Interaction method and device, medium, equipment and computer program product

Abstract

The invention relates to an interaction method and device, a medium, equipment and a computer program product, and the method comprises the steps: responding to a received content viewing operation of a user on a target range of target content, determining a target task corresponding to the target range, and enabling the target task to comprise a target question and a target answer; displaying the target question and display content corresponding to the target range in a display interface; if it is determined that the user receives the target task, obtaining answer candidate information of the target question in the display process of the display content corresponding to the target range; and in response to completion of display of the display content corresponding to the target range, determining an answer result of the user to the target task based on the target answer and the answer candidate information.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular, to an interaction method, apparatus, medium, device, and computer program product. Background Art

[0002] With the development of the times, more and more users read or watch videos through electronic reading devices or video browsing applications. However, during the reading process, users may reduce their reading concentration due to problems such as too long content or low content continuity, resulting in a poor reading experience. Summary of the invention

[0003] This summary is provided to introduce concepts in a brief form that will be described in detail in the detailed description below. This summary is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0004] In a first aspect, the present disclosure provides an interaction method, the method comprising:

[0005] In response to receiving a content viewing operation of a target range of target content by a user, determining a target task corresponding to the target range, wherein the target task includes a target question and a target answer;

[0006] Displaying the target problem and the display content corresponding to the target range in the display interface;

[0007] If it is determined that the user takes up the target task, in the process of displaying the display content corresponding to the target range, obtaining the candidate answer information of the target question;

[0008] In response to completion of display of the display content corresponding to the target range, a result of the user's answer to the target task is determined based on the target answer and the answer candidate information.

[0009] In a second aspect, the present disclosure provides an interactive device, the device comprising:

[0010] A first determination module, configured to, in response to receiving a content viewing operation of a user on a target range of target content, determine a target task corresponding to the target range, wherein the target task includes a target question and a target answer;

[0011] A first display module, used to display the display content corresponding to the target problem and the target scope in a display interface;

[0012] An acquisition module, configured to acquire candidate answer information of the target question during the display process of the display content corresponding to the target range if it is determined that the user has received the target task;

[0013] The second determination module is used to determine the user's answer result to the target task based on the target answer and the answer candidate information in response to the display content corresponding to the target range being displayed.

[0014] In a third aspect, the present disclosure provides a computer-readable medium having a computer program stored thereon, which implements the steps of the method described in the first aspect when the computer program is executed by a processing device.

[0015] In a fourth aspect, the present disclosure provides an electronic device, including:

[0016] a storage device having a computer program stored thereon;

[0017] A processing device is used to execute the computer program in the storage device to implement the steps of the method described in the first aspect.

[0018] In a fifth aspect, the present disclosure provides a computer program product, including a computer program, which implements the steps of the method described in the first aspect when executed by a processor.

[0019] Therefore, in the above technical solution, the user can receive the corresponding target task during the content browsing process, so that the user can browse in combination with the target question during the content interaction process, so as to find the answer corresponding to the target question during the browsing process and submit it to verify the user's answer result. Therefore, through the above technical solution, based on the target task, the user's browsing interest and browsing concentration can be effectively improved, and the interaction diversity of the user during the content browsing process can be improved. In addition, when generating the target task, the display content corresponding to the target range is generated. On the one hand, the correlation between the generated target task and the user's current browsing content can be improved, and the user's interest in interaction can be improved. On the other hand, the efficiency of task generation can be improved, and the user experience can be improved.

[0020] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] 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 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. In the drawings:

[0022] Figure 1 It is a flowchart of an interaction method provided according to an embodiment of the present disclosure.

[0023] Figure 2It is a schematic diagram of a display interface provided according to an embodiment of the present disclosure.

[0024] Figure 3 It is a schematic diagram of a display interface provided according to another embodiment of the present disclosure.

[0025] Figure 4 It is a schematic diagram of a display interface provided according to another embodiment of the present disclosure.

[0026] Figure 5 It is a schematic diagram of a display interface provided according to another embodiment of the present disclosure.

[0027] Figure 6 It is a schematic diagram of a display interface provided according to another embodiment of the present disclosure.

[0028] Figure 7 It is a timing diagram of an interaction method provided according to an embodiment of the present disclosure.

[0029] Fig. 8A and 8B It is an interactive schematic diagram of a task generation model provided according to an embodiment of the present disclosure.

[0030] Fig. 9 It is a block diagram of an interaction device provided according to an embodiment of the present disclosure.

[0031] Fig.10 A schematic diagram of the structure of an electronic device suitable for implementing the embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0032] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0033] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0034] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0035] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0036] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0037] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0038] It is understandable that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, scope of use, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0039] For example, in response to receiving an active request from a user, a prompt message is sent to the user to clearly prompt the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, application, server, or storage medium that performs the operation of the technical solution of the present disclosure according to the prompt message.

[0040] As an optional but non-limiting implementation, in response to receiving an active request from the user, the prompt information may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form. In addition, the pop-up window may also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0041] It is understandable that the above notification and the process of obtaining user authorization are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that meet the relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0042] At the same time, it is understandable that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws, regulations and relevant provisions.

[0043] Figure 1 As shown, it is a flowchart of an interactive method provided according to an embodiment of the present disclosure, such as Figure 1 As shown, the method may include:

[0044] In step 11, in response to receiving a content viewing operation of a user on a target range of target content, a target task corresponding to the target range is determined, where the target task includes a target question and a target answer.

[0045] The target content may be a novel text, or an audio or video content, etc. The method may be applied to a reading application or a reading terminal or a video application that supports electronic reading. Taking a novel text as an example, for each novel text in the reading library, the novel may be pre-divided into chapters to determine the range of each chapter in the novel that needs to trigger task generation.

[0046] As an example, each novel can be pre-split based on a chapter splitting model, and the chapter splitting model can be implemented based on a neural network model, by pre-labeling the novel to obtain training data, and training the neural network model based on the training data, and the chapter splitting model. Among them, for short stories, all chapters can be used as a chapter range, that is, there is no need to split the chapters. The short story can be determined based on the length division method commonly used in the novel field, such as determining whether it is a short story based on the attributes corresponding to the novel, or the number of chapters can be pre-set. If the total number of chapters in the novel is less than the number of chapters, the novel does not need to be split into chapters. If the total number of chapters in the novel is greater than or equal to the number of chapters, the novel is split into chapters to determine the multiple chapter ranges corresponding to the novel.

[0047] As an example, when a user reads a text, he or she usually reads sequentially from the beginning to the end. Therefore, this step may be to determine the target task corresponding to the target range in response to receiving the user's viewing operation on the text of the first chapter in the target range of the target novel, so as to generate the corresponding target task before entering the text reading of the target range to interact with the user.

[0048] For example, the target novel A is pre-divided into chapters to obtain its corresponding 5 chapter ranges, which are recorded as A1-A5. When the user reads the target novel, he usually enters the first page of the novel, that is, the first page of the first chapter in the first chapter range. At this time, the target task corresponding to the target range A1 can be determined.

[0049] As another example, the target content may be audio or video content, which is usually the content of an episode. In this embodiment, the content of each episode may be used as a target range to generate a target task corresponding to each episode.

[0050] In step 12, display content corresponding to the target problem and the target scope is displayed in the display interface.

[0051] As an example, the display interface may be a novel reading interface, in which the text content of the novel may be displayed, and the text content may be determined based on the user's current text reading progress. At the same time, the target question may be displayed before the text content, such as Figure 2 As shown, the target question can be displayed in area B1, and the content can be displayed in area B2 to interact with the user, so that the user can determine whether to accept the target task.

[0052] In step 13, if it is determined that the user accepts the target task, candidate answer information for the target question is obtained during the display process of the display content corresponding to the target scope.

[0053] The display content corresponding to the target range may be the content in the target content that belongs to the target range, such as the text content of chapters 1 to 3 in the novel text that belongs to the target range, or the video content of the third episode in the video content, etc. As an example, Figure 2 As shown, a collection control may be displayed in area B1, and the user may collect the target task by clicking the collection control. In response to receiving the user's click operation on the collection control, it may be determined that the user collects the target task. As another example, the user may collect the target task by double-clicking the card of the target task. In response to receiving a predetermined collection operation, it may be determined that the user collects the target task.

