Document generation method, system and device based on intelligent agent, electronic equipment, medium and program product

Abstract

The invention provides an agent-based document generation method, system and device, electronic equipment, a medium and a program product, and relates to the field of artificial intelligence, in particular to the fields of large language models, agents, smart documents and the like. According to the implementation scheme, in response to a received document generation request, multiple agents used for processing the document generation request to generate a target document are determined, and the multiple agents comprise a first agent used for generating a document framework and a second agent used for generating document content; calling the first agent to generate architecture content associated with chapters of a target document; and calling the second agent to generate the target document based on the architecture content.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, particularly to large language models, intelligent agents, and smart documents. Specifically, it relates to an agent-based document generation method, an agent-based document generation system, an apparatus, an electronic device, a computer-readable storage medium, and a computer program product. Background Art

[0002] Artificial intelligence (AI) is the study of how computers can simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, and planning). It encompasses both hardware and software technologies. AI hardware technologies generally include sensors, specialized AI chips, cloud computing, distributed storage, and big data processing. AI software technologies primarily encompass computer vision, speech recognition, natural language processing, machine learning / deep learning, big data processing, and knowledge graphs.

[0003] Intelligent agents are the manifestation of artificial intelligence in specific task scenarios. AI provides the technologies and methods for agents to operate effectively. In recent years, with the development of large language model technology, intelligent agents based on large language models have gradually evolved from a simple tool role to a role closer to human partners, capable of deeply understanding human intentions and needs and proactively providing assistance and support. Currently, intelligent agents have become valuable assistants to humans in work and life.

[0004] The approaches described in this section are not necessarily approaches that have been previously conceived or employed. Unless otherwise indicated, it should not be assumed that any approach described in this section is prior art simply by virtue of its inclusion in this section. Similarly, unless otherwise indicated, the issues raised in this section should not be considered as having been recognized in any prior art. Summary of the Invention

[0005] The present disclosure provides an agent-based document generation method, an agent-based document generation system, an apparatus, an electronic device, a computer-readable storage medium, and a computer program product.

[0006] According to one aspect of the present disclosure, an agent-based document generation method is provided, comprising: in response to receiving a document generation request, determining multiple agents for processing the document generation request to generate a target document, wherein the multiple agents include a first agent for generating a document architecture, and a second agent for generating document content; calling the first agent to generate architecture content associated with a chapter of the target document; and calling the second agent to generate the target document based on the architecture content.

[0007] According to another aspect of the present disclosure, an agent-based document generation system is provided for receiving a document generation request to generate a target document, including: a first agent for generating a document architecture to generate architecture content associated with a chapter of a target document when called; and a second agent for generating document content to generate a target document based on the architecture content when called.

[0008] According to another aspect of the present disclosure, an agent-based document generation device is provided, including: an agent determination module, configured to determine, in response to receiving a document generation request, multiple agents for processing the document generation request to generate a target document, wherein the multiple agents include a first agent for generating a document architecture, and a second agent for generating document content; a first calling module, configured to call the first agent to generate architecture content associated with a chapter of the target document; and a second calling module, configured to call the second agent to generate the target document based on the architecture content.

[0009] According to another aspect of the present disclosure, an electronic device is provided, comprising at least one processor; and a memory communicatively connected to the at least one processor; the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the agent-based document generation method of the present disclosure as described above.

[0010] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, where the computer instructions are used to cause a computer to execute the agent-based document generation method described above in the present disclosure.

[0011] According to another aspect of the present disclosure, a computer program product is provided, including a computer program, which implements the agent-based document generation method of the present disclosure as described above when executed by a processor.

[0012] According to one or more embodiments of the present disclosure, document system construction and document generation can be performed efficiently.

[0013] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The accompanying drawings illustrate exemplary embodiments and constitute a part of the specification. Together with the description of the specification, they serve to explain exemplary implementation of the embodiments. The illustrated embodiments are for illustrative purposes only and do not limit the scope of the claims. Throughout the drawings, the same reference numerals designate similar, but not necessarily identical, elements.

