A target outline generation method, system and related device

Abstract

The application discloses a target outline generation method, system and related device, and the method comprises the following steps: obtaining a first requirement input by a user, generating at least one title corresponding to a root level based on the first requirement; in response to obtaining a second requirement input by the user, generating a title of a next level corresponding to a current level based on the second requirement and at least part of the generated title, until a target outline composed of titles of all levels is obtained. In the foregoing manner, the application can improve the fitting degree of the generated outline and the user requirement.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method, system and related apparatus for generating target outlines. Background Technology

[0002] With the advent of the data age, the amount of text that users need to generate is increasing. Therefore, users increasingly expect to efficiently generate outlines for their texts before writing, thus improving subsequent writing efficiency. Current technology typically recommends outlines from historical texts in a database with the highest similarity to the type of text the user needs to write, providing a reference for the user's writing. However, since there are often differences between the text the user needs to write and historical texts, the outlines extracted from historical texts are difficult to accurately match the user's needs. Therefore, how to improve the relevance of generated outlines to user needs has become an urgent problem to be solved. Summary of the Invention

[0003] The main technical problem addressed by this application is to provide a method, system, and related apparatus for generating target outlines, which can improve the degree to which the generated outlines match user needs.

[0004] To solve the above-mentioned technical problems, one technical solution adopted in this application is: to provide a target outline generation method, including: obtaining a first requirement input by a user, generating at least one title corresponding to the root level based on the first requirement; in response to obtaining a second requirement input by a user, generating at least one title of the next level corresponding to the current level based on the second requirement and at least some of the generated titles, until a target outline composed of titles of all levels is obtained.

[0005] To solve the above-mentioned technical problems, another technical solution adopted in this application is: to provide a target outline generation system, including: an acquisition module, used to acquire a first requirement input by a user, and generate at least one title corresponding to the root level based on the first requirement; and a processing module, in response to acquiring a second requirement input by a user, the processing module is used to generate at least the title of the next level corresponding to the current level based on the second requirement and at least some of the generated titles, until a target outline composed of titles of all levels is obtained.

[0006] To solve the above-mentioned technical problems, another technical solution adopted in this application is: to provide a mobile terminal, including: a display screen, a memory, and a processor, wherein the display screen and the memory are respectively coupled to the processor, the display screen is used at least to display content to the user and to allow the user to input or select content, the memory stores program instructions, and the processor is used to execute the program instructions to implement the target outline generation method mentioned in the above technical solution.

[0007] To solve the above-mentioned technical problems, another technical solution adopted in this application is to provide a computer-readable storage medium storing program instructions thereon, wherein the program instructions, when executed by a processor, implement the target outline generation method mentioned in the above technical solution.

[0008] The beneficial effects of this application are as follows: Unlike existing technologies, the target outline generation method proposed in this application, after generating at least one title corresponding to the root level matching the user's first requirement based on the user's input, in response to obtaining the user's second requirement, generates at least one title for the next level below the current level based on the second requirement and at least some of the already generated titles. This scheme generates titles for a portion of the levels matching the user's requirements each time; furthermore, during the generation of multi-level titles, the already generated titles are modified according to the user's requirements to improve the final generated outline's fit with the user's needs. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0010] Figure 1 This is a flowchart illustrating one implementation method of the objective outline generation method of this application;

[0011] Figure 2 This is a flowchart illustrating step S101 corresponding to one embodiment.

[0012] Figure 3 This is a schematic diagram illustrating an application scenario for the implementation of Method 1 for generating the objective outline of this application;

[0013] Figure 4 This is a flowchart of an embodiment prior to step S102;

[0014] Figure 5 This is a flowchart of an implementation method following step S302;

[0015] Figure 6 This is a schematic diagram illustrating another application scenario of the target outline generation method of this application;

[0016] Figure 7 This is a flowchart of another implementation method following step S302;

[0017] Figure 8 This is a flowchart of an embodiment following step S102;

[0018] Figure 9 This is a flowchart of an implementation method following step S602;

[0019] Figure 10 This is a schematic diagram of one embodiment of the target outline generation system of this application;

[0020] Figure 11 This is a schematic diagram of the structure of one embodiment of the mobile terminal of this application;

[0021] Figure 12 This is a schematic diagram of one embodiment of the computer-readable storage medium of this application. Detailed Implementation

[0022] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0023] In this paper, the terms "system" and "network" are often used interchangeably. The term "and / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Additionally, the character " / " generally indicates that the preceding and following related objects have an "or" relationship. Furthermore, "many" in this paper means two or more.

[0024] The target outline generation method provided in this application is used to generate text outlines that meet the needs of users. The execution subject is a processor or server that can obtain user input and has text processing functions. The text type can be at least one of official documents, academic papers and graduation theses. This application will not list them all or make specific restrictions.

[0025] Please see Figure 1 , Figure 1 This is a flowchart illustrating one implementation method of the objective outline generation method of this application. The method includes:

[0026] S101: Obtain the user's first requirement and, based on the first requirement, generate at least one title corresponding to the root level.

[0027] Specifically, to improve the efficiency of text writing, it is necessary to obtain the user's primary needs before writing. These primary needs may include at least one of the following: the topic, type, and content information of the article the user needs to write.