[0054] After the user receives the target task, he can determine the answer to the target question by browsing the content and submit it, and then the candidate answer information can be obtained.

[0055] In step 14, in response to the completion of display of the display content corresponding to the target range, the user's answer result to the target task is determined based on the target answer and the answer candidate information.

[0056] Among them, the completion of display of the display content corresponding to the target range can be the completion of the display of the latest content in the display content corresponding to the target range in terms of time sequence, such as the display of the last page within the target range in the novel, and the last play of the video within the target range in the video. At this time, it can be considered that the user has browsed all the content within the target range. In this scenario, the answer candidate information determined by the user can be matched with the target answer to determine whether the answer candidate information is the correct answer. As an example, the target answer and the answer candidate information can be input into the judgment model, and the answer result is obtained by determining whether the two match. For example, if the two match, the answer result is correct, otherwise the answer result is wrong. As another example, the display content, target question, target answer and answer candidate information can be constructed into a prompt text to call the model for judgment. Among them, the model can be implemented as a large language model. For example, taking the novel text as an example, the prompt text can be constructed in the following way:

[0057] In the novel content XXXX, the target question is: XX1, the target answer is XX2, and the user's answer candidate information is YY1. Please ask if the answer is correct, output yes / no

[0058] Then the prompt text is input into the model for interaction to obtain the output of the model, and if the output is yse, it means that the answer result is correct. Among them, the prompt text under different content types can be pre-set, and the present disclosure does not limit this.

[0059] As an example, a prompt message may be output in the display interface, and the prompt message is used to prompt the user's answer result, such as a pop-up output prompt message to prompt the user whether the answer result is correct or wrong. As another example, the result identification corresponding to the answer result may be displayed in the display interface, such as outputting an error identification or a correct identification, etc., to further improve the interactivity with the user. Among them, if the user receives the target task, based on the target answer and the answer candidate information, when determining the user's answer result for the target task, the answer candidate information is empty, that is, the user has not selected the corresponding information for submission, and there is no need for feedback prompts at this time.

[0060] Therefore, in the above technical solution, the user can receive the corresponding target task during the content browsing process, so that the user can browse in combination with the target question during the content interaction process, so as to find the answer corresponding to the target question during the browsing process and submit it to verify the user's answer result. Therefore, through the above technical solution, based on the target task, the user's browsing interest and browsing concentration can be effectively improved, and the interaction diversity of the user during the content browsing process can be improved. In addition, when generating the target task, the display content corresponding to the target range is generated. On the one hand, the correlation between the generated target task and the user's current browsing content can be improved, and the user's interest in interaction can be improved. On the other hand, the efficiency of task generation can be improved, and the user experience can be improved.

[0061] In a possible embodiment, the type of the target task is a question-and-answer type, that is, a target question is raised in the target task, and the user determines the answer to the question from the browsed content during the content browsing process.

[0062] Accordingly, in the process of presenting the presentation content corresponding to the target range, obtaining the candidate answer information of the target question may include:

[0063] During the display process of the display content corresponding to the target range, in response to receiving a selection operation of the user on the display content, the answer candidate information is determined according to Western Sydney corresponding to the display content selected by the selection operation.

[0064] The target question is generated based on the display content corresponding to the target range, so it is related to the display content corresponding to the target range, and the information for answering the target question should also be in the display content. Therefore, in this embodiment, the answer candidate information submitted by the user can be received while the user is browsing the display content corresponding to the target range.

[0065] As an example, the target content is a target novel, and the displayed content is text content;

[0066] The determining the answer candidate information according to the information corresponding to the display content selected by the selection operation includes: using the text content selected by the selection operation as the answer candidate information.

[0067] For example, Figure 2 As shown in FIG. 1 , after the user receives the target task, he or she can read in conjunction with the target question to determine the text that can answer the target question from the text he or she reads. For example, the user can turn the page to read the next page. When the user reads to the text shown in FIG. Figure 3In the interface shown, if the user sees a text that he thinks answers the target question, he can select the corresponding text. For example, the selection operation can be a long press operation. If the corresponding text can be selected by a long press operation, the selected text content can be used as the answer candidate information submitted by the user in this scenario. Another example is that the selection operation can be a double-click operation to select text, which can be set based on the actual application scenario, and the present disclosure does not limit this.

[0068] For example, the target content in the above scenario can be suspense, reasoning, solving crimes, etc. If the target content is a novel, it can be determined based on the genre type or classification attribute of the novel. Accordingly, for this type of novel, the user can receive the target task to obtain the target question while reading the novel, so that the user can find the answer to the target question during the reading process. It will not spoil the plot for the user, but can interact with the user to facilitate the user to sort out the plot, thereby improving the user's reading interest and reading efficiency.

[0069] As another example, the target content is a target video, and the display content is a video content;

[0070] The determining the answer candidate information according to the information corresponding to the display content selected by the selection operation includes:

[0071] The text corresponding to the video content selected by the selection operation is used as the answer candidate information, and the text corresponding to the video content includes text corresponding to subtitles in the video content and / or text corresponding to audio in the video content.

[0072] In the process of browsing videos, if the user believes that the video content currently browsed can be used to answer the question, the user can select part of the video content by taking a screenshot or capturing the video. If the user believes that the target question can be answered by the information in the current picture, the user can use the screenshot operation as a selection operation to obtain a video image. Furthermore, the subtitles in the video image can be extracted to obtain the text corresponding to the video image, and then obtain the answer candidate information.

[0073] As another example, if the user believes that the target question can be answered by the information in the current video clip, the user can use the video capture operation as a selection operation to obtain a video clip. Furthermore, the subtitles in the image of the video clip can be extracted to obtain the text corresponding to the video clip, and then the answer candidate information can be obtained. Alternatively, the audio in the video clip can be recognized to obtain the text, such as recognizing the audio based on speech recognition technology to obtain the corresponding text as the answer candidate information.

[0074] As a result, users can submit answers based on the browsed content during the content browsing process, which facilitates user interaction and increases user interest in content browsing.

[0075] Therefore, through the above technical solution, after the user receives the target task, the user can read in conjunction with the target question during the reading process, which is convenient for the user to sort out the plot, and the answer to the target question in the target task can be determined from the reading text during the reading process and submitted without the user performing input operations. This improves the diversity of user interaction while facilitating user operation, enhances user interest and interaction while avoiding plot reminders for the user, thereby ensuring the user's usage experience.

[0076] In another possible embodiment, if the user makes content selections multiple times, the information corresponding to the content selected by the user for the last time is used as the answer candidate information. That is, after each content selection, the information corresponding to the most recently selected content is used as the answer candidate information, so that the user can correct his or her answer during the content browsing process to meet the user's usage needs.

[0077] As another example, the steps may include:

[0078] In response to receiving the user's selection operation on the displayed content, outputting selection prompt information, the selection prompt information is used to prompt the user whether to submit the selected content. Figure 3 As shown, in response to receiving the selection operation, the selection prompt information, such as "Submit answer", can be output at the associated position of the selected content. Then, in response to receiving the user's confirmation operation on the selection prompt information, the answer candidate information is determined according to the information of the displayed content selected by the selection operation. If the user can click on the selection prompt information to trigger the confirmation operation, the answer candidate information can be determined according to the content selected by the user.

[0079] In a possible embodiment, the type of the target task is a selection type, and the target task also includes wrong answers, that is, the target question of the target task corresponds to a target answer, i.e., a correct answer, and at least one wrong answer, so as to provide the user with multiple options when the target task is displayed.

[0080] Accordingly, displaying the target problem and the display content corresponding to the target range in the display interface may include:

[0081] The target question is displayed in the display interface, and the target answer and the wrong answer are displayed as answer candidates corresponding to the target question. Figure 4As shown in C1, the target question and its corresponding answer candidates may be displayed in the display interface so that the user can select from the answer candidates.

[0082] The display content is displayed in an area below the display area of ​​the target question and the answer candidates.

[0083] By displaying the target questions and answer candidates before displaying the content, users can be reminded of their tasks before they read, thereby improving the effectiveness and convenience of users taking on tasks.

