer corresponding to the topic title, generating, using the language model, a second question for the topic title; and presenting the key point and the second question. According to the method of the embodiments of the present disclosure, the user is actively initiated to ask questions for the topic title, which can help the user to select a suitable angle of attack for the topic title when the user is not clear about the key points of the question, guide the user to solve the topic title through the way of questioning, and allow the user to review the key information of the historical interaction content at any time, thereby improving the efficiency of solving problems.

Description

Method, device, medium and program product for guided questioning TECHNICAL FIELD

[0001] The present disclosure relates generally to the field of computers, and more specifically to a method, an electronic device, a computer-readable storage medium, and a computer program product for guided questioning. BACKGROUND

[0002] With the rapid development and continuous breakthroughs in deep learning technology, a large number of artificial intelligence question-answering products have emerged. Nowadays, more and more users begin to actively use these artificial intelligence question-answering products for autonomous learning. These artificial intelligence question-answering products can quickly and accurately analyze questions and provide users with detailed problem-solving ideas and methods.

[0003] For users themselves, this autonomous learning approach gives them great flexibility and autonomy. They can choose appropriate questions for practice at any time and any place according to their learning progress and needs, and are no longer restricted by time and space. On the other hand, from the perspective of educational resources, the widespread use of artificial intelligence question-answering products saves teacher resources.

[0004] SUMMARY

[0005] According to example embodiments of the present disclosure, a method, an electronic device, a computer storage medium, and a computer program product for guided questioning are provided.

[0006] In a first aspect of the present disclosure, a method for guided questioning is provided, comprising: generating, using a language model, a first question for a question title; obtaining a first reply of a user reply to the first question; determining, using the language model, a key point for the question title according to the first question and an answer corresponding to the first question; in response to the first reply not including a target answer corresponding to the question title, generating, using the language model, a second question for the question title; and presenting the key point and the second question.

[0007] In a second aspect of the present disclosure, an electronic device is provided, comprising: at least one processing unit; at least one memory coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit, which, when executed by the at least one processing unit, cause the electronic device to perform the method described in the first aspect of the present disclosure.

[0008] In a third aspect of the present disclosure, a computer-readable storage medium is provided, having machine-executable instructions stored thereon, which, when executed by a device, cause the device to perform the method described in the first aspect of the present disclosure.

[0009] In a fourth aspect of the present disclosure, a computer program product is provided, comprising computer executable instructions, wherein the computer executable instructions, when executed by a processor, implement the method described according to the first aspect of the present disclosure.

[0010] The summary is provided to introduce a selection of concepts that are further described in the detailed description below. This summary is not intended to identify key or essential features of the disclosure, nor is it intended to limit the scope of the disclosure. Other features will be more fully described in the following detailed description, examples of which are provided for illustration of the various features and aspects of the disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0011] The above and other features, aspects, and advantages of various embodiments of the present disclosure will become more apparent from the following detailed description, taken in conjunction with the accompanying drawings, in which like reference numbers represent like elements throughout. In the drawings:

[0012] FIG. 1 shows a schematic diagram of an example system in which embodiments of the present disclosure can be implemented;

[0013] FIG. 2 shows a flowchart of a method for guided questioning according to embodiments of the present disclosure;

[0014] FIG. 3A shows a schematic diagram of an interface for a method for guided questioning according to embodiments of the present disclosure;

[0015] FIG. 3B shows a schematic diagram of determining a key point according to embodiments of the present disclosure;

[0016] FIGS. 4A-4B show iterative process diagrams for a method for guided questioning according to embodiments of the present disclosure;

[0017] FIG. 5A shows a schematic diagram of collecting a question according to embodiments of the present disclosure;

[0018] FIG. 5B shows a schematic diagram of an interface for a method for guided questioning according to embodiments of the present disclosure;

[0019] FIGS. 5C-5J show further schematic diagrams of interfaces for a method for guided questioning according to embodiments of the present disclosure;

[0020] FIG. 6 shows a schematic diagram of responding to a user reply according to embodiments of the present disclosure;

[0021] FIG. 7 shows a schematic block diagram of an example apparatus according to some embodiments of the present disclosure; and

[0022] FIG. 8 shows a block diagram of an example device that can be used to implement embodiments of the present disclosure. DETAILED DESCRIPTION

[0023] Embodiments of the present disclosure will be described below in greater detail with reference to the accompanying drawings. While certain embodiments of the present disclosure are shown in the drawings, it is understood that the present disclosure can be embodied in various forms and should not be interpreted as being limited to the embodiments set forth herein; rather, these embodiments are provided so as to facilitate a more thorough and complete understanding of the present disclosure. It is understood that the drawings and embodiments of the present disclosure are for exemplary purposes only and are not intended to limit the scope of protection of the present disclosure.