[0028] In one application approach, the user's input request text is obtained, resulting in the user's initial request.

[0029] In another application, the user's voice input of their needs is obtained, and the voice input is converted into text to obtain the user's first needs.

[0030] In another application, the user inputs a request image, and the request text is identified from the user input request image to obtain the first request input by the user.

[0031] In one application scenario, an input box is used to receive the user's initial request. In response to the user entering the request text in the input box, the user's initial request is obtained.

[0032] In another application scenario, the system uses an input box and a voice receiving interface to receive the user's first requirement. In response to the user entering the requirement text in the input box, the system obtains the user's first requirement. In response to the user's voice requirement input through the voice receiving interface, the system converts the voice requirement into the requirement text and displays it in the input box, thus obtaining the user's first requirement.

[0033] In another application scenario, the system uses an input box, a voice receiving interface, and an image receiving interface to receive the user's initial requirement. In response to the user entering the requirement text in the input box, the system obtains the user's initial requirement. In response to the user's voice input via the voice receiving interface, the system converts the voice into text and displays it in the input box, thus obtaining the user's initial requirement. In response to the user's image input via the image receiving interface, the system identifies the requirement text from the image and displays it in the input box, thus obtaining the user's initial requirement.

[0034] In a specific application scenario, to improve the accuracy of the requirement text, after the requirement is transcribed into text and displayed in the input box, a prompt message is generated to prompt the user to confirm whether the transcribed requirement text is accurate. After the requirement text on the requirement image is extracted and displayed in the input box, a prompt message is generated to prompt the user to confirm whether the recognized requirement text is accurate. The system also receives modifications made by the user to the requirement text in the input box based on the prompt message.

[0035] Furthermore, at least one title corresponding to the root level is generated based on the obtained first requirement. The title corresponding to the root level is the title of the highest level of the article the user is to write, and this title can be a word or a sentence composed of multiple words.

[0036] In one application, after obtaining the first requirement, a first-level heading generation requirement is derived based on it. This first-level heading generation requirement is then input into a pre-trained outline generation model. The model outputs at least one heading corresponding to the root level based on the first-level heading generation requirement. The output headings corresponding to the root level match the first requirement. The outline generation model is trained using outlines from historical texts to obtain at least one heading that matches the first-level heading generation requirement.

[0037] In one application scenario, the outline generation model is a large language model. Based on the first requirement, a model prompt is generated and provided to the large language model. The large language model is then required to output at least one title that matches the model prompt, so as to improve the relevance of the obtained title to the user's needs.

[0038] In a specific application scenario, the aforementioned large language model is a generative pre-trained Transformer model (GPT).

[0039] S102: In response to the second request obtained from user input, based on the second request and at least some of the generated titles, generate at least the title of the next level corresponding to the current level, until the target outline composed of titles of all levels is obtained.

[0040] Specifically, in response to having generated at least some headings, a second user input is obtained to continue generating headings at the next level corresponding to at least some of the headings, or headings at multiple levels under at least some of the headings, based on the generated headings. In response to obtaining the headings at all levels, the headings at all levels are combined into a target outline.

[0041] In one application scenario, after generating at least one heading corresponding to the root level in step S101, the heading corresponding to the root level is used as a first-level heading. A second user input requirement is obtained, and headings for the next level corresponding to at least some of the first-level headings are generated. Alternatively, in response to obtaining at least one first-level heading, headings for all levels under at least some of the first-level headings are directly generated according to the user requirement. In response to no longer obtaining a second user input requirement, all generated headings are combined into a target outline. Here, the level corresponding to a first-level heading is higher than the level corresponding to a second-level heading, and a second-level heading is a subheading of a corresponding first-level heading.

[0042] In another application scenario, in response to the already generated headings corresponding to some levels, the lowest level among all generated headings is taken as the current level, or any level among the generated headings is taken as the current level. Based on the second requirement input by the user, headings for at least some of the next level below the current level are generated; or, headings for all levels under at least some of the current level's headings are generated. All generated headings are then combined to form the target outline.

[0043] The target outline generation method proposed in this application, after generating at least one title corresponding to the root level matching the first requirement based on user input, in response to obtaining a second requirement input by the user, generates at least one title for the next level below the current level based on the second requirement and at least some of the already generated titles. This scheme generates titles for a portion of the levels matching the user's requirements each time; and, during the generation of multi-level titles, the already generated titles are modified according to the user's requirements to improve the final generated outline's fit with the user's needs.

[0044] In one application method, please refer to Figure 2 , Figure 2 This is a flowchart illustrating step S101 corresponding to one embodiment. This step includes:

[0045] S201: Obtain the first requirement from user input.

[0046] Specifically, the first requirement is to obtain the title of the article to be written from the user's input. The specific implementation process can be found in the corresponding implementation methods described above, and will not be repeated here.

[0047] For a specific application scenario, please refer to Figure 3 , Figure 3 This is a schematic diagram illustrating an application scenario for the implementation of Method 1 for generating the objective outline of this application. In response to the user entering their first requirement in input box 100, the method retrieves the user's first requirement.