[0084] Accordingly, if it is determined that the user receives the target task, obtaining the candidate answer information of the target question during the display process of the display content corresponding to the target range may include:

[0085] In response to receiving a selection operation of the user among the answer candidates, it is determined that the user receives the target task, and the answer candidate information is determined based on the selected answer candidate.

[0086] Taking a novel text as an example, the type of the novel can be online literature, drama, story, publishing and other types of novels, which can be predetermined based on the classification attributes of the novel. The target task is based on the user's prediction of the next plot, so the answer candidate options can be displayed before the content is displayed, and the user can choose from the answer candidate options to predict the plot of the target question within the target range.

[0087] As an example, in response to receiving a user's selection operation, that is, the user predicts the plot based on the displayed target question and answer candidates, it can be considered that the user has received the target task, and the answer candidate selected by the user's selection operation is determined as the answer candidate information. As another example, each answer candidate has its corresponding candidate identifier, and the candidate identifier can also be used as the answer candidate information. When the answer result is determined later, the candidate identifier can be directly compared with the identifier corresponding to the target answer. If the two are the same, the answer result is correct, and if the two are different, the answer result is wrong, thereby further improving the efficiency of answer result determination.

[0088] Therefore, through the above technical solution, when displaying the target task, multiple answer candidates corresponding to the target question can be displayed at the same time as the target question, so that users can make plot predictions based on the answer candidates to submit answers, further improving the interactive diversity of users during the reading process and enhancing users' reading interest.

[0089] In a possible embodiment, the method may further include:

[0090] If it is determined that the user has taken up the target task, and during the display process of the display content corresponding to the target range, the user receives a closing operation on the target content, the display progress of the target content, the target question and the candidate answer information are stored, wherein the candidate answer information is empty when the user has not selected any content.

[0091] In a possible embodiment, the user may exit during the process of browsing the displayed content corresponding to the target range. For example, the user can end the browsing of the content within the target through a return operation or a close operation, that is, trigger the close operation of the target content and exit the content browsing. After it is determined that the user has received the target task, if it is determined that the user has exited reading, the display progress of the target content can be stored to mark the user's reading position so that the target content can be continued to be browsed later. At the same time, the target question and answer candidate information can also be stored to retain the target task received by the user and the answer candidate information submitted, so as to continue the target task when the user browses later. Among them, if the user does not select the corresponding content after receiving the target task, the answer candidate information is empty at this time, and the user can select the corresponding content to answer the target question during the continued browsing process.

[0092] Afterwards, in response to receiving a viewing operation of the user on the target content, the target content is displayed based on the display progress, and a display identifier corresponding to the target question is displayed.

[0093] For example, the user can select the target content again to browse, and the target content can be displayed based on the display progress to continue displaying from the position where the user last exited browsing, so that the user can continue browsing, and at the same time, the display identifier C2 corresponding to the target question is displayed in the display interface, such as Figure 5 As shown, the user can view the target question by clicking on the display logo, so that the user can understand the target question while continuing to browse and complete the target task. This is convenient for the user to quickly adjust the browsing and achieve the continuity of content connection between multiple browsings, and can also prompt the user to accept the target task, so that the user can view the target question while browsing the target content, so that the user can select the corresponding content to answer.

[0094] In a possible embodiment, the method may further include:

[0095] If it is determined that the user has received the target task, the display mark corresponding to the target question is displayed in the display interface. Among them, the target question will be displayed before the display content corresponding to the target range. When the user browses the target content, the content will be switched through page turning or sliding operations. During this process, the target question will be hidden. However, the user may forget the specific content of the target question during browsing. In this embodiment, in order to facilitate the user to view the target question, the display mark corresponding to the target question can be displayed in the display interface. As an example, the display method can be a floating display or an edge display, etc., which can be set based on the actual application scenario.

[0096] Afterwards, in response to the user's selection operation on the display mark, the target question is displayed in a display card, wherein the display card is displayed in the form of a floating layer on the upper layer of the display interface.

[0097] When the user wants to view the specific content of the target question during browsing, he can select the display mark to view the target question again, such as Figure 6 As shown. As an example, the display time of the display card can be preset, such as 3s. After the display time of the display card reaches the display time, the display card is closed and the display mark corresponding to the target question is displayed in the display interface. The closing operation of the display card can also be preset, such as a double-click operation. The user double-clicks the display card to close the display card and continue reading. A close control can also be displayed in the display card, and the user closes the display card and continues reading by clicking the close control.

[0098] Therefore, through the above technical solution, the display of the target question can be turned off during the user's reading process to avoid blocking the displayed content. At the same time, by displaying the display mark of the target question, the user can quickly and conveniently view the target question so that the user can find the answer to the target question, thereby improving the user's interactive experience.

[0099] In a possible embodiment, the method may further include:

[0100] In response to completion of display of the display content corresponding to the target range, deleting the target task;

[0101] Among them, after the display content corresponding to the target range is displayed, the user's answer to the target question can be determined, that is, the user has performed corresponding interactions based on the question, and the subsequent reading content is not related to the target question. At this time, the target task can be deleted to avoid prompting the user for the target task again.

[0102] Accordingly, in response to receiving a content viewing operation of a target range of target content by a user, determining a target task corresponding to the target range may include:

[0103] In response to receiving a content viewing operation of the target range by the user for the first time, determining the target task.

[0104] Among them, the user browses in conjunction with the target question during the content browsing process, so that the user can find the answer to the target question while increasing the user's interest. When the user has browsed the target content, the answer to the target question is usually known to the user, and there is no need to display the target question to the user. Based on this, in the present disclosure, the target task can be determined in response to the first reception of the user's content viewing operation on the target range, that is, the target task is determined during the user's first browsing of the target content, ensuring user interaction while avoiding repeated task prompts to the user.

[0105] In a possible embodiment, the method may further include:

[0106] If it is determined that the user's answer to the target question is correct, a prompt message is output, wherein the prompt message is used to provide the user with feedback on the answer result obtained by the user in receiving the target task. The specific implementation method has been described above and will not be repeated here.

[0107] The user's account information is updated based on the target reward information associated with the target task.

[0108] Among them, the reward information associated with the target task can be pre-set, and the reward information associated with different target tasks can be the same or different. It can be set based on the actual application scenario, and the present disclosure does not limit this. When it is determined that the answer result is a correct result, it can be considered that the target task is completed, and the associated reward information can be distributed to the user's account, that is, the user's account information is updated. For example, if the target reward information associated with the target task is N chapter reading cards, then N chapter reading cards can be added to the user's account information, so that the user can subsequently unlock the content of the corresponding chapter based on the chapter reading card for reading.

[0109] Therefore, through the above technical solution, corresponding reward information can be distributed to users after they receive and complete the target task, thereby further increasing the user's interest in receiving the target task, and at the same time also increasing the user's reading participation and experience, facilitating the user's subsequent reading process.

[0110] Figure 7 As shown, it is a timing diagram of an interaction method provided based on an embodiment of the present disclosure, such as Figure 7As shown, before the user triggers the content viewing operation to browse the content, a target task can be generated. During the browsing process, the user can choose to receive the target task or not. If the target task is not received, it is determined that there is no feedback; if it is received, the answer candidate information can be found and submitted during the browsing process. If the user submits the content, the display of the display content of the target range is completed, that is, the user determines the answer result after browsing the display content. If the user does not submit the content, it is determined that there is no feedback and the target task is deleted after browsing the display content corresponding to the target range. Then browse the display content of the next target range and repeat the above process. Among them, the specific steps have been described in detail above and will not be repeated here.

[0111] In a possible embodiment, in response to receiving a content viewing operation of a user on a target range of target content, determining a target task corresponding to the target range may include:

[0112] In response to receiving the user's operation of viewing the contents in the target range, the contents in the target range are acquired.

[0113] Among them, content extraction can be performed from the target content based on the target range. For example, if the target range is Chapters 5 to 10, the chapter texts of Chapters 5 to 10 can be obtained to analyze the chapter texts.

[0114] Afterwards, a target task corresponding to the target range is generated according to the content within the target range and a task generation model corresponding to the target content, wherein the task generation model is implemented based on a large language model.

