Text abstract generation method and device, equipment, storage medium and product

Abstract

The invention discloses a text abstract generation method and device, equipment, a storage medium and a product, and relates to the technical field of computers, and the text abstract generation method comprises the steps that an initial text is acquired; according to a preset structured template, cue word conversion operation is carried out on the initial text to obtain structured cue words, and the structured cue words reflect semantics and structures of the initial text; and sending the structured cue word to the large language model, and receiving text summary information output by the large language model. According to the method and the device, the prompt word conversion operation is carried out on the initial text to be converted into the text abstract, so that the structured prompt word reflecting the semantics and the structure of the initial text is obtained, and the structured prompt word reflects the semantics and the structure of the initial text; the large language model can comprehensively consider the full-text structure and semantics of the initial text according to the structured cue word, so that the accuracy of text abstract generation is improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a text summary generation method, device, equipment, storage medium and product. Background Art

[0002] At present, with the development of computer technology, the Internet and digital information are growing explosively. Users are faced with the challenge of reading and processing large amounts of multilingual information, which puts forward the need to quickly and efficiently obtain and understand large amounts of text information. Among them, text summarization can greatly improve the efficiency of information acquisition by displaying the text content in a concise, accurate and structured manner. Therefore, how to generate text summaries is a technical problem that urgently needs to be solved.

[0003] In the related art, key sentences or phrases are usually extracted from the original text to generate a summary. However, this extractive summary generation method does not comprehensively consider the semantics and structure of the text, and the extracted sentences often cannot accurately represent the full text content, resulting in low accuracy of the generated text summary.

[0004] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of this application is to provide a text summary generation method, aiming to solve the technical problem of low accuracy of the generated text summary.

[0006] To achieve the above purpose, the present application proposes a text summary generation method, which is applied to a summary generation platform. A large language model is deployed in the summary generation platform. The text summary generation method includes:

[0007] Get the initial text;

[0008] According to a preset structured template, the initial text is converted into a prompt word to obtain a structured prompt word, wherein the structured prompt word reflects the semantics and structure of the initial text;

[0009] The structured prompt word is sent to the large language model, and text summary information output by the large language model is received, wherein the text summary information is generated by the large language model performing summary text conversion on the structured prompt word based on a summary output instruction.

[0010] Optionally, the step of performing a prompt word conversion operation on the initial text according to a preset structured template to obtain a structured prompt word includes:

[0011] Extracting attribute word information of each level from a preset structured template, wherein the attribute word information is a keyword identifying different categories of parts in the structured prompt word;

[0012] The initial text and the attribute word information are sent to the large language model, and the structured prompt words output by the large language model are received, wherein the structured prompt words are generated by the large language model extracting the description text corresponding to the attribute word information from the initial text based on the prompt word output instruction, obtaining the target semantic description text, and converting the target semantic description text into the prompt words.

[0013] Optionally, the step of sending the structured prompt word to the large language model and receiving text summary information output by the large language model includes:

[0014] Receive demand information input by a user and convert the demand information into demand parameters;

[0015] The structured prompt word and the restriction parameter are sent to the large language model, and text summary information output by the large language model is received, wherein the text summary information is generated by the large language model performing summary text conversion on the structured prompt word based on a summary output instruction and the requirement parameter.

[0016] Optionally, after the step of sending the structured prompt word to the large language model and receiving text summary information output by the large language model, the method includes:

[0017] Based on a preset context detection template, the text summary information is detected to obtain a detection result;

[0018] It is determined whether the detection result meets a preset detection standard. If the detection result does not meet the detection standard, the text summary information is adjusted based on the context detection template to obtain adjusted text summary information.

[0019] Optionally, after the step of sending the structured prompt word to the large language model and receiving text summary information output by the large language model, the method includes:

[0020] Displaying the text summary information on the user terminal;

[0021] Responding to a text editing instruction issued by a user at the user terminal, wherein the text editing instruction includes a start field, an end field, and editing operation information;

[0022] Integrate the fields between the start field and the end field to obtain a target edit field;

[0023] An editing operation corresponding to the editing operation information is performed on the target editing field to obtain edited text summary information.