[0048] S202: Based on the first requirement, generate at least one candidate title corresponding to the root level.

[0049] Specifically, based on the obtained first requirement, at least one candidate title corresponding to the root level is generated.

[0050] In one application approach, based on a primary requirement, a reference outline related to that requirement is obtained. Based on the reference outline, at least one candidate title corresponding to the root level is obtained.

[0051] In one application scenario, the above-mentioned candidate title generation method can also refer to the title generation method corresponding to the root level in step S101, which will not be elaborated here.

[0052] In another application scenario, at least one reference article related to the primary requirement is obtained. The titles of these articles are extracted to create a reference outline composed of multiple titles. The titles corresponding to the highest level of this reference outline are then used as candidate titles corresponding to the primary requirement. These reference articles can be pre-stored in memory or a relevant database, or they can be obtained from the internet. The methods for obtaining reference articles related to the primary requirement can include keyword matching or semantic matching. Combining the reference articles with the root-level titles improves the accuracy of the obtained titles.

[0053] In another application scenario, select some headings related to the primary requirement from the highest-level headings in the reference outline as the aforementioned candidate headings. Alternatively, select headings related to the primary requirement from the highest-level headings in the reference outline as the aforementioned candidate headings.

[0054] In another application scenario, a predetermined number of reference articles related to the primary requirement are obtained. Titles are extracted from all reference articles to obtain a list of the highest-level reference titles for each article. Based on these multiple lists of reference titles, the aforementioned candidate titles are derived.

[0055] In a specific application scenario, identical titles from multiple reference title lists are directly used as candidate titles; related or similar titles from multiple reference title lists are merged, and the merged title is used as the remaining candidate title.

[0056] Additionally, the candidate titles are sorted according to their order in the reference title list; alternatively, the candidate titles are input into a pre-trained title sorting model, which outputs the sorted candidate titles. The title sorting model is trained using the highest-level title order in historical text to sort the input titles.

[0057] Furthermore, the generated sorted candidate titles are displayed on the display interface of the corresponding device.

[0058] In one application scenario, candidate titles are displayed in an editable box on the display interface, allowing users to modify the candidate titles within the editable box.

[0059] For a specific application scenario, please refer to [link / reference]. Figure 3 The candidate titles are generated and displayed in the display box 101, which is an editable box.

[0060] S203: Generate a first prompt text to prompt the user to modify the candidate title.

[0061] Specifically, after generating at least one candidate title, a first prompt text is generated to prompt the user to modify the generated candidate title, such as "Do you need to make a modification?".

[0062] For a specific application scenario, please refer to [link / reference]. Figure 3 Generate the first prompt text and simultaneously display the first prompt text in the display box 101 that is displayed in the candidate title.

[0063] Of course, in other specific application scenarios, the first prompt text can also be displayed separately from the candidate titles. That is, after the generated candidate titles are displayed in the edit box, the first prompt text is generated and displayed in the prompt box.

[0064] S204: In response to receiving a first modification request from the user based on the first prompt text input, modify the candidate titles at the root level based on the first modification request until at least one title corresponding to the root level is obtained.

[0065] In one application approach, the user's initial modification request may include at least one of the following: an addition request, a deletion request, and a reordering request. The addition request includes the user's desire to add at least one more title based on the at least one candidate title already generated; the deletion request includes the user's desire to delete at least one candidate title based on the generated candidate titles; and the reordering request includes the user's desire to reorder the generated candidate titles.

[0066] In a specific application scenario, in response to the user's first modification request entered in the input box 100, "VII. Acknowledgements" is added to the generated candidate title based on the user's first modification request and displayed in the display box 101.

[0067] Furthermore, the modified candidate title is used as the final title corresponding to the root level. Alternatively, in response to user confirmation, the modified candidate title is used as the final title corresponding to the root level.

[0068] Of course, in other application scenarios, after modifying the candidate titles at the root level based on the first modification request, users can continue to input the first modification request based on the modified candidate titles to further modify the modified candidate titles.

[0069] The above solution, after generating candidate titles corresponding to the root level, modifies the candidate titles according to user needs to improve the relevance of the root level titles to the user's primary needs.

[0070] In one application method, please refer to Figure 4 , Figure 4This is a flowchart illustrating an implementation method prior to step S102. The steps prior to S102 specifically include:

[0071] S301: In response to obtaining the title of the current level, generate a second prompt text to prompt the user to generate at least the title of the next level corresponding to the current level.

[0072] Specifically, after generating the title of the current level, a second prompt text is generated to indicate to the user the title of the next level or multiple levels below the current level that can be generated.

[0073] In a specific application scenario, the current level is the root level. After obtaining at least one title corresponding to the root level, a second prompt text is generated: "Do you want to generate the title of the next level below the root level title?" This prompts the user to generate the title of the next level below the current level. Alternatively, a second prompt text is generated: "Do you want to generate the titles of all levels below the root level title?"

[0074] In other specific application scenarios, the current level can also be the second level, the third level, or the fourth level, etc. Among them, the second level is the level below the root level, and the third level is the level below the second level.

[0075] S302: Obtain the user's second request based on the second prompt text input.