[0115] Among them, the corresponding task generation model can be pre-configured according to the type of target content. As an example, for suspense, reasoning, solving crimes and other types of content, it is usually more focused on the reasoning process of the content, and the plot cannot be revealed to the user during the interaction process. Therefore, for this type of content, the corresponding task generation model can be set to the question-and-answer type, that is, the target question is generated for the user and the user determines the answer during the reading process. For online articles, plots, stories, publishing and other types of content, it can be interacted with the user so that the user predicts the plot. Therefore, for this type of content, the corresponding task generation model can be set to the selection type, that is, the target question and answer candidates are generated for the user, so that the user predicts the plot by selection. The task generation model is implemented based on a large language model so that the model analysis and processing is performed inside the model based on the constructed natural language input, and the results are returned in natural language according to the specified format. Analysis based on the large language model can effectively improve the understanding ability and processing efficiency of the post-task generation model for the content within the target range.

[0116] Therefore, in this embodiment, after obtaining the content within the target range, a task generation model corresponding to the target content can be used to obtain the corresponding target task, thereby improving the correlation between the target task and the content within the target range while ensuring the matching degree between the target task and the target content, thereby ensuring the user's reading experience for the target content while improving user interaction.

[0117] In a possible embodiment, the type of the task generation model corresponding to the target content is a question-answering type.

[0118] Accordingly, generating a target task corresponding to the target range according to the content within the target range and the task generation model corresponding to the target content may include:

[0119] A first prompt text is constructed based on the content within the target scope, and the task generation model is input to obtain the event information of the target event in the content within the target scope, wherein the first prompt text contains the dimensions of the event information. The event information may generally contain the plot of the complete reasoning process corresponding to the target event, including information such as the occurrence of the event, clue collection, reasoning analysis, and result description. Therefore, the structure of the first prompt text can be expressed as follows:

[0120] Please sort out the derivation process of the results of the target event in the following content {content within the target scope}, including the process of discovering clues, phenomena, testing, verification, and proof

[0121] Thus, the content within the target range can be spliced ​​into the first prompt text and input into the task generation model, so that the task generation model can analyze the content within the target range to obtain the event information of the target event corresponding to the content within the target range. Wherein, if the content within the target range is video content, the text corresponding to the video content can be obtained based on the subtitles or audio in the video content, so as to construct the first prompt text based on the text corresponding to the video content.

[0122] like Fig. 8A As shown. Among them, the dimensions that need to be reflected in the event information can be input in the first prompt text, so that the task generation model can analyze according to the dimensions prompted in the first prompt text, improve the effectiveness and comprehensiveness of the event information, and effectively understand the specific information of the content within the target scope, thereby providing effective data support for the subsequent generation of target questions.

[0123] Based on the event information and the task generation model, target clue information that triggers the target event is determined.

[0124] As an example, the task generation model can remember the conversation context, so in this step, a prompt text of this conversation can be constructed to input into the task generation model to prompt the task generation model to perform analysis operations, such as:

[0125] Based on the above analysis results, please extract the key clues to the incident (a phenomenon, object, character performance, etc.)

[0126] The task generation model can analyze the prompt text of this conversation combined with the event information output above to determine the target clue information. Figure 8B shown.

[0127] As another example, if the task generation model has no context memory, the prompt text of this conversation can be constructed based on the event information and input into the task generation model, such as:

[0128] Please extract the key clues of the incident (a phenomenon, object, character performance, etc.) based on the following event analysis results {event information}

[0129] Therefore, the event information determined above can be spliced ​​into the prompt text and input into the task generation model, so that the task generation model can analyze based on the event information to obtain the target clue information.

[0130] Afterwards, the target task is generated based on the event information, the target clue information and the task generation model, and the target answer in the target task is generated based on the target clue information.

[0131] The target clue information may be directly output as the target answer in the target task, or may be analyzed based on a task generation model to obtain the target question and the target answer at the same time.

[0132] As an example, the task generation model can remember the conversation context, so in this step, a prompt text of this conversation can be constructed to input into the task generation model to prompt the task generation model to perform analysis operations, such as:

[0133] According to the plot content, ask the reader a question with the target clue information as the answer, and output a set of questions and answers in the format of:

[0134] Question: [question], Answer: [answer]

[0135] Then the task generation model can analyze the prompt text of this conversation in combination with the event information and target clue information output above, and generate a target question with the target clue information as the answer, so that the task generation model can output the target question and the target answer at the same time. Figure 8B shown.

[0136] As another example, if the task generation model has no context memory, the prompt text of this conversation can be constructed based on the event information and the target clue information and input into the task generation model, such as:

[0137] According to the plot content {event information}, a question is asked to the reader with the target clue information {target clue information} as the answer, and a set of questions and answers are output in the format of:

[0138] Question: [question], Answer: [answer]

[0139] Therefore, the event information and target clue information determined above can be spliced ​​into the prompt text and input into the task generation model, so that the task generation model can analyze the event information and target clue information and generate a target question with the target clue information as the answer.

[0140] As another example, when the task generation model does not have contextual memory, in order to further improve the accuracy of the model output results, the chapter content can be spliced ​​and input into the task generation model when constructing the prompt text, so as to further improve the comprehensiveness of the data analyzed by the model, thereby improving the accuracy of the determined target clue information, so as to improve the accuracy of the generation of target questions and target answers, provide reliable data support for the subsequent distribution of target tasks to users, and ensure the user's interactive experience.

[0141] Therefore, through the above technical solution, the content within the target scope can be analyzed in multiple steps through the task generation model to ultimately determine the target question and target answer corresponding to the content within the target scope, thereby ensuring the correlation between the generated target question and target answer and the content within the target scope, making it easier for users to understand the requirements of the target task and ensuring the interactive experience of users when looking for answers during the reading process.

[0142] In a possible embodiment, the task generation model is constructed in the following manner:

[0143] Obtain training text corpus. As an example, for task generation models corresponding to question-answering types, novels of the types of mystery, suspense, mystery, and crime-solving can be obtained in advance as training text corpus, which can be obtained from open source resources.

[0144] Event recognition is performed on the training text corpus based on a large language model to obtain training events corresponding to the training text corpus.

[0145] Wherein, in this step, for the long text in the training text corpus, event recognition can be performed based on the training corpus, so that multiple events corresponding to the long text can be determined. For the short text in the training corpus, the content of the short text can be regarded as an event without splitting. Wherein, it can be determined whether it is a long text or a short text based on the classification attribute of the training text corpus. If the classification attribute is not marked, it can be determined whether it is a long text or a short text based on the number of chapters contained in the training text corpus. The number of chapters can be pre-set. If the total number of chapters in the training text corpus is less than the number of chapters, the training text corpus is a short text. If the total number of chapters in the training text corpus is greater than or equal to the number of chapters, the training text corpus is a long text. In this process, the prompt text can be constructed and implemented by a universal large language model. The present disclosure is not limited to this.

[0146] Afterwards, for each of the training events, the training text corresponding to the training event is obtained from the training text corpus corresponding to the training event, and training answers and training questions are generated based on the large language model and the training text.

[0147] If the training text corpus is a short text, the training text corpus can be used as the training text to construct the prompt text; if the training text corpus is a long text, the chapter corresponding to the training event can be extracted from the training text corpus as the training text to construct the prompt text. Then, the training answers and training questions are generated based on the constructed prompt text and the large language model.

[0148] For example, the structure of the prompt text is as follows:

[0149] Please analyze the following content {training text} and output N sets of questions and answers in the following format:

[0150] Question: [question], Answer: [answer]

[0151] Among them, N can be set based on the actual application scenario, wherein the question corresponds to the training question, and the output answer corresponds to the training answer. Through the above steps, multiple sets of questions and answers can be automatically generated based on the large language model.

[0152] Afterwards, based on the training answers and the training questions corresponding to the training text, the annotated answers and annotated questions corresponding to the training text are determined to obtain training samples, wherein each of the training samples contains the training text and the annotated questions and annotated answers corresponding to the training text.

[0153] Among them, there may be combinations of questions and answers that are not highly related to the chapter text or have incorrect answers. In this step, for each set of training answers and training questions, the training answers and training questions can be output and displayed so that they can be reviewed by the reviewer, and the approved training answers and training questions can be used as the marked answers and marked questions. For another example, the reviewer can also correct the training answers, and the corrected training answers and training questions can be used as the marked answers and marked questions, thereby obtaining the training samples.