[0024] Optionally, the step of performing an editing operation corresponding to the editing operation information on the target editing field to obtain edited text summary information includes:

[0025] Convert the target edit field and the edit operation information into a prompt word to obtain a target prompt word;

[0026] The target prompt word is sent to the large language model, and edited text summary information output by the large language model is received, wherein the edited text summary information is generated by the large language model performing an editing operation corresponding to the editing operation information on the target edit field based on a summary adjustment instruction.

[0027] In addition, to achieve the above purpose, the present application also proposes a text summary generation device, the text summary generation device comprising:

[0028] An acquisition module is used to acquire the initial text;

[0029] A conversion module, used to perform a prompt word conversion operation on the initial text according to a preset structured template to obtain a structured prompt word, wherein the structured prompt word reflects the semantics and structure of the initial text;

[0030] A generation module is used to send the structured prompt word to the large language model and receive text summary information output by the large language model, wherein the text summary information is generated by the large language model performing summary text conversion on the structured prompt word based on a summary output instruction.

[0031] In addition, to achieve the above-mentioned purpose, the present application also proposes a text summary generation device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, and the computer program is configured to implement the steps of the text summary generation method described above.

[0032] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the text summary generation method described above are implemented.

[0033] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the steps of the text summary generation method described above are implemented.

[0034] One or more technical solutions proposed in this application have at least the following technical effects:

[0035] Compared with the related art, which usually extracts key sentences or phrases from the original text to generate a summary. However, the extractive summary generation method does not comprehensively consider the semantics and structure of the text, and the extracted sentences often cannot accurately represent the full text content, resulting in low accuracy of the generated text summary. In comparison, the present application uses a summary generation platform to perform a prompt word conversion operation on the initial text to be converted into a text summary, obtains a structured prompt word reflecting the semantics and structure of the initial text, and then sends the structured prompt word to the large language model, and receives the text summary information output by the large language model. It can be understood that on the one hand, the summary generation platform uses the deployed large language model to generate text summary information for the initial text, wherein the large language model can more accurately summarize the full text, and on the other hand, because the structured prompt word reflects the semantics and structure of the initial text, after the summary generation platform sends the structured prompt word to the large language model, the large language model can comprehensively consider the full text structure and semantics of the initial text according to the structured prompt word, so that the generated text summary information is more complete and fits the original text, thereby improving the accuracy of text summary generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0037] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0038] Figure 1 A flowchart of the first embodiment of the method for generating a text summary of the present application is provided;

[0039] Figure 2 This is a schematic diagram of a summary generation module in the text summary generation method of this application;

[0040] Figure 3 A flowchart of the second embodiment of the method for generating a text summary of the present application is provided;

[0041] Figure 4This is a schematic diagram of the JSON structure of the summary document in the text summary generation method of this application;

[0042] Figure 5 This is a schematic diagram of the AI ​​creation module in the text summary generation method of this application;

[0043] Figure 6 This is a schematic diagram of the module structure of the text summary generation device in the embodiment of the present application;

[0044] Figure 7 Schematic diagram of the device structure of the hardware operating environment involved in the text summary generation method in the embodiment of the present application.

[0045] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0046] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0047] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0048] The main solution of the embodiment of the present application is: obtaining an initial text; performing a prompt word conversion operation on the initial text according to a preset structured template to obtain a structured prompt word, wherein the structured prompt word reflects the semantics and structure of the initial text; sending the structured prompt word to the large language model, and receiving text summary information output by the large language model, wherein the text summary information is generated by the large language model performing a summary text conversion on the structured prompt word based on a summary output instruction.

[0049] In this embodiment, the summary generation platform is used as the execution entity. For the convenience of description, it will be described hereinafter as “platform”.

[0050] In the related art, key sentences or phrases are usually extracted from the original text to generate a summary. However, this extractive summary generation method does not comprehensively consider the semantics and structure of the text, and the extracted sentences often cannot accurately represent the full text content, resulting in low accuracy of the generated text summary.

[0051] This application provides a solution to achieve automatic generation of text summaries and improve the accuracy of text summary generation.