[0015] Figure 1 A schematic diagram illustrating an exemplary system in which the various methods described herein may be implemented according to an embodiment of the present disclosure;

[0016] Figure 2 A flowchart of an agent-based document generation method according to an embodiment of the present disclosure is shown;

[0017] Figure 3 A schematic diagram of calling an agent to generate a document list according to an embodiment of the present disclosure is shown;

[0018] Figure 4 A schematic diagram showing the architecture content of calling an intelligent agent to generate a target document according to an embodiment of the present disclosure;

[0019] Figure 5 A schematic diagram illustrating a document reference mark according to an embodiment of the present disclosure is shown;

[0020] Figure 6 A schematic diagram showing illustration marks according to an embodiment of the present disclosure;

[0021] Figure 7 A schematic diagram of calling a drawing tool to generate an illustration according to an embodiment of the present disclosure is shown;

[0022] Figure 8 A schematic diagram of calling an intelligent agent to evaluate a target document according to an embodiment of the present disclosure is shown;

[0023] Figure 9 A structural block diagram of an agent-based document generation system according to an embodiment of the present disclosure is shown;

[0024] Figure 10 shows a structural block diagram of an agent-based document generation system according to another embodiment of the present disclosure;

[0025] Figure 11 A structural block diagram of an agent-based document generation device according to an embodiment of the present disclosure is shown;

[0026] Figure 12 shows a structural block diagram of an agent-based document generation device according to another embodiment of the present disclosure;

[0027] Figure 13 A structural block diagram of an exemplary electronic device that can be used to implement the embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0028] The following description of exemplary embodiments of the present disclosure is made in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be appreciated by those skilled in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0029] In this disclosure, unless otherwise specified, the use of terms such as "first" and "second" to describe various elements is not intended to limit the positional relationship, temporal relationship, or importance relationship of these elements. Such terms are only used to distinguish one element from another. In some examples, the first element and the second element may refer to the same instance of the element, while in some cases, based on the context of the description, they may also refer to different instances.

[0030] The terms used in the descriptions of the various examples described in this disclosure are for the purpose of describing specific examples only and are not intended to be limiting. Unless the context clearly indicates otherwise, if the number of elements is not specifically limited, the element may be one or more. In addition, the term "and / or" used in this disclosure encompasses any one and all possible combinations of the listed items.

[0031] In related technologies, document construction often uses a model directly generated by a large language model. This model can only generate a single document, making it difficult to create a document system that can be referenced and verified. Furthermore, due to the length limit of context tokens, the document content may be too general and lack a lot of details during the generation of a single document.

[0032] To this end, embodiments of the present disclosure provide a more efficient document system generation technology.

[0033] The embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0034] Figure 1 FIG2 is a schematic diagram of an exemplary system 100 in which the various methods and apparatuses described herein may be implemented according to an embodiment of the present disclosure. Figure 1 , the system 100 includes one or more client devices 101, 102, 103, 104, 105, and 106, a server 120, and one or more communication networks 110 coupling the one or more client devices to the server 120. The client devices 101, 102, 103, 104, 105, and 106 can be configured to execute one or more applications.

[0035] In an embodiment of the present disclosure, the server 120 may run to enable execution of one or more services or software applications described in the embodiment of the present disclosure.

[0036] In some embodiments, server 120 may also provide other services or software applications, which may include non-virtualized environments and virtualized environments. In some embodiments, these services may be provided as web-based services or cloud services, such as provided to users of client devices 101, 102, 103, 104, 105, and / or 106 under a software as a service (SaaS) model.

[0037] exist Figure 1 In the configuration shown, the server 120 may include one or more components that implement the functions performed by the server 120. These components may include software components, hardware components, or a combination thereof that can be executed by one or more processors. Users operating client devices 101, 102, 103, 104, 105, and / or 106 may, in turn, utilize one or more client applications to interact with the server 120 to utilize the services provided by these components. It should be understood that a variety of different system configurations are possible, which may differ from the system 100. Therefore, Figure 1 is one example of a system for implementing the various methods described herein and is not intended to be limiting.