[0024] Current problem-solving products (or solutions) require users (e.g., students) to first raise confusions about the problem, and then the problem-solving products (e.g., a problem-reading software or a problem-reading device integrated with the problem-reading software) answer the users' confusions. The effective application of such products requires that the user must be able to accurately recognize matters that the user does not know and clearly express the matters as confusions. However, in many cases, especially when the user is faced with a problem (e.g., a question) that the user does not understand, the user does not know which confusions should be raised to solve the problem, thereby reducing the efficiency of solving the problem and the user experience. Other problem-solving products (or solutions) only record the content of the interaction with the user, resulting in a large amount of text data in the process of solving the problem, and the user cannot quickly review the key points of the previous interaction.

[0025] In view of this, embodiments of the present disclosure provide a method for guided questioning. The method automatically generates questions about the subject of the question by using a language model, without the user needing to accurately express the question in advance, but rather the user answers the generated questions, and then the language model determines whether the user's answer contains the answer to the subject of the question. If not, the questions can continue to be generated to guide the user to solve the subject of the question. During the process of asking questions, key points are generated and recorded according to the content of the questions and the content of the answers corresponding to the questions. Therefore, according to the method of the embodiments of the present disclosure, the user is actively asked questions about the subject of the question, which can help the user select a suitable angle of attack for the subject of the question when the user is not clear about the subject of the question, guide the user to solve the subject of the question through the questioning method, and allow the user to review the key information of the historical interaction content at any time, thereby improving the efficiency of solving the problem.

[0026] Embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings, in which FIG. 1 shows a schematic diagram of an example environment 100 in which embodiments of the present disclosure can be implemented. The example environment 100 includes computing devices 110 and 120. The computing device 110 can be deployed with a language model (e.g., a large language model) that is trained on a large set of text data and is capable of understanding and generating human language. The computing device 120 is also shown in FIG. 1. In some embodiments, the computing device 120 communicates with the computing device 110 through a network 130. The network 130 can include a wired network, a wireless network, or a combination thereof, to provide communication between the computing device 120 and the computing device 110. In some embodiments, the computing device 120 can be connected to the computing device 110 through a data line, and the present disclosure does not limit the connection manner between the computing device 110 and the computing device 120.

[0027] The computing devices 110 and 120 can include, but are not limited to, a personal computer, a server computer, a handheld or laptop device, a mobile device such as a mobile phone, a personal digital assistant (PDA), a media player, etc., a multi-processor system, a consumer electronic product, a wearable electronic device, a smart home device, a mini computer, a mainframe computer, an edge computing device, a distributed computing system including any of the above systems or devices, etc.

[0028] An application (e.g., a client program) for invoking a language model to generate guided questions can be installed in the computing device 120. Taking the system 100 in FIG. 1 as an example, the computing device 120 can communicate with the computing device 110 through the network 130, and send a request for explaining a topic to the language model in the computing device 110. The language model in the computing device 110 can generate questions in response to the request from the computing device 120 and provide the generated questions to the computing device 120.

[0029] In some embodiments, the computing device 120 can invoke an image sensor (e.g., a camera) to capture an image 132 of the page 140 containing the title of the question. In some embodiments, the computing device 120 performs text recognition on the image 132 to extract textual content as the title 134 of the question. In some embodiments, upon recognizing the title 134 of the question, the title 134 can be presented in an interface (e.g., a main interface) of the application and the user can confirm whether the title 134 needs to be explained. The interface is further provided with an interface 136 for explaining the title. The user can trigger the interface 136 for explaining the title to send a request for explaining the title to the computing device 110 through the computing device 120. In some embodiments, the display interface of the computing device 120 jumps from an upper interface to a lower interface, and the computing device 110 generates a question for the title 134 using the language model and transmits the question to the computing device 120 through the network 130 and displays the question in the interface. In some embodiments, the user can provide an answer to the question to the computing device 110 through the computing device 120 and the network 130. In some embodiments, the computing device 120 determines a key point 138 for the title 134 based on the first question and the answer corresponding to the first question using the language model. In some embodiments, the computing device 120 obtains the answer to the question provided by the user and generates a new question for the title 134 using the language model if the answer does not include the target answer corresponding to the title 134. In some embodiments, the computing device 120 displays the new question and the key point 138 in the interface for the user to view. In some embodiments, the computing device 110 transmits the new question to the computing device 120 through the network 130 for the user to answer. By continuously asking such questions, the user is guided to solve the title 134 step by step.