[0076] Specifically, in response to the user inputting a second request on the relevant device based on the second prompt text, the second request input by the user is obtained.

[0077] In one application method, please refer to Figure 5 , Figure 5 This is a flowchart illustrating an implementation method following step S302. Step S302 specifically includes:

[0078] S401: In response to receiving a second request from user input, and the second request includes a request to generate a next-level heading, based on the second request, obtain at least one heading to be refined from the headings of the current level.

[0079] In one application approach, in response to the second requirement being that only the generation requirement for generating the next level of headings is included, at least one heading to be refined is obtained from the headings of the current level based on the second requirement.

[0080] In one application scenario, the title to be refined can be included in the second requirement. For example, the current level of titles includes Title A, Title B, and Title C, etc. In response to the user's second requirement, "Give me the title of the next level corresponding to Title A", Title A is taken as the title to be refined, and the title of the next level of Title A is generated. Among them, there can be one or more titles of the next level of Title A.

[0081] In another application scenario, the titles to be refined can also be selected by the user. For example, if the user's second request is "Generate the next level of titles," in response to the user's selection action on the display interface of the corresponding device, the selected titles of the current level are used as the titles to be refined. Specifically, after obtaining the user's second request, a selection box is displayed before each title of the current level, and the user triggers the selection action by clicking the selection box; or, in response to the user's selection action such as marking or circling on some titles of the current level, the titles corresponding to the user's selection action are used as the titles to be refined.

[0082] In another application, based on the second requirement and the reference outline template, at least one heading to be refined is obtained from the headings at the current level.

[0083] Specifically, the current level of headings is compared with the reference outline template to identify the headings that need further refinement within the current level of headings.

[0084] In one application scenario, the process of obtaining the aforementioned reference outline template includes: responding to the user's input of a first requirement, generating at least one title corresponding to the root level based on the first requirement, and then obtaining a reference outline template related to the first requirement. The reference outline template includes titles that need to be refined, as well as titles that do not need to be refined; for example, for various types of articles, the first title is generally "Introduction," and the final title is generally "Acknowledgements," which do not require further refinement; while the titles in the middle sections often contain more sub-arguments.

[0085] In a specific application scenario, after obtaining the initial requirement, a template generation requirement is derived based on it. This requirement is then input into a pre-trained template generation model, which outputs a reference outline template that matches the initial requirement. The template generation model is trained using outlines from different types of training texts to obtain reference outline templates of a type that match the initial requirement.

[0086] S402: Based on the second requirement and at least some of the generated headings, generate at least the heading for the next level corresponding to the current level.

[0087] Specifically, based on at least some of the already generated titles corresponding to the current level and the levels above it, generate the title for the next level corresponding to the title to be refined.

[0088] In one application approach, the next level heading is generated only based on the heading to be refined.

[0089] In another application, the next level heading corresponding to the heading to be refined is generated based on the heading to be refined at the current level and the headings at some levels above it. For example, if the root level heading, second-level headings, and third-level headings have already been generated, and the heading to be refined is a heading within the third-level headings, then the next level heading to be refined is generated based on the heading to be refined, its corresponding second-level heading, and the root level heading.

[0090] For a specific application method, please refer to Figure 6 , Figure 6 This is a schematic diagram illustrating another implementation of the outline generation method of this application. In response to the user entering a second requirement in the input box 100, and the second requirement being to generate the titles of all the next level titles below the root level titles, the generated next level titles are displayed in the display box 101.

[0091] Furthermore, a third prompt text is generated to prompt the user to modify the next level of the title corresponding to the refined title. In response to receiving a second modification request from the user based on the third prompt text, the next level of the title corresponding to the refined title is modified based on the second modification request.

[0092] The specific implementation process of modifying the next level title corresponding to the title to be refined based on the second modification requirement can refer to the process of modifying the generated root level candidate titles in steps S203 and S204 above, and will not be elaborated in detail here.

[0093] In another application, in response to detecting that a user has deleted all sub-level headings corresponding to any heading to be refined, a first termination marker is generated for the corresponding heading to be refined. The first termination marker is used to mark the corresponding heading to be refined as the lowest level heading.

[0094] In a specific application scenario, the current level of headings includes Heading A, Heading B, Heading C, and Heading D. Based on the user's second requirement, Heading B and Heading C are designated as headings to be refined, and the next level of headings corresponding to Heading B and Heading C are generated. If the user inputs a command to delete all headings at the next level of Heading B, it is assumed that the user defaults to no longer needing to generate headings below Heading B, and Heading B is assigned a first termination marker. Furthermore, when generating headings at the next level or all levels below some headings based on user requirements in the future, headings below Heading B will no longer be generated.

[0095] In another application method, please refer to Figure 7 , Figure 7 This is a flowchart illustrating another implementation method following step S302. Step S302 specifically includes:

[0096] S501: In response to a second requirement obtained from user input, wherein the second requirement includes the requirement to generate titles for all levels below the current level, based on the second requirement, at least one title to be refined is obtained from the titles of the current level.

[0097] Specifically, in response to the obtained second requirement, which includes the generation requirement for all levels of headings, at least one heading to be refined is obtained from the headings of the current level based on the second requirement. The method for obtaining the heading to be refined can refer to the corresponding application methods described above.