[0038] The user can use the client devices 101, 102, 103, 104, 105 and / or 106 to input multimodal information such as text, pictures, documents for document generation. The client device can provide an interface that enables the user of the client device to interact with the client device. The client device can also output information to the user via the interface. Although Figure 1 Only six client devices are depicted, but one skilled in the art will appreciate that the present disclosure can support any number of client devices.

[0039] Client devices 101, 102, 103, 104, 105, and / or 106 may include various types of computer devices, such as portable handheld devices, general-purpose computers (such as personal computers and laptops), workstation computers, wearable devices, smart screen devices, self-service terminal devices, service robots, gaming systems, thin clients, various messaging devices, sensors or other sensing devices, etc. These computer devices may run various types and versions of software applications and operating systems, such as Microsoft Windows, Apple iOS, UNIX-like operating systems, Linux, or Linux-like operating systems (such as Google Chrome OS); or include various mobile operating systems, such as Microsoft Windows Mobile OS, iOS, Windows Phone, and Android. Portable handheld devices may include cellular phones, smartphones, tablet computers, personal digital assistants (PDAs), etc. Wearable devices may include head-mounted displays (such as smart glasses) and other devices. Gaming systems may include various handheld gaming devices, internet-enabled gaming devices, etc. Client devices are capable of executing a variety of different applications, such as various internet-related applications, communication applications (such as email applications), and short message service (SMS) applications, and may use various communication protocols.

[0040] The network 110 may be any type of network known to those skilled in the art that can support data communications using any of a variety of available protocols, including but not limited to TCP / IP, SNA, IPX, etc. By way of example only, the one or more networks 110 may be a local area network (LAN), an Ethernet-based network, a token ring, a wide area network (WAN), the Internet, a virtual network, a virtual private network (VPN), an intranet, an extranet, a blockchain network, a public switched telephone network (PSTN), an infrared network, a wireless network (e.g., Bluetooth, WIFI), and / or any combination of these and / or other networks.

[0041] Server 120 may include one or more general-purpose computers, specialized server computers (e.g., PC (personal computer) servers, UNIX servers, mid-range servers), blade servers, mainframe computers, server clusters, or any other suitable arrangement and / or combination. Server 120 may include one or more virtual machines running virtual operating systems, or other computing architectures involving virtualization (e.g., one or more flexible pools of logical storage devices that may be virtualized to maintain a server's virtual storage device). In various embodiments, server 120 may run one or more services or software applications that provide the functionality described below.

[0042] The computing units in the server 120 may run one or more operating systems including any of the operating systems described above as well as any commercially available server operating systems. The server 120 may also run any of a variety of additional server applications and / or middle-tier applications, including HTTP servers, FTP servers, CGI servers, JAVA servers, database servers, and the like.

[0043] In some implementations, server 120 may include one or more applications to analyze and consolidate data feeds and / or event updates received from users of client devices 101, 102, 103, 104, 105, and / or 106. Server 120 may also include one or more applications to display the data feeds and / or real-time events via one or more display devices of client devices 101, 102, 103, 104, 105, and / or 106.

[0044] In some embodiments, server 120 may be a distributed system server or a server integrated with blockchain. Server 120 may also be a cloud server, or an intelligent cloud computing server or intelligent cloud host equipped with artificial intelligence technology. A cloud server is a host product within the cloud computing service system that addresses the management difficulties and poor scalability of traditional physical hosts and virtual private servers (VPS) services.

[0045] The system 100 may also include one or more databases 130. In some embodiments, these databases may be used to store data and other information. For example, one or more of the databases 130 may be used to store information such as audio files and video files. The databases 130 may reside in a variety of locations. For example, the database used by the server 120 may be local to the server 120, or may be remote from the server 120 and communicate with the server 120 via a network-based or dedicated connection. The databases 130 may be of different types. In some embodiments, the databases used by the server 120 may be, for example, relational databases. One or more of these databases may store, update, and retrieve data to and from the databases in response to commands.

[0046] In some embodiments, one or more of the databases 130 may also be used by applications to store application data. The databases used by the applications may be different types of databases, such as a key-value store, an object store, or a conventional store backed by a file system.