[0030] It can be understood that although the language model is deployed in the computing device 110 in FIG. 1, the language model can be split into multiple sub-models and each sub-model can be deployed in a corresponding computing device according to actual needs, achieving distributed deployment, for example, to support large-scale models. Accordingly, the corresponding computing device can send the generated information to the computing device 120 and display the information in response to a request or a command from the computing device 120.

[0031] Further, although it is shown in FIG. 1 that the language model is deployed separately from the computing device 120, it can be understood that, as the language model is lightened, the language model can also be deployed locally at the computing device 120, so as to be able to respond to the user's request more quickly and generate corresponding information. According to the method of the embodiments of the present disclosure, the user is actively initiated to ask questions for the topic, which can help the user to select a suitable angle of attack for the topic when the user is not clear about the key points of the question, guide the user to solve the topic by the way of asking questions, and allow the user to review the key information of the historical interaction content at any time, thereby improving the efficiency of solving the problem.

[0032] The block diagram of the example environment 100 in which the embodiments of the present disclosure can be implemented is described above in combination with FIG. 1. The method for guided questioning according to the embodiments of the present disclosure is described below in combination with FIG. 2. FIG. 2 shows a flowchart of the method 200 for guided questioning according to the embodiments of the present disclosure. The method 200 can be executed at the computing device 120 in FIG. 1 and any suitable computing device. It should be understood that the numbering in the flowchart of the method 200 does not represent the order of execution of these steps, some or all of these steps can be executed in parallel, or the order of execution can be interchanged, and the present disclosure does not limit this. In addition, the method 200 in FIG. 2 can also include additional steps not shown and / or can omit the steps shown, and the scope of the present disclosure is not limited in this respect.

[0033] In block 202, the computing device 120 can generate a first question for the topic using a language model. The language model may, for example, be a large language model (LLM). The large language model is a deep learning model trained using a large amount of text data, and can generate natural language text or understand the meaning of language text. The first question is for the topic 134, that is, the first question is a question generated around the topic 134, and the answer corresponding to the first question can be a solution condition mined from the topic 134, etc. Therefore, the first question can guide the user to think in the direction of the answer corresponding to the topic 134. As shown in FIG. 1, the topic 134 can be "Two cars A and B start from two places A and B at the same time, and meet at a distance of 12 kilometers from the midpoint. Given that the speed ratio of the two cars A and B is 4:7, how far are the two places A and B apart?" The first question can be "Are the times traveled by the two cars A and B the same when they meet?"

[0034] In some embodiments, an application program (for example, a client program) that calls the language model to generate the question can be installed in the computing device 120, so as to execute the method for guided questioning according to the embodiments of the present disclosure.

[0035] In some embodiments, the computing device 120 can launch an application and display a question-asking control in an interface after the application is launched (e.g., a launch interface or a main interface of the application). The user can start a question-asking by clicking the question-asking control. For example, the question-asking control can be a button control displayed in a floating manner in the current interface. The disclosure does not limit the specific implementation of the triggering operation.

[0036] The computing device 120 can display a dialog interface for guided questioning in response to the triggering operation of the question-asking control by the user. In the dialog interface, an input area and an information display area can be included. The input area can receive the input information of the user and transmit the input information (i.e., a reply) to the language model in the computing device 110 to generate a new question by the model 112 according to the input information of the user and send the new question to the computing device 120. The computing device 120 can display the received new question in the information display area to facilitate the user to browse.

[0037] In block 204, the computing device 120 obtains a first reply of the user to the first question. The first reply is a response of the user to the first question. As shown in FIG. 1, the first reply can be “same”. The first reply can also be “not same”, “unknown”, and the like. In some embodiments, the user can reply the first reply by voice input, and can also reply the first reply by text input.

[0038] In block 206, the computing device 120 determines a key point for the subject of the question using the language model according to the first question and the answer corresponding to the first question. As an example, the current question (i.e., the first question) is “Are the times taken by the two cars of A and B to meet the same?” The answer corresponding to the question is same. Then the key point can be determined as “Are the times taken by the two cars of A and B to meet the same?” using the language model. In some embodiments, the key point can be a combination of text data and image data, i.e., the key point is displayed in a combination of text and image.

[0039] In block 208, the computing device 120 generates a second question for the subject of the question using the language model in response to the first reply not including the target answer corresponding to the subject of the question. The reply of the user (i.e., the first reply) can be correct, incorrect, and the like, as long as it does not include the target answer corresponding to the subject of the question, which means that the user has not correctly solved the question 134. Therefore, it is necessary to continue to generate a question for the question 134 using the language model.