[0052] From the above embodiments, it can be seen that the present application uses a summary generation platform to perform a prompt word conversion operation on the initial text to be converted into a text summary, obtains a structured prompt word that reflects the semantics and structure of the initial text, and then sends the structured prompt word to the large language model, and receives the text summary information output by the large language model. It can be understood that on the one hand, the summary generation platform uses the deployed large language model to generate text summary information for the initial text, wherein the large language model can more accurately summarize the full text, and on the other hand, since the structured prompt word reflects the semantics and structure of the initial text, after the summary generation platform sends the structured prompt word to the large language model, the large language model can comprehensively consider the full text structure and semantics of the initial text based on the structured prompt word, so that the text summary information generated by it is more complete and fits the original text, thereby improving the accuracy of text summary generation.

[0053] It should be noted that the execution subject of this embodiment may be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device, a terminal system, etc. capable of realizing the above functions. The following takes the summary generation platform as an example to illustrate this embodiment and the following embodiments.

[0054] Based on this, the present application embodiment provides a text summary generation method, referring to Figure 1 , Figure 1 This is a flowchart of the first embodiment of the text summary generation method of this application.

[0055] In this embodiment, the text summary generation method includes steps S100 to S300:

[0056] Step S100, obtaining initial text;

[0057] It should be noted that the initial text refers to the text to be converted into a text summary. The text summary generation method of the present application is applied to a summary generation platform, in which a large language model is deployed. It is understandable that the above text summary generation platform is a tool or service that uses artificial intelligence technology, especially natural language processing (NLP) technology, to automatically compress long text into short summaries. These platforms can quickly extract key information from texts, generate summaries, help users save time, and improve reading and comprehension efficiency. The working principle of the text summary generation platform is usually based on language models, which learn a large amount of text data, master patterns, grammar and vocabulary, and then generate human-like text according to given prompts or inputs. There are two main ways to generate summaries: extractive and generative. The extractive summary is to form a summary by extracting key sentences from the original text, while the generative summary is based on the understanding of the original text and reorganizing the language to generate new summary content. The application scenarios of the text summary generation platform are very wide, including but not limited to content digestion, research efficiency improvement, enhanced understanding, resource optimization, reader participation improvement, content creation assistance, news updates, meeting preparation, and language learning. In general, the text summary generation platform uses advanced AI technology to provide users with an effective means to quickly obtain the core content of the text, which is suitable for a variety of scenarios and needs.

[0058] In a specific implementation, the platform obtains the initial text by the user inputting the initial text and uploading the initial file through the interactive page of the platform to obtain the initial text; or the platform obtains the initial text by selecting a knowledge base file as the original text for summary generation, which is not specifically limited here.

[0059] Step S200, performing a prompt word conversion operation on the initial text according to a preset structured template to obtain structured prompt words, wherein the structured prompt words reflect the semantics and structure of the initial text;

[0060] It should be noted that the structured template is a framework for structured prompt words, which helps users create high-quality prompt words in a standardized and modular way to guide large language models (LLMs) to generate accurate responses. The prompt word (Prompt) refers to the word or sentence after the query sentence is input when a person interacts with a large language model, which is used to guide the large model to output the response content required by the person. In the process of interaction between the user and the large model, the accuracy of the prompt word is crucial, and high-quality prompt words can guide the model to output high-quality content. The structured prompt word (StructuredPrompt) is a methodology for writing high-quality prompt words, which helps users interact with large language models at a deeper level in a modular and standardized way.

[0061] In a specific implementation, the structured prompt words reflect the semantics and structure of the initial text. After the summary generation platform sends the structured prompt words to the large language model, the large language model can comprehensively consider the full text structure and semantics of the initial text based on the structured prompt words, so that the text summary information generated by it is more complete and fits the original text, thereby improving the accuracy of text summary generation.

[0062] In a specific implementation, the platform performs a prompt word conversion operation on the initial text according to a preset structured template to obtain a structured prompt word, including:

[0063] Extracting attribute word information of each level from a preset structured template, wherein the attribute word information is a keyword identifying different categories of parts in the structured prompt word; sending the initial text and the attribute word information to the large language model, and receiving the structured prompt word output by the large language model, wherein the structured prompt word is generated by the large language model extracting the description text corresponding to the attribute word information from the initial text based on the prompt word output instruction, obtaining the target semantic description text, and converting the target semantic description text into the prompt word.