[0047] Figure 1The system 100 may be configured and operated in various ways to enable application of the various methods and apparatuses described in accordance with the present disclosure.

[0048] Various aspects of the agent-based document generation method according to an embodiment of the present disclosure will be described in detail below.

[0049] Figure 2 A flowchart of an agent-based document generation method 200 according to an embodiment of the present disclosure is shown.

[0050] like Figure 2 As shown, method 200 includes step S201, step S202 and step S203.

[0051] In step S201, in response to receiving a document generation request, multiple agents are determined for processing the document generation request to generate a target document, including a first agent for generating a document structure and a second agent for generating document content.

[0052] In the example, in some application scenarios, with the rapid development of artificial intelligence technology, the text generation capabilities of large language models are becoming stronger and stronger. The intelligent agents in the embodiments of the present disclosure can call on these capabilities of large language models to perform their respective processing tasks.

[0053] In this example, a document generation request indicates a user's need for document generation, allowing for the identification of multiple relevant agents to handle the request based on this user's needs. Based on analysis of the document generation request, the document generation task can be broken down into several subtasks, and multiple agents can be identified to generate the target document based on these subtasks. These agents each have their own functional characteristics and specializations. Through multi-agent collaboration, the limitations of document generation based directly on large language models due to word length restrictions can be overcome, contributing to the generation of high-quality target documents.

[0054] In an example, the first agent may be an agent for generating a document architecture determined according to a document generation request. The goal of the first agent is to build a framework for the target document so as to facilitate subsequent expansion or modification of the target document based on the framework.

[0055] In this example, the second agent may be an agent determined based on the document generation request to generate document content. The second agent's goal is to generate document content for the target document based on the document architecture of the target document generated by the first agent, so as to obtain a target document with rich details and clear structure.

[0056] In step S202, the first agent is called to generate the schema content associated with the chapter of the target document.

[0057] In the example, after determining the first agent based on the document generation request, the first agent is called to generate the document architecture content in the document chapter. Architecture content refers to the way of designing and organizing document content, format and structure, with the aim of ensuring that information can be presented in a logically clear, easy-to-understand and usable form. Document architecture can involve all aspects of the document, including the layout, style, hierarchy and navigation design of elements such as text, images and tables. Document architecture can be used to ensure the consistency, accessibility and usability of documents.

[0058] In step S203, the second agent is called to generate a target document based on the architecture content.

[0059] In this example, after the first agent completes the generation of the document's structure, it calls upon the second agent to expand the document's content based on the structure, resulting in a target document with rich details and a clear structure. Document content refers to the information and data contained within a document, primarily including elements such as text, images, tables, and multimedia. Document content conveys information through textual explanations, image presentations, or data lists, clearly and accurately expressing a specific topic.

[0060] According to an embodiment of the present disclosure, an agent-based document generation method is proposed. The method parses the document generation request to determine multiple agents with different specific skills required for the target document generation task, so as to achieve adaptive splitting of the document generation task, thereby breaking through the limitations of traditional solutions in document generation directly based on large language models, solving the problem of missing details in the generated documents, and improving the quality of the generated target documents.

[0061] In some embodiments, the plurality of agents may include a third agent for planning a document generation task. The document generation request may include raw data for generating a target document and a target category for the target document. Figure 2 The agent-based document generation method 200 may further include: calling a third agent to generate a document list indicating at least one target document. The document list may be determined based on at least a portion of the original data associated with the target category.

[0062] Figure 3 A schematic diagram of calling an agent to generate a document list according to an embodiment of the present disclosure is shown. The agent may be the third agent used to plan the document generation task.

[0063] In an example, the document generation request may include a basis for generating the target document. Figure 3As shown, the document generation request may include raw data 301 for generating the target document and target category 302 for determining the target document generation field and direction. The third agent may be a dedicated agent for planning the document generation task. The third agent may be called first to plan the document generation task before calling the first agent. Based on the raw data 301 and the target category 302, the third agent may be called to generate a document list 303. The document list 303 may indicate which target documents need to be generated in the subsequent generation process in the form of a list. Figure 3 As shown, the document list 303 may include a document 1 to be generated 3031, a document 2 to be generated 3032, and a document 3 to be generated 3033. It is understandable that Figure 3 Three documents to be generated are shown only by way of example, but the scope of the present disclosure is not limited thereto.