[0040] In block 210, the key point and the second question are presented. For example, the key point and the second question are presented in an interface. Although a new question (i.e., the second question) needs to be generated here, this is not the only operation performed when it is determined that the first reply does not include the target answer corresponding to the target topic. That is, before the new question is generated, some other operations can also be performed according to the characteristics and content of the first reply. For example, if the first reply is a user's confusion "I don't know, why do you want to know the speed ratio?", the confusion can be answered first and then a new question is generated. According to the method of the embodiments of the present disclosure, the user is actively initiated to ask questions for the target topic, which can help the user to select a suitable angle of attack for the target topic when the user is not clear about the key point of the question, guide the user to solve the target topic through the question, and allow the user to review the key information of the historical interaction content at any time, thereby improving the efficiency of solving the problem.

[0041] In some embodiments, when generating the new question, the historical question (e.g., the first generated question) and / or the historical reply (e.g., the reply to the first generated question) are obtained to generate the new question. This embodiment further includes determining the new question based on the historical question and / or the historical reply using a language model.

[0042] In this way, the computing device 120 can use the language model to generate the new question according to the context content (i.e., the historical question and / or the historical reply). In this way, the language model can know which knowledge the user has mastered and which knowledge needs to be mastered in order to solve the target problem, so that the new question generated with reference to the context content can have strong coherence and guidance, which is conducive to helping the user to solve the target problem.

[0043] As an example, the historical question obtained by the computing device 120 includes "Are the times of the two cars of A and B when they meet the same?" The historical reply obtained includes "The times are the same!" and the new question generated by the computing device 120 using the language model can be "What is the relationship between the speed ratio and the distance ratio when the times are the same?" It can be seen that the new question is a further extension of the historical question, and the extension direction is towards a direction conducive to solving the target problem. This is conducive to helping the user to solve the target problem.

[0044] FIG. 3A illustrates a schematic diagram of an interface 300 for guided questioning, according to an embodiment of the present disclosure. In this embodiment, the interface 300 includes a plurality of controls. In some embodiments, the interface 300 is provided with at least one of a voice interface 302 for receiving voice input from a user, a quit interface 306 for quitting the first interface 300, a feedback interface 308 for receiving user-provided opinions, and a prompt interface 304 for providing a prompt to the user. If the user triggers the prompt control 304, the computing device 120 can generate a prompt for the current question using the language model and present the prompt in the interface 300.

[0045] During the application of the method of the embodiments of the present disclosure, multiple rounds of interactions between the computing device 120 and the user can be generated. For the sake of brevity, in some embodiments, these contextual contents can not be presented immediately, but a record presentation control 310 is provided on the interface 300, and when the user triggers the record presentation control 310, the contextual contents including the historical questions and the historical answers are presented. For example, the first question, the first answer, and the second question are presented in the interface. In some embodiments, the interface 300 further includes a topic presentation area 312 for presenting the topic 134. In some embodiments, the interface 300 further includes a key point presentation area 314 for presenting the determined key points. In some embodiments, the interface 300 further includes an interaction area 316. In this interaction area 316, the latest generated question is kept presented for the user to view the question at any time during the process of thinking of an answer. That is, when the current question is generated, the current question is presented in the interaction area 316. After a new question is generated, the previous question is replaced with the newly generated question. This is beneficial for the user to quickly locate the latest question.

[0046] FIG. 3B illustrates a schematic diagram of determining key points, according to an embodiment of the present disclosure. At 318, a language model is used to determine whether the current question (i.e., the question just presented to the user) is related to the topic (i.e., the topic title). If the current question is not related to the topic, it is not necessary to summarize key points for the question and the answer, and the process ends at 320. If the current question is related to the topic, a question summary for the current question is determined at 322 using the language model, and an answer summary for the answer corresponding to the current question is determined at 324. The key points are determined at 326 according to the question summary and the answer summary. In this embodiment, the key points can be recorded while the guided questioning is generated, which can serve a similar function as board writing, and is beneficial for the user to quickly review the key points of the historical questions and the historical answers.

[0047] As an example, for example, the current question is "Are the times taken by the two cars to meet the same?", and the answer is the same. Then the language model can be used to determine the question summary as "The time taken by the two cars to meet", and the answer summary as "the same", so as to determine the key point as "The time taken by the two cars to meet is the same?" according to the question summary and the answer summary. In some embodiments, the key point can be a combination of text data and image data.

[0048] FIG. 3C shows a schematic diagram of a key point according to an embodiment of the present disclosure. The edge length of the cube 330 is a, and the edge length of the cube 332 is 3a. The proportional diagrams of the cube 330 and the cube 332 in the key point display area 314 are shown respectively. And the text explanation is provided below, so that the user can fully perceive the difference in surface area of the two cubes. This is helpful for the user to quickly grasp the key point.