[0154] Fine-tune the large language model based on the training samples, and use the trained model as the task generation model.

[0155] Among them, the training sample includes the training sample and its corresponding questions and answers, and the large language model can be optimized based on the training sample to realize the task generation in the task generation scenario based on the large language model. Fine-tuning training can be realized based on the training sample combined with the fine-tuning algorithm of the large language model commonly used in the field, and the present disclosure is not limited to this. Therefore, through the above technical solution, the initial questions and answers can be generated for the training text corpus through the large language model, so as to effectively reduce the workload of the user for analysis and improve the efficiency of training sample generation. At the same time, the generated initial questions and answers are screened or corrected to improve the accuracy of the labeled questions and labeled answers, thereby improving the accuracy and effectiveness of the training samples, and further improving the accuracy of the target tasks subsequently generated by the task generation model, and ensuring effective interaction with the user in the subsequent reading process.

[0156] In a possible embodiment, the type of the task generation model corresponding to the target content is a selection type, and the target task also includes wrong answers.

[0157] Accordingly, generating a target task corresponding to the target range according to the content within the target range and the task generation model corresponding to the target content may include:

[0158] A second prompt text is constructed based on the content within the target range, and is input into the task generation model to obtain plot information corresponding to the content within the target range and key plot change information corresponding to the plot information, wherein the second prompt text contains the dimension of the plot information.

[0159] As an example, the structure of the second prompt text can be expressed as follows:

[0160] Summarize the plot in the following content {content within the target scope}, and focus on explaining the new characters, things or new changes in the plot, as well as the key turning points

[0161] Accordingly, the content within the target range can be spliced ​​into the second prompt text to input the task generation model to obtain plot information and key plot change information. Among them, the dimensions that need to be reflected in the plot information can be input in the second prompt text, so that the task generation model can be analyzed according to the dimensions prompted in the second prompt text, improve the effectiveness and comprehensiveness of the plot information, and effectively understand the specific content of the content within the target range, thereby providing effective data support for the subsequent generation of target questions. Among them, the content within the target range is video content, then the text corresponding to the video content can be obtained based on the subtitles or audio in the video content, so as to construct the second prompt text based on the text corresponding to the video content.

[0162] Afterwards, based on the key plot change information, the plot information and the task generation model, the factor information triggering the key plot change information and the result information corresponding to the key plot change information are determined.

[0163] As an example, the task generation model can remember the conversation context, so in this step, a prompt text of this conversation can be constructed to input into the task generation model to prompt the task generation model to perform analysis operations, such as:

[0164] In the key plot change information, what are the behaviors and things that lead to the changes, and what are the results of the changes?

[0165] The task generation model can then analyze the prompt text of this conversation in combination with the plot information and key plot change information output above to determine the factor information and result information.

[0166] As another example, if the task generation model has no context memory, the prompt text of this dialogue can be constructed based on the plot information and key plot turning information and input into the task generation model, such as:

[0167] Based on the following plot information {plot information}, please determine what factors lead to the changes in the key plot change information {key plot change information}, and what are the results of the changes?

[0168] Therefore, the plot information and key plot turning information determined above can be spliced ​​into the prompt text and input into the task generation model, so that the task generation model can analyze the information based on the information to obtain factor information and result information.

[0169] Afterwards, the target problem is determined based on the key plot change information, the factor information, the result information and the task generation model.

[0170] As an example, the task generation model can remember the conversation context, so in this step, a prompt text of this conversation can be constructed to input into the task generation model to prompt the task generation model to perform analysis operations, such as:

[0171] In the critical transition, the behavior that led to the change is {factor information}, please ask in a questioning tone

[0172] The task generation model can analyze the prompt text of this dialogue in combination with the key plot change information, the factor information, and the result information output above, and generate a target question based on the factor information prompted in the prompt text.

[0173] As another example, if the task generation model has no context memory, the prompt text of this dialogue can be constructed based on the key plot change information, the factor information, and the result information and input into the task generation model, such as:

[0174] In the key turning point {key plot change information}, the behavior that causes the change is {factor information}, and its corresponding result is {result information}. Please ask questions in a questioning tone.

[0175] Therefore, the key plot change information, the factor information, and the result information determined above can be spliced ​​into the prompt text and input into the task generation model, so that the task generation model can analyze the key plot change information, the factor information, and the result information to generate a target question corresponding to the factor information.

[0176] As another example, when the task generation model does not have contextual memory, in order to further improve the accuracy of the model output results, the chapter content can be spliced ​​and input into the task generation model when constructing the prompt text, so as to further improve the comprehensiveness of the data analyzed by the model, thereby improving the accuracy of the determined target clue information, so as to improve the accuracy of the generation of target questions and target answers, provide reliable data support for the subsequent distribution of target tasks to users, and ensure the user's interactive experience.

[0177] Afterwards, based on the target question, the factor information, the plot information and the task generation model, a target answer and an incorrect answer corresponding to the target question are determined, wherein the target answer is generated based on the factor information, and the incorrect answer is different from the target answer.

[0178] As an example, the task generation model can remember the conversation context, so in this step, a prompt text of this conversation can be constructed to input into the task generation model to prompt the task generation model to perform analysis operations, such as:

[0179] Complete options for supplementary questions, requiring that the character behaviors in different options should not be repeated:

[0180] Question: {target question}

[0181] Option 1: {Factor Information}

[0182] Option 2: ...

[0183] Option 3: ...

[0184] The task generation model can analyze the prompt text of this conversation in combination with the target question, the factor information, and the plot information output above to generate a target answer based on the factor information prompted in the prompt text, and generate an incorrect answer based on the target question and the plot information.

[0185] As another example, if the task generation model does not have context memory, the prompt text of this dialogue can be constructed based on the target question, the factor information, and the plot information and input into the task generation model. Thus, the target question, the factor information, and the plot information determined above can be spliced ​​into the corresponding prompt text and input into the task generation model, so that the task generation model can analyze the target question, the factor information, and the plot information, generate the target answer based on the factor information prompted in the prompt text, and generate the wrong answer based on the target question and the plot information.

[0186] Therefore, through the above technical solution, the content of the chapter text can be analyzed in multiple steps through the task generation model to finally determine the target question, target answer and wrong answer corresponding to the chapter text, so as to ensure the relevance between the generated target question and target answer and the chapter text, and at the same time, the similarity between the wrong answer and the correct answer can be improved to increase the difficulty of users' predictions and ensure the interactive experience of users when looking for answers during the reading process.

[0187] In a possible embodiment, the task generation model is constructed in the following manner:

[0188] Obtain training text corpus. As an example, for the task generation model corresponding to the selected type, novels of the types of online articles, plots, stories, and publications can be obtained in advance as training text corpus, which can be obtained from open source resources. After obtaining the novel text, it can be divided into chapters according to the chapter directory of the novel, so that the text of each chapter is used as a training text corpus.

[0189] Generate training correct answers, training questions and training incorrect answers based on the large language model and the training text corpus.

[0190] For example, the structure of the prompt text is as follows:

[0191] Please analyze the following {training text corpus} and output questions and multiple answers in the following format:

[0192] Question: [question],

[0193] Correct answer: [answer],

[0194] Wrong answer: [answer],

[0195] Wrong answer: [answer]

[0196] The question corresponds to the training question, the correct answer output corresponds to the training correct answer, and the wrong answer output corresponds to the training wrong answer. Thus, through the above steps, questions and answers can be automatically generated based on the large language model.

[0197] Afterwards, based on the training correct answers, the training questions and the training incorrect answers corresponding to the training text corpus, the annotated correct answers, annotated questions and annotated incorrect answers corresponding to the training text corpus are determined to obtain training samples.

[0198] Among them, there may be combinations of questions and answers that are not highly related to the chapter text or have incorrect answers. In this step, for each set of training correct answers, training incorrect answers, and training questions, they can be output and displayed so that they can be reviewed by the reviewer, and the training correct answers, training questions, and training incorrect answers that have passed the review can be used as the marked correct answers, marked questions, and marked incorrect answers. For another example, the reviewer can also correct the above questions or answers, and the corrected training correct answers, training questions, and training incorrect answers can be used as the marked correct answers, marked questions, and marked incorrect answers, thereby obtaining the training sample.