[0098] In one application approach, the title to be refined can be included in the second requirement. For example, the current level of titles includes Title A, Title B, and Title C, etc. In response to the user's second requirement being: "Give me titles for all levels corresponding to Title A," Title A is taken as the title to be refined, and titles for all levels of Title A are generated. Title A includes at least one title at one level, or titles corresponding to multiple levels.

[0099] In another application method, the title to be refined can also be selected by the user. The specific process can be referred to the corresponding application method mentioned above.

[0100] In another application method, based on the second requirement and the reference outline template, at least one heading to be refined is obtained from the headings at the current level. The specific process can be referenced from the corresponding application methods described above.

[0101] S502: Based on the second requirement and at least some of the generated headings, generate at least the heading for the next level corresponding to the current level.

[0102] Specifically, based on at least some of the already generated titles corresponding to the current level and the levels above it, titles for all levels corresponding to the titles to be refined are generated.

[0103] In one application approach, headings for all levels below the heading to be refined are generated only based on the heading to be refined.

[0104] In another application, based on the current level's heading to be refined and the headings of some levels above it, the next level's heading corresponding to the heading to be refined is generated. For example, if a root level heading, second-level headings, and third-level headings have already been generated, and the heading to be refined is a third-level heading, then based on the heading to be refined, its corresponding second-level heading, and the root level heading, all levels of headings below the heading to be refined are generated. These generated levels of headings below the heading to be refined can include one level of headings, two levels of headings, etc., depending on the specific circumstances.

[0105] Furthermore, a fourth prompt text is generated to prompt the user to modify the titles at all levels corresponding to the refined title. In response to receiving a third modification request from the user based on the fourth prompt text, the titles at at least some levels corresponding to the refined title are modified based on the third modification request.

[0106] The specific implementation process of modifying the titles of at least some levels below the title corresponding to the title to be refined based on the third modification requirement can also refer to the process of modifying the generated candidate titles of the root level in steps S203 and S204 above, and will not be elaborated in detail here.

[0107] In another application, in response to detecting that the user has confirmed all levels of headings corresponding to any heading to be refined, a second termination marker is generated for the corresponding heading to be refined. The second termination marker is used to indicate that all levels of headings have been generated for the corresponding heading to be refined.

[0108] In a specific application scenario, the current level of headings includes Heading A, Heading B, Heading C, and Heading D. Based on the user's second requirement, Heading B and Heading C are selected as headings to be refined, and headings for all levels corresponding to Heading B and Heading C are generated. If the user confirms all headings at all levels under Heading B, then the headings at all levels under Heading B generated for the second requirement are considered to meet the user's needs and require no modification. A second termination marker is then assigned to Heading B.

[0109] It should be noted that users can confirm all levels of headings corresponding to any heading to be refined, or they can confirm all levels of headings corresponding to multiple headings to be refined.

[0110] Please see Figure 8 , Figure 8 This is a flowchart illustrating an implementation method following step S102. Step S102 further includes:

[0111] S601: Obtain recommended text from the target outline that matches at least some of the headings.

[0112] Specifically, in response to generating a target outline through the above steps, recommended text that matches some of the headings in the target outline is generated.

[0113] In one application, after obtaining the target outline, a recommendation text generation requirement is derived based on the target outline. This requirement is then input into a pre-trained recommendation text acquisition model to receive recommendation text corresponding to at least a portion of the titles. The recommendation text acquisition model is trained using training text and corresponding titles or general topics to acquire at least a portion of the recommendation text that semantically or keywordly matches some titles in the target outline.

[0114] In one application scenario, the recommended text retrieval model can generate recommended text only for the lowest-level headings in the target outline, or it can generate recommended text for all headings in the target outline.

[0115] In another application scenario, after generating the target outline, the user's third requirement is obtained. Based on the third requirement and the target outline, the recommendation text generation requirement is obtained. The recommendation text generation requirement is then input into a pre-trained recommendation text acquisition model to receive the recommendation text corresponding to the partial titles that match the user's third requirement.

[0116] In another application scenario, the recommended text acquisition model is a large language model. Based on the recommended text generation requirements, model prompts are generated and provided to the large language model. The large language model is required to output at least part of the recommended text that matches the recommended text generation requirements, so as to provide corresponding content references for the generated title, thereby improving the efficiency of users' subsequent writing.

[0117] S602: Generate an attachment identifier corresponding to the recommended text after the corresponding title. The recommended text is displayed after the user selects the attachment identifier.

[0118] Specifically, after obtaining the recommended text, it is displayed hidden as an attachment icon. The recommended text is only displayed on the relevant device's interface after receiving the user's selection instruction. Users can generate the corresponding selection instruction by clicking the attachment icon on the displayed interface.

[0119] The above method avoids displaying a large amount of text content directly on the display interface. Users can select the recommended text they need to know for display, which improves the efficiency of users' information access.

[0120] Of course, in other applications, the generated recommendation text can also be displayed directly below the corresponding title for user reference.

[0121] In one application scenario, the attachment identifier for the recommended text is positioned relative to the title of that recommended text. Alternatively, the attachment identifier can be placed on the title, allowing users to click the title to display the corresponding recommended text.