[0064] Therefore, a third agent for planning the document generation task is determined based on the document generation request, and the third agent is called before calling the first agent. A document list of target documents can be generated to prompt the number of target documents that need to be generated in the subsequent document generation process and the generation direction of the target documents, so as to help generate a document system that can be related and referenced to each other.

[0065] In some embodiments, as combined Figure 2 The step S202 shown calls the first agent to generate the structural content associated with the chapter of the target document, which may include: transmitting the document list to the first agent so that the first agent generates the structural content of at least one target document in response to receiving the document list.

[0066] In this example, as previously described, the first agent can be a dedicated agent for document schema generation. A document list can be used to indicate which target documents need to be generated, and thus can include a list of names of the target documents to be generated. Schema content can be used as a basis for expanding or updating the target documents, and thus can include outlines for each target document to be generated in the document list. After transmitting the document list to the first agent, the first agent can generate schema content for all target documents based on the raw data in the document generation request and the document list generated by the third agent.

[0067] Figure 4 A schematic diagram of calling an agent to generate the architecture content of a target document according to an embodiment of the present disclosure is shown. The agent may be the first agent for generating the document architecture.

[0068] like Figure 4As shown, a document list 401 can be transmitted to a first agent 410 for generating a document architecture to obtain the architecture content 402 of all target documents. Document list 401 may, for example, include document one 4011, document two 4012, and document three 4013. The architecture content 402 generated after document list 401 is processed by first agent 410 may, for example, include the outline of document one 4021, the outline of document two 4022, and the outline of document three 4023. Figure 4 The specific contents of the three major document outlines 4023 are schematically shown, and may include Chapter 1 title 4023a, Chapter 1 overview information 4023b, Chapter 1 section 1 title 4023c, and Chapter 1 section 1 overview information 4023d. Chapter 1 overview information 4023b may be a brief description of the contents of Chapter 1, and Chapter 1 section 1 overview information 4023d may be a brief description of the contents of Chapter 1, Section 1. Similarly, Document Three Major Document Outlines 4023 may also include Chapter 2 title 4023f and other contents.

[0069] Therefore, by calling the first agent to process the document list to generate the respective structural contents of the target documents, the framework of the target document to be generated can be preliminarily determined, providing a basis for the generation of the target document.

[0070] In some embodiments, as combined Figure 2 The step S203 shown calls the second agent to generate a target document based on the architecture content, which may include: transmitting the architecture content of at least one target document to the second agent, so that the second agent generates the target document by expanding the architecture content in response to receiving the architecture content.

[0071] In this example, the second agent can be a dedicated agent for content expansion. After the first agent completes generating the schema content for all target documents, the schema content, including the outline of the document to be generated, can be transmitted to the second agent. The second agent then expands the target documents based on the schema content of the target documents, thereby generating the target documents.

[0072] Therefore, by providing the architecture content to a second agent dedicated to content expansion to generate a target document, the quality of the generated target document can be effectively improved.

[0073] In some embodiments, the schema content may include a document reference tag. The document reference tag may be used to indicate references between different target documents.

[0074] Figure 5 A schematic diagram of a document reference mark according to an embodiment of the present disclosure is shown.

[0075] like Figure 5As shown, the schematic diagram shows that the architecture content 501 includes three document outlines 5011. At the same time, the schematic diagram shows that the three document outlines 5011 include the seventh chapter title 5011a and the seventh chapter overview information 5011b, wherein the seventh chapter overview information 5011b may include a document reference mark 5011c. Figure 5 As shown in , the document reference mark 5011c is "[reference document one]", indicating that the text at the reference mark is associated with the content in document one.

[0076] In the example, the schema content may include a document reference tag for indicating a reference relationship between documents. The document reference tag carries an indicator for indicating a reference, so the document reference tag can be represented by a symbol wrapped around text, such as "[ ]", "[]", "<>", etc. Accordingly, the document reference tag can be represented as "[Reference Document Two]", "[Reference Document Three]", "<Reference Document Four>", etc. The representation of the document reference tag can be preset through the document reference tag parameters of the agent before performing the document generation task.