[0049] For ease of understanding, the embodiment shown in FIG. 2 only illustrates an example of generating a first question and determining whether to generate a second question according to the user's reply to the question, but this does not limit the method to only including the generation of the two questions or only the generation of the first question. In fact, from the above it can be inferred that as long as the user has not replied to the correct answer to the question, the above question and reply process can be iteratively performed, that is, the user is constantly asked questions and the user's replies are obtained, and the user is guided to think to get the correct answer through multiple questions. That is, the above process can be iteratively performed.

[0050] FIG. 4A shows an iterative process diagram of the method for guided questioning according to an embodiment of the present disclosure. At 402, the topic that the user wants to explain is obtained. At 404, the language model is called and the topic is input into the language model. At 406, a question for the topic is generated, for example, which can be provided to the user in the tone of a teacher. At 408, the user's reply to the answer to the topic is obtained. At 410, it is determined whether the user's reply includes the answer corresponding to the topic. If it does, the process ends at 412. If not, the language model is called back at 404 and a new question is generated at 406, entering the iterative process.

[0051] FIG. 4B illustrates another iterative process diagram of the method for guided questioning, according to an embodiment of the present disclosure. As shown in FIG. 4B, at 414, the question is acquired by the user using the computing device 120 for capturing an object including the question using an image sensor to obtain a question image, and by extracting textual content in the question image. For example, the question image can be subjected to optical character recognition to determine the textual content. FIG. 5A illustrates a schematic diagram of capturing a question, according to an embodiment of the present disclosure. As shown in FIG. 5A, at 502, the captured question image is presented. At 504, the question and answer identified from the question image are presented. At 506, the analysis process and detailed answer to the question are presented. At 508, a question explanation interface is presented. The user can jump to the question explanation interface by triggering the question explanation interface.

[0052] Referring back to the embodiment of FIG. 4B, at 416, the question for the question is generated by the computing device 120 using a language model based on the contextual content. For example, the computing device 120 can input the textual content of the question into a local language model or a language model on the computing device 110 to generate the question. FIG. 5B illustrates a schematic diagram of an interface for the method for guided questioning, according to an embodiment of the present disclosure. The interface is the question explanation interface, which is similar to the interface 300. When the question is first generated, “Explanations are being prepared… ” can be presented in the interactive area 316 to prompt the user to be patient. In this process, the computing device 120 can invoke the language model to generate the first question for the question 134. The question generated at 416 is the current question, which can represent not only the first question, but also the subsequently generated questions, depending on the stage of the question explanation. FIG. 5C illustrates another schematic diagram of an interface for the method for guided questioning, according to an embodiment of the present disclosure. After obtaining the question at 416, the current question “Are the times taken by the two cars to meet the same?” is presented in the interactive area 316 of the interface.

[0053] Referring back to the embodiment of FIG. 4B, at 418, the reply to the question in reply to the question is acquired. For example, the user can input the voice content representing the reply through the voice interface 302. FIG. 5D illustrates another schematic diagram of an interface for the method for guided questioning, according to an embodiment of the present disclosure. The user can click the voice interface 302 to input the voice data. After obtaining the voice data, the computing device 120 can process the voice data into textual data as the reply.

[0054] After acquiring the reply, the reply can be displayed on the interface. FIG. 5E illustrates another schematic diagram of an interface for the method for guided questioning, according to an embodiment of the present disclosure. As can be seen, the reply of the user is “No!”

[0055] Referring back to the embodiment of FIG. 4B, at 420, a key point “The time taken by both cars A and B to meet is the same.” is generated based on the question summary for the question and the answer summary for the answer corresponding to the question.

[0056] FIG. 5F shows another schematic view of an interface of the method for guided questioning, according to an embodiment of the present disclosure. In conjunction with the aforementioned example, the user’s reply is “No way!” The reply does not include the target answer and the reply does not include the answer corresponding to the current question (which is “the same”). At this point, the computing device 120 can announce the answer corresponding to the current question “the same” to the user, for example, in a voice played manner.

[0057] Referring back to the embodiment of FIG. 4B, at 422, it is determined whether the reply includes the target answer. If the reply includes the target answer, it means that the user has directly solved the question through the question, at 424, the iteration process can be directly ended. If the reply does not include the target answer, it returns to 416 to continue to generate a new question. The key point and the newly generated question are displayed in the interface. Thus, one iteration process is completed. FIG. 5G shows another schematic view of an interface of the method for guided questioning, according to an embodiment of the present disclosure. It shows a new question “Since the time is the same, what is the relationship between the speed ratio and the distance?”