[0122] Please see Figure 9 , Figure 9 This is a flowchart illustrating an embodiment following step S602. Step S602 further includes:

[0123] S701: Based on all recommended texts, adjust the titles matched by at least a portion of the recommended texts according to a preset adjustment method. The preset adjustment method includes at least one of the following: adding a next-level title, adjusting the content of the title, and adjusting the order of the titles.

[0124] Specifically, based on at least part of the recommended text, the matched titles are adjusted to improve the accuracy of the obtained target outlines.

[0125] In one application approach, adjustment requirements are generated based on the recommended text and its corresponding title. These requirements are then input into a pre-trained adjustment model, which outputs the adjustment results based on these requirements. The adjustment model comprises a first adjustment module, a second adjustment module, and a third adjustment module.

[0126] In one application scenario, the first adjustment module in the adjustment model summarizes or extracts keywords from the recommended text to obtain at least one corresponding keyword, and uses the obtained summarized keyword as the next level title of the title corresponding to the recommended text.

[0127] In another application scenario, the second adjustment module in the adjustment model supplements the matching title based on the content of the recommended text, so as to summarize part of the content of the recommended text into the corresponding title, thereby improving the completeness of the corresponding title.

[0128] In another application scenario, the third adjustment module in the adjustment model extracts events from the recommended text corresponding to the same level of titles, and adjusts the order of the extracted events according to the order of event development. Based on the adjusted event order, the order of the titles at that level is then adjusted.

[0129] S702: Generate a fifth prompt text to prompt the user to confirm the revised target outline.

[0130] Specifically, in response to adjustments made to the target outline based on the recommended text, a fifth prompt text is generated to remind the user to modify the adjusted target outline.

[0131] Furthermore, in response to the fourth requirement obtained from the user's input of the fifth prompt text, the adjusted target outline is modified based on the fourth requirement.

[0132] The specific process of generating the fifth prompt text and modifying the adjusted target outline based on the fourth requirement can be referred to the corresponding implementation method described above.

[0133] S703: In response to receiving a confirmation instruction from the user based on the fifth prompt text, display the adjusted target outline.

[0134] Specifically, in response to receiving a confirmation instruction from the user based on the fifth prompt text input, the adjusted target outline is displayed on the display interface of the corresponding device to provide a reference for the user's writing, thereby improving the user's writing efficiency.

[0135] Please see Figure 10 , Figure 10 This is a schematic diagram of one embodiment of the target outline generation system of this application. The target outline generation system includes: a first processing module 10 and a second processing module 20.

[0136] Specifically, the first processing module 10 is used to obtain the first requirement input by the user, and based on the first requirement, generate at least one title corresponding to the root level.

[0137] In one application mode, the first processing module 10 is used to generate at least one candidate title corresponding to the root level based on a first requirement; generate a first prompt text to prompt the user to modify the candidate title; and in response to obtaining the user's first modification request based on the first prompt text, modify the candidate title at the root level based on the first modification request until at least one title corresponding to the root level is finally obtained.

[0138] The process of generating at least one candidate title corresponding to the root level based on the first requirement includes: obtaining a reference outline related to the first requirement based on the first requirement; obtaining at least one candidate title corresponding to the root level based on the reference outline; and displaying all candidate titles using an editable box.

[0139] In response to the second request received from user input, the second processing module 20 generates at least the next level's title corresponding to the current level based on the second request and at least some of the already generated titles, until a target outline composed of titles from all levels is obtained.

[0140] In one application method, in response to obtaining a second request input by the user, before the second processing module 20 generates at least the title of the next level corresponding to the current level based on the second request and at least part of the generated titles, the method further includes: in response to obtaining the title of the current level, generating a second prompt text to prompt the user to generate at least the title of the next level corresponding to the current level; and obtaining the second request input by the user based on the second prompt text.

[0141] In one application scenario, after the second processing module 20 obtains the second requirement input by the user based on the second prompt text, it further includes: in response to obtaining the second requirement input by the user, and the second requirement includes the requirement to generate the next level title, obtaining at least one title to be refined from the title of the current level based on the second requirement.

[0142] Specifically, based on the second requirement and at least some of the generated titles, at least the next level title corresponding to the current level is generated, including: based on at least some of the generated titles corresponding to the current level and the levels above it, generating the next level title corresponding to the title to be refined.

[0143] The process includes, after generating the title of the next level corresponding to the title to be refined based on at least some of the titles already generated at the current level and the levels above it, the method further includes: generating a third prompt text to prompt the user to modify the title of the next level corresponding to the title to be refined; in response to obtaining a second modification request input by the user based on the third prompt text, modifying the title of the next level corresponding to the title to be refined based on the second modification request; in response to detecting that the user deletes all the titles of the next level corresponding to any title to be refined, generating a first termination marker for the corresponding title with refinement; wherein the first termination marker is used to mark the corresponding title to be refined as the title of the lowest level.

[0144] In another application scenario, after the second processing module 20 obtains the second requirement input by the user based on the second prompt text, it further includes: in response to obtaining the second requirement input by the user, and the second requirement includes the generation requirement of all level titles, obtaining at least one title to be refined from the title of the current level based on the second requirement.