[0064] It should be noted that the attribute words are keywords or phrases used to identify and describe different parts of the prompt word. They are usually used as titles in the Markdown format to clearly organize and display the structure of the prompt word.

[0065] Among them, attribute words have the following characteristics and functions:

[0066] Identification: Attribute words use specific formats (such as #title in Markdown) to identify different parts of the prompt word, such as #Role, ##Skills, etc.

[0067] Semantic description: Each attribute word contains specific semantics, which describes the nature and content of the text paragraph or sentence that follows it. For example, the Skills attribute word is usually followed by a description of the skills or abilities that the model should have.

[0068] Organizational structure: Attribute words help organize the structure of the prompt word, making the various parts of the prompt word clearer and more orderly, making it easier to read and understand.

[0069] Guiding the model: Attribute words guide the model to a certain extent, helping it understand the intent and importance of each part, so as to better process and respond to user input.

[0070] Easy to expand and modify: Due to the clarity and identity of attribute words, users can more easily expand and modify prompt words to adapt to different needs and scenarios.

[0071] Improve readability: In longer or more complex prompts, the use of attribute words improves overall readability, allowing other users or developers to quickly grasp the structure and key points of the prompt.

[0072] Furthermore, attribute words include but are not limited to:

[0073] Role: defines the role or identity that the model plays in the interaction.

[0074] Profile: Provides background information or an overview of the character.

[0075] Skills: Skills describe the professional skills or abilities that a character possesses.

[0076] Rules: Set the rules or constraints that roles should follow during interactions.

[0077] Workflow: describes the steps or process for a role to complete a task.

[0078] Goals: Goals define what the character should achieve in the interaction.

[0079] Initialization: Initialization, setting the behavior or steps when the character starts to interact.

[0080] Understandably, attribute words play a vital role in structuring prompt words, and they help build a prompt word framework that is clear, organized, easy to understand and use.

[0081] In the specific implementation, the platform uses the deployed large language model and the system instructions of prompt engineering to enable the large language model to convert the generation requirements into structured prompt words based on the initial text and attribute word information without targeted fine-tuning, thereby guiding the input, processing and output of the summary generation results of the large model. Prompt engineering is designed to help the large model achieve high accuracy and relevance in summary generation tasks, and the quality of prompt engineering will have a great impact on the generation effect.

[0082] Furthermore, the structured prompt words of the present application are generated by the large language model based on the prompt word output instruction, extracting the description text corresponding to the attribute word information from the initial text, obtaining the target semantic description text, and converting the target semantic description text into prompt words. Specifically, the large language model extracts the description text that meets the attribute word information in the initial text to obtain the target semantic description text, and finally summarizes the target semantic description texts of each level and converts them into prompt words. Among them, the characteristics of the structured prompt words are the use of specific attribute words, the use of identifiers of Markdown markup syntax, the control of content hierarchy, identification of structure and variables, etc., and the use of the CoT thinking chain method. Among them, the attribute words realize the semantic prompt and induction of the prompt word content, and alleviate the interference of inappropriate content. The identifier realizes the role of gathering the same semantics and reduces the difficulty of model understanding.

[0083] Step S300: sending the structured prompt word to the large language model, and receiving text summary information output by the large language model, wherein the text summary information is generated by the large language model performing summary text conversion on the structured prompt word based on a summary output instruction.

[0084] In a specific implementation, the platform sends the structured prompt words to the large language model, and receives the text summary information output by the large language model. Specifically, the present application uses the deployed large language model to generate text summary information for the initial text, wherein the large language model can more accurately summarize the full text, thereby improving the accuracy of text summary generation, that is, through the prompt word project, the large language model is used to perform customized summary generation for multilingual texts, without the need for additional training models, thereby reducing costs and being more efficient and accurate.