[0058] FIG. 5H shows another schematic view of an interface of the method for guided questioning, according to an embodiment of the present disclosure. In FIG. 5H, it can be seen that through at least 5 iterations, the user finally solves the question and summarizes 5 key points, which are “The time taken by both cars A and B to meet is the same.” “In the same time, the speed ratio is 4:7 and the distance ratio is also 4:7” “The distance traveled by car B is more than half of the total distance by 12 kilometers, and the distance traveled by car A is less than half of the total distance by 12 kilometers.” “The distance difference between the two cars is 24 kilometers.” “The distance ratio is 4:7, the total distance is considered as 4+7=11 parts, the distance difference is 7-4=3 parts, 3 parts is 24 kilometers, one part is 24÷3=8 kilometers, and the total distance is 8x11=88 kilometers.”

[0059] After the iteration is completed, in some embodiments, an interface is presented, which provides a plurality of scoring options for selection by the user. FIG. 5I shows another schematic view of an interface of the method for guided questioning, according to an embodiment of the present disclosure. In FIG. 5I, it can be seen that after the user replies with the correct answer, the user is invited to score the explanation and three scoring options are provided, which are “good”, “average” and “bad”. After the user gives a selection, the computing device 120 can transmit the user’s selection to the developer for optimizing the language model.

[0060] At the end of the iteration, in some embodiments, if the user's reply includes the target answer corresponding to the question, positive feedback is provided. The embodiment further includes presenting an interface, where the interface is provided with an option indicating that the language model is used to regenerate a third question for the question 134 for the question 134. FIG. 5J shows another schematic diagram of an interface for the method of guided questioning according to an embodiment of the present disclosure. As can be seen in FIG. 5J, after the user replies with the correct answer, an option of "Learn again" is provided for reinitiating the iterative process as shown in FIG. 4B.

[0061] The above describes the process and interface changes for guiding the user to solve the question through multiple iterations in the form of questions. Those of ordinary skill in the art can understand that various functions or components can be combined with each other or appropriately transformed to obtain other embodiments.

[0062] FIG. 6 shows a schematic diagram of responding to the user's reply according to an embodiment of the present disclosure. At 602, the computing device 120 generates a current question using the language model. At 604, the reply of the user's reply to the current question is obtained. The computing device 120 can have different responses to the specific content of the reply.

[0063] The result of the judgment at 606 is no and the result of the judgment at 610 is yes, then the reply provided by the user includes the answer corresponding to the current question, but does not include the answer corresponding to the question 134. For example, the current question can be "Are the times of the two cars A and B the same when they meet?" The user's reply can be "Yes". Because the user does not provide the answer corresponding to the question 134, it is necessary to continue to generate questions around the question 134 at 602, that is, to continue the iterative process of the question. In some embodiments, the computing device 120 can provide positive feedback "You are right!" and optionally provide an image and voice prompt indicating affirmation, and generate a new question using the language model, if the first reply of the user includes the answer corresponding to the first reply, but does not include the target answer. The new question can be "Since the times are the same, what is the relationship between the speed ratio and the distance for this?"

[0064] The result of the judgment at 606 is yes, then the reply provided by the user includes the answer corresponding to the question, but does not include the answer corresponding to the question. For example, the current question can be "Are the times of the two cars A and B the same when they meet?" The user's reply can be "The total distance is 88 kilometers." In this case, because the current question provides a suitable angle of attack for solving the question under the guidance of the current question, the user thinks based on this and gets the correct answer, so there is no need for further guided questioning, and the iterative process can be ended. At 608, the question presentation process is ended.

[0065] If the result of the determination at 606 is no and the result of the determination at 610 is no, the user's reply does not include the answer corresponding to the current question and does not include the answer corresponding to the question 134. For example, the current question can be "Are the times taken by the two cars to meet the same?" and the user's reply can be "No, they are not the same!" Since the user does not provide the answer corresponding to the question 134, and the reply is determined not to be the user's question at 612, it is necessary to provide negative feedback and optionally provide negative images and voice prompts at 616, present the answer corresponding to the current question, and return to 602 to continue generating new questions around the question 134, i.e., continue the iteration of the question process.

[0066] In some embodiments, if the user's reply is determined not to include the answer corresponding to the current question, the reply is checked at 612 using a language model to determine whether it belongs to the user's question. If the reply belongs to the user's question at 614, the answer corresponding to the reply is determined and presented. Returning to 602, a new question for the question is generated based on the reply. This embodiment allows the user's personalized question to be solved according to the user's needs during the question process, which is helpful to help the user solve the question.

[0067] On the basis of this embodiment, if the user's question is also related to the question, the key points of the corresponding answer can be determined and recorded in the interface. This is helpful for the user to review the solution at any time.