[0077] Therefore, by adding reference marks between documents when generating architectural content, the second intelligent agent in the subsequent document generation step can be prompted to generate reference text between documents, thereby corresponding the text at the document reference mark with the specific content in the associated chapter of the referenced target document, to help achieve mutual reference and verification between documents.

[0078] In some embodiments, the document reference mark can be expanded to display the associated section information of the referenced target document.

[0079] In the example, when the second agent expands the content of the target document based on the architecture content, the document reference mark can be expanded into a specific text content, which is used to display the associated chapter information of the referenced target document. Figure 5 The document reference mark 5011c shown is "[reference document one]" before expansion, and after being expanded by the second agent, it can be, for example, "Chapter 2, Section 3 of Reference Document One", etc.

[0080] Therefore, by expanding the document reference markup, cross-references between documents can be achieved. This is also beneficial for document modification or update processes. When a referenced document changes, the document reference markup can be used to synchronously reflect changes in all related documents, thereby improving the efficiency of document modification or update.

[0081] In some embodiments, the schema content may include figure markup that may be used to indicate where the figure is to be embedded in the target document.

[0082] Figure 6A schematic diagram illustrating inset markings according to an embodiment of the present disclosure is shown.

[0083] like Figure 6 As shown, the schematic diagram shows that the architecture content 601 includes three document outlines 6011, and the schematic diagram shows that the three document outlines 6011 include the fourth chapter title 6011a and the fourth chapter overview information 6011b, wherein the fourth chapter overview information 6011b includes not only text descriptions but also illustration marks 6011c. Figure 6 As shown in , the illustration mark 6011c can be "[need a flow chart]", indicating that the corresponding flow chart needs to be inserted at the position of the mark in the subsequent document generation process. Similarly, the symbols used for illustration marks can also be "[ ]", "[]", "<>", etc. Accordingly, the illustration mark can be expressed as "[need a flow chart]", "[need a structure diagram]", "<need a schematic diagram>", "[need a table]", etc. The representation form of the illustration mark can be preset through the illustration reference mark parameter of the intelligent agent before performing the document generation task.

[0084] Therefore, by adding illustration tags when generating architectural content, corresponding illustrations can be generated in subsequent document generation steps to enhance the expressive effect of the text content in the illustrations, making the target document content more intuitive and easy to understand.

[0085] In some embodiments, the illustration markup may be expanded into a graphic description language for representing illustrations, so that the second agent can call a drawing tool to generate an illustration based on the graphic description language.

[0086] Figure 7 A schematic diagram of calling a drawing tool to generate an illustration according to an embodiment of the present disclosure is shown.

[0087] like Figure 7 As shown, the graphic description language 701 of the illustration can be processed by the drawing tool 710 to generate the corresponding illustration 702. Figure 7 The process and content of generating an illustration 702 by using a graphic description language 701 for illustrations are schematically shown in FIG. The graphic description language 701 for illustrations may be a Mermaid language, which is composed of “ <flowchart> "and"< / flowchart> "The package consists of words or symbols that conform to the specific grammar of the Mermaid language. For example Figure 7The illustrated graphical description language 701 exemplarily describes a flowchart generation language based on the Mermaid language. A second agent can invoke a drawing tool 710 to process the graphical description language 701 to generate a corresponding illustration 702. Illustration 702 contains a flowchart corresponding to the graphical description language 701. For example, after starting the process, step 1 is executed first, followed by step 2. A determination is made as to whether condition 1 is satisfied. If so, step 3 is executed; otherwise, step 4 is executed.

[0088] In this example, during the process of the second agent expanding the architecture content, the illustration markup is expanded into a graphical description language for describing the illustration. For example, the illustration markup may include flowchart markup, structure diagram markup, schematic markup, table markup, etc. Accordingly, the flowchart markup may be expanded into a flowchart description language, the structure diagram markup may be expanded into a structure diagram description language, the schematic markup may be expanded into a schematic description language, and the table markup may be expanded into a table description language, etc.