[0199] Fine-tune the large language model based on the training samples, and use the trained model as the task generation model.

[0200] Among them, the training sample includes the training sample and its corresponding questions, correct answers and wrong answers, then the large language model can be optimized based on the training sample to realize task generation in the task generation scenario based on the large language model. Fine-tuning training can be realized based on the training sample combined with the fine-tuning algorithm of the large language model commonly used in the field, and the present disclosure does not limit this. Therefore, through the above technical scheme, the initial task can be generated for the training text corpus through the large language model, so as to effectively reduce the workload of the user for analysis and improve the efficiency of training sample generation. At the same time, the information in the generated initial task is screened or corrected to improve the accuracy of marking questions, marking correct answers and marking wrong answers, thereby improving the accuracy and effectiveness of the training samples, and further improving the accuracy of the target tasks subsequently generated by the task generation model, and ensuring effective interaction with the user in the subsequent reading process.

[0201] Based on the same inventive concept, the present disclosure also provides an interactive device, such as Fig. 9 As shown, the device 10 comprises:

[0202] A first determination module 100 is configured to determine a target task corresponding to a target range of target content in response to receiving a content viewing operation of the user on the target range of target content, wherein the target task includes a target question and a target answer;

[0203] A first display module 200 is used to display the display content corresponding to the target problem and the target scope in a display interface;

[0204] An acquisition module 300 is used to acquire candidate answer information of the target question during the display process of the display content corresponding to the target range if it is determined that the user has received the target task;

[0205] The second determination module 400 is used to determine the user's answer result to the target task based on the target answer and the answer candidate information in response to the display content corresponding to the target range being displayed.

[0206] Optionally, the target task is of question-answering type;

[0207] The acquisition module comprises:

[0208] The first acquisition submodule is used to determine the answer candidate information according to information corresponding to the display content selected by the selection operation in response to receiving a selection operation of the display content by the user during the display process of the display content corresponding to the target range.

[0209] Optionally, the target content is a target novel, and the displayed content is text content;

[0210] The first acquisition submodule includes:

[0211] The first processing submodule is used to use the text content selected by the selection operation as the answer candidate information.

[0212] Optionally, the target content is a target video, and the displayed content is video content;

[0213] The first acquisition submodule includes:

[0214] The second processing submodule is used to use the text corresponding to the video content selected by the selection operation as the answer candidate information, wherein the text corresponding to the video content includes the text corresponding to the subtitles in the video content and / or the text corresponding to the audio in the video content.

[0215] Optionally, the type of the target task is a selection type, and the target task further includes wrong answers;

[0216] The first display module includes:

[0217] A first display submodule is used to display the target question in the display interface, and to display the target answer and the wrong answer as answer candidates corresponding to the target question;

[0218] A second display submodule is used to display the display content in the area below the display area of ​​the target question and the answer candidates;

[0219] The acquisition module comprises:

[0220] The second acquisition submodule is used to determine that the user takes up the target task in response to receiving a selection operation of the user among the answer candidate items, and determine the answer candidate information based on the selected answer candidate items.

[0221] Optionally, the device further comprises:

[0222] a storage module, configured to store the display progress of the target content, the target question and the candidate answer information if it is determined that the user has received the target task and a closing operation of the target content by the user is received during the display process of the display content corresponding to the target range, wherein the candidate answer information is empty when the user has not selected any content;

[0223] The second display module is used to display the target content based on the display progress in response to receiving the viewing operation of the user on the target content, and to display the display mark corresponding to the target question.

[0224] Optionally, the device further comprises:

[0225] A deletion module, configured to delete the target task in response to completion of display of the display content corresponding to the target range;

[0226] The first determining module is further configured to:

[0227] In response to receiving a content viewing operation of the target range by the user for the first time, determining the target task.

[0228] Optionally, the device further comprises:

[0229] An output module, configured to output prompt information if it is determined that the answer of the user to the target question is a correct answer;

[0230] An updating module is used to update the user's account information based on the target reward information associated with the target task.

[0231] Optionally, the device further comprises:

[0232] A third display module is used to display a display mark corresponding to the target problem in the display interface if it is determined that the user has received the target task;

[0233] A fourth display module is used to display the target question in a display card in response to the user's selection operation on the display mark, wherein the display card is displayed in the form of a floating layer on the upper layer of the display interface.

[0234] Optionally, the first determining module includes:

[0235] A third acquisition submodule is configured to acquire the content within the target range in response to receiving a content viewing operation of the user on the target range;

[0236] The first generating submodule is used to generate a target task corresponding to the target range according to the content within the target range and a task generating model corresponding to the target content, wherein the task generating model is implemented based on a large language model.

[0237] Optionally, the type of the task generation model corresponding to the target content is a question-answering type;

[0238] The first generation submodule comprises:

[0239] A first determination submodule is configured to construct a first prompt text based on the content within the target range, and input the task generation model to obtain event information of a target event in the content within the target range, wherein the first prompt text contains dimensions of the event information;

[0240] A second determination submodule is used to determine target clue information that triggers the occurrence of the target event based on the event information and the task generation model;

[0241] The second generation submodule is used to generate the target task based on the event information, the target clue information and the task generation model, and the target answer in the target task is generated based on the target clue information.

[0242] Optionally, the task generation model is constructed in the following manner:

[0243] Get training text corpus;

[0244] Performing event recognition on the training text corpus based on a large language model to obtain training events corresponding to the training text corpus;

[0245] For each of the training events, obtaining a training text corresponding to the training event from a training text corpus corresponding to the training event, and generating a training answer and a training question based on the large language model and the training text;

[0246] Determine the marked answers and marked questions corresponding to the training text according to the training answers and the training questions corresponding to the training text to obtain training samples;

[0247] Fine-tune the large language model based on the training samples, and use the trained model as the task generation model.

[0248] Optionally, the type of the task generation model corresponding to the target content is a selection type, and the target task further includes wrong answers;

[0249] The first generation submodule includes:

[0250] A third determination submodule is configured to construct a second prompt text based on the content within the target range, and input the task generation model to obtain plot information corresponding to the content within the target range and key plot change information corresponding to the plot information, wherein the second prompt text contains the dimension of the plot information;

[0251] A fourth determination submodule, configured to determine, based on the key plot change information, the plot information, and the task generation model, factor information triggering the key plot change information and result information corresponding to the key plot change information;

[0252] A fifth determination submodule, configured to determine the target problem based on the key plot change information, the factor information, the result information and the task generation model;

[0253] The sixth determination submodule is used to determine the target answer and wrong answer corresponding to the target question based on the target question, the factor information, the plot information and the task generation model, wherein the target answer is generated based on the factor information, and the wrong answer is different from the target answer.

[0254] Optionally, the task generation model is constructed in the following manner:

[0255] Get training text corpus;

[0256] Generate training correct answers, training questions, and training incorrect answers based on the large language model and the training text corpus;

[0257] According to the training correct answers, the training questions and the training wrong answers corresponding to the training text corpus, determining the marked correct answers, marked questions and marked wrong answers corresponding to the training text corpus to obtain training samples;

[0258] Fine-tune the large language model based on the training samples, and use the trained model as the task generation model.

[0259] Reference below Fig.10 , which shows a schematic diagram of the structure of an electronic device (such as a terminal device or a server) 600 suitable for implementing the embodiment of the present disclosure. The terminal device in the embodiment of the present disclosure may include but is not limited to mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Fig.10 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0260] like Fig.10 As shown, the electronic device 600 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 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the electronic device 600 are also stored. 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.

[0261] 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 magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Fig.10 The electronic device 600 is shown with various devices, 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.

[0262] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device 609, or installed from a storage device 608, or installed from a ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0263] It should be noted that the computer-readable medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0264] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0265] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0266] The above-mentioned computer-readable medium carries one or more programs. When the above-mentioned one or more programs are executed by the electronic device, the electronic device: in response to receiving a content viewing operation of a user on a target range of target content, determines a target task corresponding to the target range, wherein the target task includes a target question and a target answer; displays the target question and the display content corresponding to the target range in a display interface; if it is determined that the user takes up the target task, obtains answer candidate information of the target question during the display process of the display content corresponding to the target range; in response to the display of the display content corresponding to the target range being completed, determines the user's answer result to the target task based on the target answer and the answer candidate information.