[0145] Specifically, based on the second requirement and at least some of the generated titles, at least the title of the next level corresponding to the current level is generated, including: based on at least some of the generated titles corresponding to the current level and the levels above it, generating titles for all levels corresponding to the titles to be refined.

[0146] The process, after generating titles for all levels corresponding to the title to be refined based on at least some of the already generated titles corresponding to the current level and the levels above it, further includes: generating a fourth prompt text to prompt the user to modify the titles for all levels corresponding to the title to be refined; in response to obtaining a third modification request input by the user based on the fourth prompt text, modifying the titles for at least some levels corresponding to the title to be refined based on the third modification request; and in response to detecting that the user confirms the titles for all levels corresponding to any title to be refined, generating a second termination marker for the corresponding title to be refined; wherein the second termination marker is used to mark that the corresponding title to be refined has generated titles for all levels.

[0147] Please continue reading Figure 10 The target outline generation system proposed in this application also includes a first acquisition module 11 coupled to the first processing module 10. After acquiring the first requirement input by the user and generating at least one title corresponding to the root level based on the first requirement, the first acquisition module 11 is used to: acquire a reference outline template related to the first requirement based on the first requirement.

[0148] The first processing module 10 obtains at least one title to be refined from the current level of titles based on the second requirement, including: obtaining at least one title to be refined from the current level of titles based on the second requirement and the reference outline template.

[0149] Please continue reading Figure 10 The target outline generation system proposed in this application also includes a second acquisition module 21 coupled to the second processing module 20. After obtaining the target outline composed of titles at all levels, the second acquisition module 21 is used to acquire recommended text in the target outline that matches at least some of the titles; generate an attachment identifier corresponding to the recommended text after the corresponding title; wherein, the recommended text is displayed after the attachment identifier is selected by the user.

[0150] In one application method, after obtaining the recommended text that matches at least some of the titles in the target outline, the method further includes: adjusting the titles that match at least some of the recommended text according to a preset adjustment method based on all the recommended text; wherein the preset adjustment method includes at least one of adding a next-level title, adjusting the content of the title, and adjusting the order of the titles; generating a fifth prompt text to prompt the user to confirm the adjusted target outline; and displaying the adjusted target outline in response to obtaining the confirmation instruction input by the user based on the fifth prompt text.

[0151] Please see Figure 11 , Figure 11This is a schematic diagram of the structure of a mobile terminal according to an embodiment of this application. The mobile terminal includes a display screen 30, a memory 40, and a processor 50. The display screen 30 and the memory 40 are respectively coupled to the processor 50. The display screen 30 is used at least to display content to the user and to allow the user to input or select content. The memory 40 stores program instructions, and the processor 50 is used to execute the program instructions to implement the time-generating method mentioned in any of the above embodiments. Specifically, electronic devices include, but are not limited to, desktop computers, laptops, tablets, servers, etc., and are not limited here. In addition, the processor 50 may also be called a CPU (Center Processing Unit). The processor 50 may be an integrated circuit chip with signal processing capabilities. The processor 50 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor may be a microprocessor or any conventional processor. In addition, the processor 50 may be implemented by integrated circuit chips.

[0152] Please see Figure 12 , Figure 12 This is a schematic diagram of a computer-readable storage medium according to an embodiment of the present application. The computer-readable storage medium 60 stores program instructions 70 that can be executed by a processor. When the program instructions 70 are executed by the processor, they implement the time summary generation method mentioned in any of the above embodiments.

[0153] In the several embodiments provided in this application, it should be understood that the disclosed methods and apparatus can be implemented in other ways. For example, the apparatus implementations described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0154] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0155] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0156] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods of various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0157] The above description is merely an embodiment of this application and does not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for generating a target outline, characterized in that, include: The system obtains the user's first requirement and generates at least one title corresponding to the root level based on the first requirement; wherein, the first requirement includes at least one of the topic, type and content information of the article written by the user, and the at least one title corresponding to the root level is generated using a pre-trained outline generation model. In response to receiving a second request from the user, based on the second request and at least some of the generated titles, at least the next level title corresponding to the current level is generated, until a target outline composed of titles from all levels is obtained; wherein, the second request is input by the user based on a second prompt text, the second prompt text being used to prompt the user to generate the next level title corresponding to the current level; The method further includes, after obtaining the second requirement input by the user based on the second prompt text, obtaining at least one title to be refined from the current level of titles based on the second requirement, in response to obtaining the second requirement input by the user, and the second requirement including the requirement to generate the next level of titles. The step of generating at least the next level title corresponding to the current level based on the second requirement and at least some of the generated titles includes: generating the next level title corresponding to the title to be refined based on at least some of the generated titles corresponding to the current level and the levels above it.

2. The method according to claim 1, characterized in that, The step of generating at least one title corresponding to the root level based on the first requirement includes: Based on the first requirement, at least one candidate title corresponding to the root level is generated; Generate a first prompt text to prompt the user to modify the candidate title; In response to receiving a first modification request from the user based on the first prompt text input, the candidate titles at the root level are modified based on the first modification request until at least one title corresponding to the root level is finally obtained.