[0085] In a specific implementation, the platform sends the structured prompt word to the large language model, and receives the text summary information output by the large language model, including:

[0086] Receive demand information input by a user and convert the demand information into demand parameters; send the structured prompt word and the restriction parameter to the large language model, and receive text summary information output by the large language model, wherein the text summary information is generated by the large language model by performing summary text conversion on the structured prompt word based on the summary output instruction and the demand parameters.

[0087] It should be noted that the demand information refers to the user's demand for the generated summary. After receiving the demand information input by the user, the platform converts the demand information into demand parameters, where the demand parameters include but are not limited to setting the number of summaries generated and the number of words in each summary.

[0088] In the specific implementation, refer to Figure 2 First, the user selects a summary generation template through the interactive page of the platform, and inputs text, uploads files or selects knowledge base files as the initial text for summary generation, and sets the number of summary generation and the number of words for each summary. The summary content generated by the large model will be displayed on the document editing page of the front end through streaming input for subsequent editing by users. This module uses the system instructions of prompt engineering (Prompt Engineering) to enable the large language model to convert the generation requirements such as summary generation parameters, output language and output format into structured prompt words without targeted fine-tuning, and guide the input, processing and output of the summary generation results of the large model. Prompt engineering is designed to help large models achieve high accuracy and relevance in summary generation tasks, and the quality of prompt engineering will have a great impact on the generation effect. That is, this application uses a large language model to customize summary generation for multilingual texts through prompt engineering, without the need for additional training models, which is more efficient and accurate while reducing costs.

[0089] In the specific implementation, users can customize the number of key points and the number of words in each key point of the summary according to the text situation and actual needs to achieve personalized needs. In addition, this application supports the generation of summaries of multilingual texts, and can generate accurate general summaries under the restrictions of the number of generated items and the number of words, without the need for manual translation by users, which is more convenient.

[0090] In a specific implementation, after the platform sends the structured prompt word to the large language model and receives the text summary information output by the large language model, the method includes:

[0091] Based on a preset context detection template, the text summary information is detected to obtain a detection result; whether the detection result meets a preset detection standard is determined; if the detection result does not meet the detection standard, the text summary information is adjusted based on the context detection template to obtain adjusted text summary information.

[0092] It should be noted that the context detection template is a preset template for detecting the readability (reading fluency) and word and sentence errors of the text summary context information, which is usually constructed manually through experience. If the detection result of the text summary information does not meet the detection standard, it means that the text summary information has sentence problems, including but not limited to unsmooth sentence reading, incorrect usage of conjunctions, etc.

[0093] In a specific implementation, the platform detects the text summary information based on a preset context detection template, obtains a detection result, and determines whether the detection result meets the preset detection standard. If the detection result does not meet the detection standard, it means that there is a sentence problem in the text summary information. Then, the platform adjusts the text summary information based on the context detection template to obtain the adjusted text summary information. That is, the platform will fine-tune it according to the prompt word performance to achieve the desired effect. The underlying pre-trained large language model has powerful natural language processing capabilities, so the use of high-performance structured prompt words allows the system to generate high-quality summaries without the need for a large amount of corpus training. The model can handle multiple languages, reducing the complexity and cost of multilingual text processing.

[0094] Compared with the related art, which usually extracts key sentences or phrases from the original text to generate a summary. However, the extractive summary generation method does not comprehensively consider the semantics and structure of the text, and the extracted sentences often cannot accurately represent the full text content, resulting in low accuracy of the generated text summary. In comparison, the present application uses a summary generation platform to perform a prompt word conversion operation on the initial text to be converted into a text summary, obtains a structured prompt word reflecting the semantics and structure of the initial text, and then sends the structured prompt word to the large language model, and receives the text summary information output by the large language model. It can be understood that on the one hand, the summary generation platform uses the deployed large language model to generate text summary information for the initial text, wherein the large language model can more accurately summarize the full text, and on the other hand, because the structured prompt word reflects the semantics and structure of the initial text, after the summary generation platform sends the structured prompt word to the large language model, the large language model can comprehensively consider the full text structure and semantics of the initial text according to the structured prompt word, so that the generated text summary information is more complete and fits the original text, thereby improving the accuracy of text summary generation.