[0267] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0268] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0269] The modules involved in the embodiments described in the present disclosure may be implemented by software or hardware. The name of the module does not limit the module itself in some cases. For example, the first determination module may also be described as "a module for determining a target task corresponding to a target range of target content in response to receiving a content viewing operation of the user on the target range of target content".

[0270] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0271] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0272] According to one or more embodiments of the present disclosure, Example 1 provides an interaction method, wherein the method includes:

[0273] In response to receiving a content viewing operation of a target range of target content by a user, determining a target task corresponding to the target range, wherein the target task includes a target question and a target answer;

[0274] Displaying the target problem and the display content corresponding to the target range in the display interface;

[0275] If it is determined that the user takes up the target task, in the process of displaying the display content corresponding to the target range, obtaining the candidate answer information of the target question;

[0276] In response to completion of display of the display content corresponding to the target range, a result of the user's answer to the target task is determined based on the target answer and the answer candidate information.

[0277] According to one or more embodiments of the present disclosure, Example 2 provides the method of Example 1, wherein the type of the target task is a question-answering type;

[0278] In the process of displaying the display content corresponding to the target range, obtaining the candidate answer information of the target question includes:

[0279] During the display process of the display content corresponding to the target range, in response to receiving a selection operation of the user on the display content, the answer candidate information is determined according to information corresponding to the display content selected by the selection operation.

[0280] According to one or more embodiments of the present disclosure, Example 3 provides the method of Example 2, wherein the target content is a target novel, and the displayed content is text content;

[0281] The determining the answer candidate information according to the information corresponding to the display content selected by the selection operation includes:

[0282] The text content selected by the selection operation is used as the answer candidate information.

[0283] According to one or more embodiments of the present disclosure, Example 4 provides the method of Example 2, wherein the target content is a target video, and the display content is a video content;

[0284] The determining the answer candidate information according to the information corresponding to the display content selected by the selection operation includes:

[0285] The text corresponding to the video content selected by the selection operation is used as the answer candidate information, and the text corresponding to the video content includes text corresponding to subtitles in the video content and / or text corresponding to audio in the video content.

[0286] According to one or more embodiments of the present disclosure, Example 5 provides the method of Example 1, wherein the type of the target task is a selection type, and the target task further includes a wrong answer;

[0287] The displaying of the target problem and the display content corresponding to the target range in the display interface includes:

[0288] Displaying the target question in the display interface, and displaying the target answer and the wrong answer as answer candidates corresponding to the target question;

[0289] Displaying the display content in the area below the display area of ​​the target question and the answer candidates;

[0290] If it is determined that the user has received the target task, in the process of displaying the display content corresponding to the target range, obtaining the candidate answer information of the target question includes:

[0291] In response to receiving a selection operation of the user among the answer candidates, it is determined that the user receives the target task, and the answer candidate information is determined based on the selected answer candidate.

[0292] According to one or more embodiments of the present disclosure, Example 6 provides the method of Example 1, wherein the method further includes:

[0293] If it is determined that the user has received the target task, and during the display process of the display content corresponding to the target range, the user closes the target content, storing the display progress of the target content, the target question and the answer candidate information, wherein the answer candidate information is empty when the user has not selected any content;

[0294] In response to receiving a viewing operation of the user on the target content, the target content is displayed based on the display progress, and a display identifier corresponding to the target question is displayed.

[0295] According to one or more embodiments of the present disclosure, Example 7 provides the method of Example 1, wherein the method further includes:

[0296] In response to completion of display of the display content corresponding to the target range, deleting the target task;

[0297] In response to receiving a content viewing operation of a target range of target content by a user, determining a target task corresponding to the target range includes:

[0298] In response to receiving a content viewing operation of the target range by the user for the first time, determining the target task.

[0299] According to one or more embodiments of the present disclosure, Example 8 provides the method of Example 1, wherein the method further includes:

[0300] If it is determined that the user's answer to the target question is a correct answer, outputting a prompt message;

[0301] The user's account information is updated based on the target reward information associated with the target task.

[0302] According to one or more embodiments of the present disclosure, Example 9 provides the method of Example 1, wherein the method further includes:

[0303] If it is determined that the user has received the target task, a display mark corresponding to the target problem is displayed in the display interface;

[0304] In response to the user's selection operation on the display mark, the target question is displayed in a display card, wherein the display card is displayed in the form of a floating layer on the upper layer of the display interface.

[0305] According to one or more embodiments of the present disclosure, Example 10 provides the method of Example 1, wherein, in response to receiving a content viewing operation of a user on a target range of target content, determining a target task corresponding to the target range includes:

[0306] In response to receiving a content viewing operation of the user on the target range, acquiring the content within the target range;

[0307] A target task corresponding to the target scope is generated according to the content within the target scope and a task generation model corresponding to the target content, wherein the task generation model is implemented based on a large language model.

[0308] According to one or more embodiments of the present disclosure, Example 11 provides the method of Example 10, wherein the type of the task generation model corresponding to the target content is a question-answering type;

[0309] The generating a target task corresponding to the target range according to the content within the target range and the task generating model corresponding to the target content comprises:

[0310] Constructing a first prompt text based on the content within the target range, and inputting the first prompt text into the task generation model to obtain event information of a target event in the content within the target range, wherein the first prompt text contains the dimension of the event information;

[0311] Based on the event information and the task generation model, determining target clue information that triggers the occurrence of the target event;

[0312] The target task is generated based on the event information, the target clue information and the task generation model, and the target answer in the target task is generated based on the target clue information.

[0313] According to one or more embodiments of the present disclosure, Example 12 provides the method of Example 11, wherein the task generation model is constructed in the following manner:

[0314] Get training text corpus;

[0315] Performing event recognition on the training text corpus based on a large language model to obtain training events corresponding to the training text corpus;

[0316] For each of the training events, obtaining a training text corresponding to the training event from a training text corpus corresponding to the training event, and generating a training answer and a training question based on the large language model and the training text;

[0317] Determine the marked answers and marked questions corresponding to the training text according to the training answers and the training questions corresponding to the training text to obtain training samples;

[0318] Fine-tune the large language model based on the training samples, and use the trained model as the task generation model.

[0319] According to one or more embodiments of the present disclosure, Example 13 provides the method of Example 10, wherein the type of the task generation model corresponding to the target content is a selection type, and the target task further includes a wrong answer;

[0320] The generating a target task corresponding to the target range according to the content within the target range and the task generating model corresponding to the target content comprises:

[0321] Constructing a second prompt text based on the content within the target range, and inputting the second prompt text into the task generation model to obtain plot information corresponding to the content within the target range and key plot change information corresponding to the plot information, wherein the second prompt text contains the dimension of the plot information;

[0322] Based on the key plot change information, the plot information and the task generation model, determining factor information triggering the key plot change information and result information corresponding to the key plot change information;

[0323] Determining the target problem based on the key plot change information, the factor information, the result information and the task generation model;

[0324] Based on the target question, the factor information, the plot information and the task generation model, a target answer and an incorrect answer corresponding to the target question are determined, wherein the target answer is generated based on the factor information, and the incorrect answer is different from the target answer.

[0325] According to one or more embodiments of the present disclosure, Example 14 provides the method of Example 13, wherein the task generation model is constructed in the following manner:

[0326] Get training text corpus;

[0327] Generate training correct answers, training questions, and training incorrect answers based on the large language model and the training text corpus;

[0328] According to the training correct answers, the training questions and the training wrong answers corresponding to the training text corpus, determining the marked correct answers, marked questions and marked wrong answers corresponding to the training text corpus to obtain training samples;

[0329] Fine-tune the large language model based on the training samples, and use the trained model as the task generation model.