3. The method according to claim 2, characterized in that, The step of generating at least one candidate title corresponding to the root level based on the first requirement includes: Based on the first requirement, obtain a reference outline related to the first requirement; Based on the reference outline, at least one candidate title corresponding to the root level is obtained, and all the candidate titles are displayed using an editable box.

4. The method according to claim 1, characterized in that, Before generating at least the title of the next level corresponding to the current level based on the second request received from the user and at least a partially generated title, the method further includes: In response to obtaining the title of the current level, generate a second prompt text to prompt the user to generate at least the title of the next level corresponding to the current level; Obtain the second requirement from the user based on the second prompt text.

5. The method according to claim 1, characterized in that, After obtaining the second requirement from the user based on the second prompt text input, the method further includes: In response to a second request received from user input, and the second request includes a request to generate headings at all levels, at least one heading to be refined is obtained from the headings at the current level based on the second request. The step of generating at least the next level's title based on the second requirement and at least some of the already generated titles includes: Based on at least some of the already generated titles corresponding to the current level and the levels above it, generate titles for all levels corresponding to the title to be refined.

6. The method according to claim 5, characterized in that, After obtaining the user's input first requirement, and generating at least one title corresponding to the root level based on the first requirement, the method further includes: Based on the first requirement, obtain a reference outline template related to the first requirement; Based on the second requirement, obtaining at least one title to be refined from the current level of titles includes: Based on the second requirement and the reference outline template, at least one title to be refined is obtained from the titles at the current level.

7. The method according to claim 1, characterized in that, After generating the title of the next level corresponding to the title to be refined based on at least some of the already generated titles corresponding to the current level and the levels above it, the method further includes: Generate a third prompt text to prompt the user to modify the next level of title corresponding to the title to be refined; In response to receiving a second modification request from the user based on the third prompt text, the title of the next level corresponding to the title to be refined is modified based on the second modification request; In response to detecting that a user deletes all next-level titles corresponding to any of the titles to be refined, a first termination marker is generated for the corresponding title to be refined; wherein the first termination marker is used to mark the corresponding title to be refined as the lowest level title.

8. The method according to claim 5, characterized in that, After generating the titles for all levels corresponding to the title to be refined based on at least some of the already generated titles corresponding to the current level and the levels above it, the method further includes: Generate a fourth prompt text to prompt the user to modify the titles at all levels corresponding to the title to be refined; In response to receiving a third modification request from the user based on the fourth prompt text input, the title at least some levels corresponding to the title to be refined is modified based on the third modification request; In response to detecting that a user has confirmed all levels of titles corresponding to any of the titles to be refined, a second termination marker is generated for the corresponding title to be refined; wherein the second termination marker is used to mark that the corresponding title to be refined has generated all levels of titles.

9. The method according to claim 1, characterized in that, After obtaining the target outline consisting of headings at all levels, the process also includes: Obtain recommended text from the target outline that matches at least some of the headings; An attachment identifier corresponding to the recommended text is generated after the corresponding title; wherein, the recommended text is displayed after the user selects the attachment identifier.

10. The method according to claim 9, characterized in that, After obtaining the recommended text in the target outline that matches at least some of the titles, the method further includes: Based on all the recommended texts, the titles matched by at least some of the recommended texts are adjusted according to a preset adjustment method; wherein, the preset adjustment method includes at least one of adding a next-level title, adjusting the content of the title, and adjusting the order of the titles; Generate a fifth prompt text to remind the user to confirm the revised target outline; In response to receiving a confirmation instruction from the user based on the fifth prompt text, the adjusted target outline is displayed.

11. A target outline generation system, characterized in that, include: The first processing module is used to obtain the first requirement input by the user and generate at least one title corresponding to the root level based on the first requirement; wherein, the first requirement includes at least one of the topic, type and content information of the article written by the user, and the at least one title corresponding to the root level is generated using a pre-trained outline generation model. The second processing module, in response to receiving a second request input by the user, is configured to generate at least the next level title corresponding to the current level based on the second request and at least some of the generated titles, until a target outline composed of titles from all levels is obtained; wherein, the second request is input by the user based on a second prompt text, the second prompt text being used to prompt the user to generate the next level title corresponding to the current level; The method further includes, after obtaining the second requirement input by the user based on the second prompt text, obtaining at least one title to be refined from the current level of titles based on the second requirement, in response to obtaining the second requirement input by the user, and the second requirement including the requirement to generate the next level of titles. The step of generating at least the next level title corresponding to the current level based on the second requirement and at least some of the generated titles includes: generating the next level title corresponding to the title to be refined based on at least some of the generated titles corresponding to the current level and the levels above it.

12. A mobile terminal, characterized in that, include: The device includes a display screen, a memory, and a processor, wherein the display screen and the memory are respectively coupled to the processor, the display screen is used at least to display content to a user and to allow the user to input or select content, the memory stores program instructions, and the processor is used to execute the program instructions to implement the target outline generation method as described in any one of claims 1 to 10.

13. A computer-readable storage medium having program instructions stored thereon, characterized in that, When the program instructions are executed by the processor, they implement the target outline generation method as described in any one of claims 1 to 10.

Images