[0089] Therefore, by expanding the illustration mark into a graphic description language for the illustration, it is convenient for the second intelligent agent to call the corresponding illustration generation tool according to the specific graphic description language to generate an illustration at the illustration mark position, which helps to improve the readability and professionalism of the target document in a graphical expression way.

[0090] In some embodiments, the plurality of agents may further include a fourth agent for quantifying the document generation effect. Figure 2 The agent-based document generation method 200 shown may further include: calling a fourth agent to evaluate the target document to obtain an evaluation result.

[0091] In this example, the fourth agent can be a dedicated agent used to quantify the effectiveness of document generation. After the second agent is called to complete the generation of the target document, the fourth agent can be called to evaluate the target document to obtain the evaluation results of the target document. The evaluation results can include information such as numerical quantification, grading, and descriptive evaluation. Numerical quantification can measure the characteristics and quality of the evaluation object, i.e., the target document, through specific numerical values. Grading can divide the target document into different levels according to certain standards. Descriptive evaluation can describe the characteristics, advantages, and disadvantages of the target document in text form.

[0092] Therefore, by calling the fourth agent to perform quantitative evaluation on the target document, it is possible to determine whether the target document meets the requirements, providing a basis for improving and updating the target document.

[0093] In some embodiments, the evaluation result may include an evaluation score. Accordingly, the step of invoking the fourth agent to evaluate the target document to obtain the evaluation result may include transmitting the target document to the fourth agent, so that the fourth agent, in response to receiving the target document, scores the target document based on a preset scoring standard to obtain the evaluation score.

[0094] In the example, the scoring criteria can be a reference standard for the fourth agent to evaluate the target document, for example, it can include scoring dimensions, scoring rules and weights. The scoring dimensions can indicate in which aspects to evaluate. The scoring rules can specify the specific scores corresponding to different performances under each scoring dimension. The weights can determine the relative importance of each scoring dimension in the overall evaluation. The scoring criteria can be provided by default by the knowledge base of the fourth agent, or provided by the user before generating the target document. The evaluation score can be the corresponding specific score obtained after evaluating the different performances of the target document under each scoring dimension based on the provisions of the scoring rules. The evaluation result can be obtained by calculating the evaluation score in a weighted sum manner.

[0095] Therefore, by scoring the target document based on the scoring criteria to obtain an evaluation score, we can clearly understand the specific performance of the target document in different dimensions and provide a basis for the optimization of the target document.

[0096] Figure 8 A schematic diagram of calling an agent to evaluate a target document according to an embodiment of the present disclosure is shown. The agent may be the fourth agent used to evaluate the target document.

[0097] like Figure 8 As shown, it schematically shows a situation in which the fourth agent 810 scores the target document 801 based on the scoring criteria 820 to obtain an evaluation score 802. The scoring criteria 820 may, for example, include two dimensions: scoring dimension one and scoring dimension two. When the target document 801 meets the judgment condition one, the scoring dimension one is scored 1 point. When the target document 801 meets the judgment condition two, the scoring dimension one is scored 2 points. When the target document 801 meets the judgment condition three, the scoring dimension one is scored 3 points. Similarly, when the target document 801 meets the judgment condition four, the scoring dimension two is scored 1 point. When the target document 801 meets the judgment condition five, the scoring dimension two is scored 2 points. When the target document 801 meets the judgment condition six, the scoring dimension two is scored 3 points. In Figure 8 In the example, based on the scoring criteria 820, the fourth agent 810 scores the target document 801 based on the scoring dimension 1 and the scoring dimension 2, respectively, to obtain an evaluation score 802 for the target document. The evaluation score 802 may include, for example: the scoring dimension 1 score: 3 points; the scoring dimension 2 score: 1 point.

[0098] In some embodiments, the evaluation results may also include prompt information for improving the target document. Figure 2 The agent-based document generation method 200 shown may also include: in response to the evaluation score not reaching a preset threshold, calling the fourth agent to determine the first target agent associated with improvement among the first to third agents based on prompt information; and calling the first target agent to improve the target document based on the prompt information.