[0095] Based on the above first embodiment, the present application also proposes another embodiment, referring to Figure 3 , the text summary generation method comprises:

[0096] In a specific implementation, after the platform sends the structured prompt word to the large language model and receives the text summary information output by the large language model, the method includes:

[0097] Step A100, displaying the text summary information on the user terminal;

[0098] In a specific implementation, after receiving the text summary information output by the large language model, the platform displays the text summary information to the user on the user side, so that the user can edit and modify it based on the text summary information, that is, the platform supports users to perform various rich text editing operations.

[0099] Step A200, responding to a text editing instruction issued by the user at the user terminal, wherein the text editing instruction includes a start field, an end field, and editing operation information;

[0100] In a specific implementation, after the platform displays the text summary information to the user on the user side, the user can issue text editing instructions on the user side to realize interaction between the user and the platform, wherein the text editing instructions include a start field, an end field, and editing operation information, wherein the start field indicates the starting field of the required editing adjustment (addition, deletion, modification, etc.), that is, the first field after the start cursor, and the end field indicates the end field of the required editing adjustment, that is, the first field before the termination cursor, and the editing operation information includes but is not limited to addition, deletion, modification, and other editing operations.

[0101] Step A300, integrating the fields between the start field and the end field to obtain a target edit field;

[0102] In a specific implementation, the continuous fields between the start field and the end field are integrated into a field set, that is, the platform integrates the fields between the start field and the end field to obtain the target edit field.

[0103] Step A400: performing an editing operation corresponding to the editing operation information on the target editing field to obtain edited text summary information.

[0104] In the specific implementation, the summary content is displayed on the document editing page, and the editing module supports users to perform various rich text editing operations. The editing module provides complex editing functions including Word basic functions, Markdown format, code blocks, quote blocks, etc. to meet the diverse needs of users. After the document is edited, it will be automatically saved to ensure the integrity and consistency of the document. The document is parsed by the editor into tree-structured JSON data, with a structure like Figure 4 As shown. Each modification saves the updated document JSON as a whole. When reading a saved document, the complete JSON is parsed by the editor. The standard input structure of the summary generated by the model is a text with a segmented format. Users can freely add document blocks with other formats to achieve personalized modification of the summary text and ensure that the final output summary document meets expectations.

[0105] In a specific implementation, the platform performs an editing operation corresponding to the editing operation information on the target editing field to obtain the edited text summary information, including:

[0106] The target edit field and the editing operation information are converted into prompt words to obtain a target prompt word; the target prompt word is sent to the large language model, and the edited text summary information output by the large language model is received, wherein the edited text summary information is generated by the large language model performing an editing operation corresponding to the editing operation information on the target edit field based on a summary adjustment instruction.

[0107] In the specific implementation, in addition to various editing functions, the platform also supports various AI creations for generating summary document content. Figure 5 , users can select text and choose the corresponding creative ability for processing. Each creative ability has a corresponding prompt word project that converts the user's selected text and instructions into a standard call format, and then requires a standardized output format, and then calls the large model to complete the corresponding creative task, and then streams the generated content to the front end for users to choose. Specific creative functions include continuation, polishing, change expectations, translation, grammar spelling check, abbreviation and expansion. Through these functions, users can make more detailed adjustments and optimizations to the generated summary documents as needed to meet more accurate information processing needs.

[0108] In a specific implementation, the editing operation information includes but is not limited to continuation, polishing, change expectations, translation, grammar and spelling checking, abbreviations and expansions, etc. The user only needs to click the corresponding button including the editing operation information on the platform interaction page. The platform will convert the target editing field selected by the user and the editing operation information into prompt words, obtain the target prompt words, and then use the large language model to perform the editing operation corresponding to the editing operation information on the target editing field, and output the edited text summary information, that is, the text summary information created by AI.

[0109] In the specific implementation, the secondary AI creation based on the big prediction model of this application allows users to freely rewrite the generated content, and various creation tasks can be completed with one click, truly providing a convenient complete text summary writing process, that is, this application provides rich AI creation functions while supporting basic rich text editing capabilities to meet the needs of rewriting generated content.