[0330] According to one or more embodiments of the present disclosure, Example 15 provides an interactive device, wherein the device includes:

[0331] A first determination module, configured to, in response to receiving a content viewing operation of a user on a target range of target content, determine a target task corresponding to the target range, wherein the target task includes a target question and a target answer;

[0332] A first display module, used to display the display content corresponding to the target problem and the target scope in a display interface;

[0333] An acquisition module, configured to acquire candidate answer information of the target question during the display process of the display content corresponding to the target range if it is determined that the user has received the target task;

[0334] The second determination module is used to determine the user's answer result to the target task based on the target answer and the answer candidate information in response to the display content corresponding to the target range being displayed.

[0335] According to one or more embodiments of the present disclosure, Example 16 provides a computer-readable medium having a computer program stored thereon, wherein the computer program implements the steps of any of the methods described in Examples 1-14 when executed by a processing device.

[0336] According to one or more embodiments of the present disclosure, Example 17 provides an electronic device, including:

[0337] a storage device having a computer program stored thereon;

[0338] A processing device is used to execute the computer program in the storage device to implement the steps of any one of the methods described in Examples 1-14.

[0339] According to one or more embodiments of the present disclosure, Example 18 provides a computer program product, including a computer program, which implements the steps of any one of the methods of Examples 1-14 when executed by a processor.

[0340] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.

[0341] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0342] Although the subject matter has been described in language specific to structural features and / or method logic actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims. Regarding the device in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.

Claims

1. An interactive method, It is characterized in that The method comprises: In response to receiving a content viewing operation of a target range of target content by a user, determining a target task corresponding to the target range, wherein the target task includes a target question and a target answer; Displaying the target problem and the display content corresponding to the target range in the display interface; If it is determined that the user takes up the target task, in the process of displaying the display content corresponding to the target range, obtaining the candidate answer information of the target question; In response to completion of display of the display content corresponding to the target range, a result of the user's answer to the target task is determined based on the target answer and the answer candidate information.

2. The method according to claim 1, It is characterized in that The type of the target task is question-answering type; In the process of displaying the display content corresponding to the target range, obtaining the candidate answer information of the target question includes: During the display process of the display content corresponding to the target range, in response to receiving a selection operation of the user on the display content, the answer candidate information is determined according to information corresponding to the display content selected by the selection operation.

3. The method according to claim 2, It is characterized in that The target content is a target novel, and the displayed content is text content; The determining the answer candidate information according to the information corresponding to the display content selected by the selection operation includes: The text content selected by the selection operation is used as the answer candidate information.

4. The method according to claim 2, It is characterized in that The target content is a target video, and the displayed content is a video content; The determining the answer candidate information according to the information corresponding to the display content selected by the selection operation includes: The text corresponding to the video content selected by the selection operation is used as the answer candidate information, and the text corresponding to the video content includes text corresponding to subtitles in the video content and / or text corresponding to audio in the video content.

5. The method according to claim 1, It is characterized in that The type of the target task is a selection type, and the target task also includes wrong answers; The displaying of the target problem and the display content corresponding to the target scope in the display interface includes: Displaying the target question in the display interface, and displaying the target answer and the wrong answer as answer candidates corresponding to the target question; Displaying the display content in the area below the display area of ​​the target question and the answer candidates; If it is determined that the user receives the target task, in the process of displaying the display content corresponding to the target range, obtaining the candidate answer information of the target question includes: In response to receiving a selection operation of the user among the answer candidates, it is determined that the user receives the target task, and the answer candidate information is determined based on the selected answer candidate.

6. The method according to claim 1, It is characterized in that The method further comprises: If it is determined that the user has received the target task, and during the display process of the display content corresponding to the target range, the user closes the target content, storing the display progress of the target content, the target question and the answer candidate information, wherein the answer candidate information is empty when the user has not selected any content; In response to receiving a viewing operation of the user on the target content, the target content is displayed based on the display progress, and a display identifier corresponding to the target question is displayed.

7. The method according to claim 1, It is characterized in that The method further comprises: In response to completion of display of the display content corresponding to the target range, deleting the target task; In response to receiving a content viewing operation of a target range of target content by a user, determining a target task corresponding to the target range includes: In response to receiving a content viewing operation of the target range by the user for the first time, determining the target task.

8. The method according to claim 1, It is characterized in that The method further comprises: If it is determined that the user's answer to the target question is a correct answer, outputting a prompt message; The user's account information is updated based on the target reward information associated with the target task.

9. The method according to claim 1, It is characterized in that The method further comprises: If it is determined that the user has received the target task, a display mark corresponding to the target problem is displayed in the display interface; In response to the user's selection operation on the display mark, the target question is displayed in a display card, wherein the display card is displayed in the form of a floating layer on the upper layer of the display interface.

10. The method according to claim 1, It is characterized in that In response to receiving a content viewing operation of a target range of target content by a user, determining a target task corresponding to the target range includes: In response to receiving a content viewing operation of the user on the target range, acquiring the content within the target range; A target task corresponding to the target scope is generated according to the content within the target scope and a task generation model corresponding to the target content, wherein the task generation model is implemented based on a large language model.

11. The method according to claim 10, It is characterized in that The type of the task generation model corresponding to the target content is a question-answering type; The generating a target task corresponding to the target range according to the content within the target range and the task generating model corresponding to the target content comprises: Constructing a first prompt text based on the content within the target range, and inputting the first prompt text into the task generation model to obtain event information of a target event in the content within the target range, wherein the first prompt text contains the dimension of the event information; Based on the event information and the task generation model, determining target clue information that triggers the target event to occur; The target task is generated based on the event information, the target clue information and the task generation model, and the target answer in the target task is generated based on the target clue information.

12. The method according to claim 11, It is characterized in that The task generation model is constructed in the following way: Get training text corpus; Performing event recognition on the training text corpus based on a large language model to obtain training events corresponding to the training text corpus; For each of the training events, obtaining a training text corresponding to the training event from a training text corpus corresponding to the training event, and generating a training answer and a training question based on the large language model and the training text; Determine the marked answers and marked questions corresponding to the training text according to the training answers and the training questions corresponding to the training text to obtain training samples; Fine-tune the large language model based on the training samples, and use the trained model as the task generation model.

13. The method according to claim 10, It is characterized in that The type of the task generation model corresponding to the target content is a selection type, and the target task also includes wrong answers; The generating a target task corresponding to the target range according to the content within the target range and the task generating model corresponding to the target content comprises: Constructing a second prompt text based on the content within the target range, and inputting the second prompt text into the task generation model to obtain plot information corresponding to the content within the target range and key plot change information corresponding to the plot information, wherein the second prompt text contains the dimension of the plot information; Based on the key plot change information, the plot information and the task generation model, determining factor information triggering the key plot change information and result information corresponding to the key plot change information; Determining the target problem based on the key plot change information, the factor information, the result information and the task generation model; Based on the target question, the factor information, the plot information and the task generation model, a target answer and an incorrect answer corresponding to the target question are determined, wherein the target answer is generated based on the factor information, and the incorrect answer is different from the target answer.

14. The method according to claim 13, It is characterized in that The task generation model is constructed in the following way: Get training text corpus; Generate training correct answers, training questions, and training incorrect answers based on the large language model and the training text corpus; According to the training correct answers, the training questions and the training wrong answers corresponding to the training text corpus, determining the marked correct answers, marked questions and marked wrong answers corresponding to the training text corpus to obtain training samples; Fine-tune the large language model based on the training samples, and use the trained model as the task generation model.

15. An interactive device, It is characterized in that The device comprises: A first determination module, configured to, in response to receiving a content viewing operation of a user on a target range of target content, determine a target task corresponding to the target range, wherein the target task includes a target question and a target answer; A first display module, used to display the display content corresponding to the target problem and the target scope in a display interface; An acquisition module, configured to acquire candidate answer information of the target question during the display process of the display content corresponding to the target range if it is determined that the user has received the target task; The second determination module is used to determine the user's answer result to the target task based on the target answer and the answer candidate information in response to the display content corresponding to the target range being displayed.

16. A computer readable medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processing device, the steps of the method according to any one of claims 1 to 14 are implemented.

17. An electronic device, It is characterized in that include: a storage device having a computer program stored thereon; A processing device, configured to execute the computer program in the storage device to implement the steps of the method according to any one of claims 1 to 14.

18. A computer program product comprising a computer program, It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 14 are implemented.

Images