[0068] FIG. 7 shows a schematic block diagram of an example apparatus 700 according to some embodiments of the present disclosure. The apparatus 700 can be implemented by software, hardware, or a combination of both. As shown in FIG. 7, the apparatus 700 includes a first generation module 710, a first acquisition module 720, and a second generation module 730.

[0069] In some embodiments, the first generation module 710 can generate a first question for a question title using a language model. The first acquisition module 720 can acquire a first reply to the first question from a user. The second generation module 730 can generate a second question for the question title using a language model in response to the first reply not including a target answer corresponding to the question title.

[0070] The apparatus 700 of FIG. 7 can be used to implement the above-described processes in conjunction with FIGS. 1-6, and for brevity, will not be described again here.

[0071] The division of modules or units in the embodiments of the present disclosure is illustrative, and is only a logical function division. When actually implemented, another division manner can be used. In addition, each functional unit in the disclosed embodiments can be integrated into one unit, or can be physically separated, or two or more units can be integrated into one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0072] FIG. 8 illustrates a block diagram of an example device 800 that can be used to implement embodiments of the present disclosure. It should be understood that the device 800 illustrated in FIG. 8 is merely an example and should not be construed as limiting the functionality and scope of the implementations described herein. For example, the device 800 can correspond to the computing device 120 described herein with respect to FIG. 1, and can be used to perform the processes of FIGS. 1-6 described above.

[0073] As shown in FIG. 8, the device 800 is in the form of a general-purpose computing device. Components of the computing device 800 can include, but are not limited to, one or more processors or processing units 810, a memory 820, a storage device 830, one or more communication units 840, one or more input devices 850, and one or more output devices 860. The processing unit 810 can be a real or virtual processor and is capable of executing various processing in accordance with programs stored in the memory 820. In a multi-processor system, multiple processing units execute computer-executable instructions in parallel to improve the parallel processing capability of the computing device 800.

[0074] The computing device 800 typically includes a plurality of computer storage media. Such media can be removable, non-removable, or a combination thereof and can be accessed by the computing device 800. The memory 820 can be volatile (such as a register, cache, or random access memory (RAM)), non-volatile (such as read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), or flash memory), or some combination thereof. The storage device 830 can be a removable or non-removable medium and can include machine-readable media, such as a flash drive, disk, or any other medium that can be used to store information and / or data (e.g., training data for training) and that can be accessed by the computing device 800.

[0075] The computing device 800 can further include additional removable / non-removable, volatile / non-volatile storage devices. Although not shown, a floppy disk drive for reading from or writing to a removable, non-removable magnetic disk (e.g., a "floppy disk"), and an optical disk drive for reading from or writing to a removable, non-removable optical disk (e.g., a CD-ROM) can be provided. In such instances, each can be connected to the bus by one or more data media interfaces. The memory 820 can include a computer program product 825 having one or more program modules configured to carry out the various methods or actions of the various implementations of the present disclosure.

[0076] The communication unit 840 enables communications with other computing devices over a communication medium. Additionally, the functionality of the components of the computing device 800 can be implemented in a single computing cluster or a plurality of computer machines that are capable of communicating over a communication connection. Thus, the computing device 800 can operate in a networked environment using logical connections to one or more other servers, network personal computers (PCs), or another network nodes in the networking environment.

[0077] The input device 850 can be one or more input devices, such as a mouse, a keyboard, a trackball, etc. The output device 860 can be one or more output devices, such as a display, a speaker, a printer, etc. The computing device 800 can also communicate with one or more external devices (not shown) such as a storage device, a display device, etc. through the communication unit 840, as needed, one or more devices that enable a user to interact with the computing device 800, or any devices (e.g., a network card, a modem, etc.) that enable the computing device 800 to communicate in a network environment. Such communication can be carried out via an Input / Output (I / O) interface (not shown).

[0078] According to example implementations of the present disclosure, a computer readable storage medium is provided having computer executable instructions stored thereon, where the computer executable instructions are executed by a processor to implement the methods described above. According to example implementations of the present disclosure, a computer program product is also provided that is tangibly stored on a non-transitory computer readable medium and includes computer executable instructions, where the computer executable instructions are executed by a processor to implement the methods described above. According to example implementations of the present disclosure, a computer program product is provided having a computer program stored thereon, which when executed by a processor implements the methods described above.

[0079] The computer readable program instructions can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks. These computer readable program instructions can also be stored in a computer readable storage medium that can include a non-transitory computer readable storage medium that can be a computer- readable storage medium having no data, programs, program modules, e.g., instructions for operation, or digital content stored thereon or therein for a short time or not at all.

[0080] These computer readable program instructions can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks. These computer readable program instructions can also be stored in a computer readable storage medium that can include a non-transitory computer readable storage medium that can be a computer- readable storage medium having no data, programs, program modules, e.g., instructions for operation, or digital content stored thereon or therein for a short time or not at all.