[0110] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the method for generating a text summary of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0111] This application also provides a text summary generation device, please refer to Figure 6, the text summary generating device comprises:

[0112] An acquisition module 10 is used to acquire an initial text;

[0113] The conversion module 20 is used to perform a prompt word conversion operation on the initial text according to a preset structured template to obtain a structured prompt word, wherein the structured prompt word reflects the semantics and structure of the initial text;

[0114] The generating module 30 is used to send the structured prompt word to the large language model and receive the text summary information output by the large language model, wherein the text summary information is generated by the large language model performing summary text conversion on the structured prompt word based on the summary output instruction.

[0115] Optionally, the conversion module 20 includes:

[0116] An extraction module, used to extract attribute word information of each level from a preset structured template, wherein the attribute word information is a keyword identifying different categories of parts in the structured prompt word;

[0117] A prompt word conversion module is used to send the initial text and the attribute word information to the large language model, and receive the structured prompt words output by the large language model, wherein the structured prompt words are generated by the large language model based on the prompt word output instruction, extracting the description text corresponding to the attribute word information from the initial text, obtaining the target semantic description text, and converting the target semantic description text into the prompt words.

[0118] Optionally, the generating module 30 includes:

[0119] A parameter conversion module, used to receive the demand information input by the user and convert the demand information into demand parameters;

[0120] A summary text conversion module is used to send the structured prompt word and the restriction parameter to the large language model, and receive text summary information output by the large language model, wherein the text summary information is generated by the large language model performing summary text conversion on the structured prompt word based on the summary output instruction and the requirement parameter.

[0121] Optionally, the text summary generating device further includes:

[0122] A detection module, used to detect the text summary information based on a preset context detection template to obtain a detection result;

[0123] The adjustment module is used to determine whether the detection result meets the preset detection standard. If the detection result does not meet the detection standard, the text summary information is adjusted based on the context detection template to obtain the adjusted text summary information.

[0124] Optionally, the text summary generating device further includes:

[0125] A display module, used to display the text summary information on the user terminal;

[0126] A response module, used to respond to a text editing instruction issued by a user at the user terminal, wherein the text editing instruction includes a start field, an end field and editing operation information;

[0127] An integration module, used for integrating the fields between the start field and the end field to obtain a target edit field;

[0128] The editing module is used to perform an editing operation corresponding to the editing operation information on the target editing field to obtain edited text summary information.

[0129] Optionally, the editing module includes:

[0130] A target prompt word determination module, used for converting the target edit field and the edit operation information into prompt words to obtain the target prompt word;

[0131] The intelligent creation module is used to send the target prompt word to the large language model and receive the edited text summary information output by the large language model, wherein the edited text summary information is generated by the large language model performing an editing operation corresponding to the editing operation information on the target editing field based on a summary adjustment instruction.

[0132] The text summary generation device provided by the present application adopts the text summary generation method in the above embodiment, which can solve the technical problem of text summary generation. Compared with the prior art, the beneficial effects of the text summary generation device provided by the present application are the same as the beneficial effects of the text summary generation method provided by the above embodiment, and other technical features in the text summary generation device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0133] The present application provides a text summary generation device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the text summary generation method in the above-mentioned embodiment 1.

[0134] Reference below Figure 7 , which shows a schematic diagram of the structure of a text summary generation device suitable for implementing the embodiment of the present application. The text summary generation device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 7 The text summary generation 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 application.

[0135] like Figure 7 As shown, the text summary generating device may include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the text summary generating device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the text summary generation device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows a text summary generation device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have alternatively.

[0136] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a 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, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0137] The text summary generation device provided by the present application adopts the text summary generation method in the above embodiment, which can solve the technical problem of text summary generation. Compared with the prior art, the beneficial effects of the text summary generation device provided by the present application are the same as the beneficial effects of the text summary generation method provided by the above embodiment, and the other technical features in the text summary generation device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0138] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0139] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0140] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the text summary generation method in the above-mentioned embodiment.

[0141] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, 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 this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0142] The computer-readable storage medium may be included in the text summary generation device; or may exist independently without being assembled into the text summary generation device.

[0143] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by a text summary generating device, the text summary generating device generates a text summary.

[0144] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including 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 the case of 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).

[0145] 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 application. 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 alternative implementations, 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.

[0146] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0147] The readable storage medium provided in this application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned text summary generation method, and can solve the technical problem of text summary generation. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the text summary generation method provided in the above-mentioned embodiment, and will not be repeated here.

[0148] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned text summary generation method when executed by a processor.

[0149] The computer program product provided by this application can solve the technical problem of text summary generation. Compared with the prior art, the beneficial effects of the computer program product provided by this application are the same as the beneficial effects of the text summary generation method provided by the above embodiment, which will not be repeated here.

[0150] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A text summary generation method, characterized in that: Applied to a summary generation platform, a large language model is deployed in the summary generation platform, and the method for generating a text summary includes: Get the initial text; According to a preset structured template, the initial text is converted into a prompt word to obtain a structured prompt word, wherein the structured prompt word reflects the semantics and structure of the initial text; The structured prompt word is sent to the large language model, and text summary information output by the large language model is received, wherein the text summary information is generated by the large language model performing summary text conversion on the structured prompt word based on a summary output instruction.

2. The text summary generation method according to claim 1, characterized in that: The step of converting the initial text into prompt words according to a preset structured template to obtain structured prompt words includes: Extracting attribute word information of each level from a preset structured template, wherein the attribute word information is a keyword identifying different categories of parts in the structured prompt word; The initial text and the attribute word information are sent to the large language model, and the structured prompt words output by the large language model are received, wherein the structured prompt words are generated by the large language model extracting the description text corresponding to the attribute word information from the initial text based on the prompt word output instruction, obtaining the target semantic description text, and converting the target semantic description text into the prompt words.

3. The text summary generation method according to claim 1, characterized in that: The step of sending the structured prompt word to the large language model and receiving the text summary information output by the large language model includes: Receive demand information input by a user and convert the demand information into demand parameters; The structured prompt word and the restriction parameter are sent to the large language model, and text summary information output by the large language model is received, wherein the text summary information is generated by the large language model performing summary text conversion on the structured prompt word based on a summary output instruction and the requirement parameter.

4. The text summary generation method according to claim 1, characterized in that: After the step of sending the structured prompt word to the large language model and receiving the text summary information output by the large language model, the method includes: Based on a preset context detection template, the text summary information is detected to obtain a detection result; It is determined whether the detection result meets a preset detection standard. If the detection result does not meet the detection standard, the text summary information is adjusted based on the context detection template to obtain adjusted text summary information.

5. The text summary generation method according to claim 1, characterized in that: After the step of sending the structured prompt word to the large language model and receiving the text summary information output by the large language model, the method includes: Displaying the text summary information on the user terminal; Responding to a text editing instruction issued by a user at the user terminal, wherein the text editing instruction includes a start field, an end field, and editing operation information; Integrate the fields between the start field and the end field to obtain a target edit field; An editing operation corresponding to the editing operation information is performed on the target editing field to obtain edited text summary information.

6. The text summary generation method according to claim 5, characterized in that: The step of performing an editing operation corresponding to the editing operation information on the target editing field to obtain edited text summary information comprises: Convert the target edit field and the edit operation information into a prompt word to obtain a target prompt word; The target prompt word is sent to the large language model, and edited text summary information output by the large language model is received, wherein the edited text summary information is generated by the large language model performing an editing operation corresponding to the editing operation information on the target edit field based on a summary adjustment instruction.

7. A text summary generation device, characterized in that: The device comprises: An acquisition module is used to acquire the initial text; A conversion module, used to perform a prompt word conversion operation on the initial text according to a preset structured template to obtain a structured prompt word, wherein the structured prompt word reflects the semantics and structure of the initial text; A generation module is used to send the structured prompt word to the large language model and receive text summary information output by the large language model, wherein the text summary information is generated by the large language model performing summary text conversion on the structured prompt word based on a summary output instruction.

8. A text summary generation device, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the text summary generation method according to any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the text summary generation method according to any one of claims 1 to 6 are implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the text summary generation method according to any one of claims 1 to 6 are implemented.

Images