[0081] These computer readable program instructions can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks. These computer readable program instructions can also be stored in a computer readable storage medium that can include a non-transitory computer readable storage medium that can be a computer- readable storage medium having no data, programs, program modules, e.g., instructions for operation, or digital content stored thereon or therein for a short time or not at all.

[0082] The computer readable program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0083] Having described various implementations of the disclosure above, the descriptions are not exhaustive and do not limit the disclosure to the disclosed implementations. Numerous modifications and adaptations thereof will be apparent to those skilled in the art without departing from the scope and spirit of the described implementations. The scope of the disclosure includes all of the implementations described above and equivalents thereto. Language used in this specification should not be used to limit the scope of the claims.

Claims

1. A method for guided questioning, comprising: generating, using a language model, a first question for a topic of interest; obtaining a first response to the first question from a user; determining, using the language model, a key point for the topic of interest from the first question and an answer corresponding to the first question; in response to the first response not including a target answer corresponding to the topic of interest, generating, using the language model, a second question for the topic of interest; and presenting the key point and the second question.

2. The method of claim 1, wherein generating, using the language model, a second question for the topic of interest comprises: obtaining the first question and the first response; and determining, using the language model, the second question based on the first question and the first response.

3. The method of claim 2, further comprising: obtaining a second response to the second question from a user; determining, using the language model, whether the second response includes an answer corresponding to the second question; in response to the second response being determined to include an answer corresponding to the second question, providing positive feedback; and in response to the answer corresponding to the second question being different from the target answer, generating, using the language model, a third question for the topic of interest.

4. The method of claim 3, further comprising: in response to the second response being determined to not include an answer corresponding to the second question, providing negative feedback and presenting an answer corresponding to the second question; and generating, using the language model, a fourth question for the topic of interest.

5. The method of claim 3, further comprising: in response to the second response being determined to not include an answer corresponding to the second question, performing, using the language model: checking whether the second response belongs to a user question; in response to the second response belonging to the user question, determining and presenting an answer corresponding to the second response; and generating, based on the second response, a fifth question for the topic of interest.

6. The method of claim 5, further comprising: in response to the second response belonging to the user question, determining, using the language model, whether the second response is related to the topic of interest; in response to the second response being related to the topic of interest, determining, using the language model, a first question summary for the second response and a first answer summary for an answer corresponding to the second response; and determining a first key point from the first question summary and the first answer summary.

7. The method of claim 1, wherein determining, using the language model, a key point for the topic of interest from the first question and an answer corresponding to the first question comprises: determining, using the language model, whether the first question is related to the topic of interest; in response to the first question being related to the topic of interest, determining, using the language model, a second question summary for the first question and a second answer summary for an answer corresponding to the first question; and ​ ​ ​ ​ ​ ​ determine a second key point as the key point for the topic of interest based on the second question summary and the second answer summary. 8.The method of claim 7, wherein the second answer summary comprises at least one of text and a picture. 9.The method of claim 1, further comprising: capturing an object image including an object of interest using an image sensor; and obtaining the topic of interest by extracting text content in the object image. 10.The method of claim 9, wherein obtaining the topic of interest by performing text processing on the object image comprises: performing optical character recognition on the object image to determine the text content. 11.The method of claim 1, further comprising: providing a first interface, wherein the first interface is configured with at least one of: a voice interface for receiving voice input from the user; a quit interface for quitting the first interface; a feedback interface for receiving an opinion provided by the user; and a prompt interface for providing a prompt to the user. 12.The method of claim 11, further comprising: recording the first question in the first interface in response to generating the first question; and replacing the first question with the second question in the first interface in response to generating the second question. 13.The method of claim 12, wherein the first interface further comprises a record presentation control, and the method further comprises: presenting the first question, the first reply, and the second question in the first interface in response to the record presentation control being triggered. 14.The method of claim 1, further comprising: providing positive feedback in response to the first reply including a target answer corresponding to the topic of interest; and presenting a second interface, wherein the second interface is configured with a first option indicating to regenerate a third question for the topic of interest using the language model. 15.The method of claim 14, further comprising: presenting a third interface, wherein the third interface is provided with a plurality of scoring options for selection by the user. 16.An electronic device comprising: at least one processing unit; at least one memory coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit, the instructions when executed by the at least one processing unit cause the electronic device to perform the method according to any one of claims 1 to 15. 17.A computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the method according to any one of claims 1 to 15. 18.A computer program product having stored thereon a computer program which, when executed by a processor, performs the method according to any one of claims 1 to 15. ​ ​

Images