[0099] In the example, the prompt information for improving the target document may be obtained by the fourth agent in the process of evaluating the target document, or may be generated by the fourth agent based on the evaluation score. The prompt information may include ideas, directions, requirements, etc. for improving the target document. The first target agent may be determined by the fourth agent based on the prompt information, which may include one or more of the first agent, the second agent, and the third agent. For example, when the prompt information indicates that the document architecture needs to be improved, the first target agent may include the first agent; when the prompt information indicates that the document content needs to be improved, the first target agent may include the second agent; when the prompt information indicates that the document generation task planning needs to be improved, the first target agent may include the third agent. After determining the first target agent, the first target agent may be called to improve the target document based on the prompt information.

[0100] In the example, the improvement methods may include: adding new content to the target document, deleting part of the content in the target document, replacing part of the content in the target document, etc. For example, in the case where the first target agent includes the first agent, the second agent, and the third agent, when the prompt information indicates that the document generation task needs to be re-planned, the third agent can be called to re-execute the planned document generation task according to the prompt information; when the prompt information indicates that the document architecture needs to be regenerated, the first agent can be called to regenerate the document architecture according to the prompt information; when the prompt information indicates that the document content needs to be regenerated, the second agent can be called to regenerate the document content according to the prompt information.

[0101] Therefore, through the prompt information in the evaluation results, the improvement direction of the target document can be determined, so as to optimize the content of the target document and improve the quality of the target document.

[0102] In some embodiments, as combined Figure 2 The agent-based document generation method 200 shown may also include: in response to receiving a request to update a target document, determining a second target agent associated with the update from the first to third agents; and calling the second target agent to update the target document based on the request.

[0103] In the example, unlike receiving a document generation request from a user, when receiving a request from a user to update a target document, an agent can be scheduled to perform an update operation on the target document. The document update request may include a document update instruction and basic data supporting the document update, etc. The document update instruction may indicate that an update operation needs to be performed on the target document. The basic data supporting the document update may include multimodal information such as new reference text, new reference pictures, new content requirements, etc. The second target agent can be determined by the update request, which may include one or more of the first agent, the second agent, and the third agent. For example, when the update request indicates that the document architecture needs to be updated, the second target agent may include the first agent; when the update request indicates that the document content needs to be updated, the second target agent may include the second agent; when the update request indicates that the document generation task plan needs to be updated, the second target agent may include the third agent. After determining the second target agent, the second target agent may be called to update the target document according to the update request.

[0104] In the example, the update method may include: adding new content to the target document, deleting part of the content in the target document, and replacing part of the content in the target document. For example, when the second target agent includes a first agent, a second agent, and a third agent, when the update request indicates that the document generation task needs to be re-planned, the third agent can be called to re-execute the planned document generation task according to the update request; when the update request indicates that the document architecture needs to be regenerated, the first agent can be called to regenerate the document architecture according to the update request; when the update request indicates that the document content needs to be regenerated, the second agent can be called to regenerate the document content according to the update request.

[0105] Therefore, by receiving update requests, it is possible to adaptively update existing target documents. For example, when an update request includes a new reference document, the portion of the target document associated with the new reference document can be updated. Furthermore, supporting update requests expands the use of this method, making it not only limited to document generation from scratch but also applicable to revising and improving existing documents.

[0106] Figure 9 A structural block diagram of an agent-based document generation system 900 according to an embodiment of the present disclosure is shown.

[0107] like Figure 9As shown, system 900 includes a first agent 901 and a second agent 902. System 900 can be used to receive a document generation request to generate a target document. First agent 901 can be used to generate a document schema to generate schema content associated with sections of the target document when called. Second agent 902 can be used to generate document content to generate the target document based on the schema content when called.

[0108] In this example, first agent 901 is a dedicated agent for document architecture generation. When called, first agent 901 can generate the target document's architecture content, such as the target document's chapter outlines, chapter names, and overview information, based on the document generation request. Second agent 902 can be a dedicated agent for document content generation. When called, second agent 902 can expand the target document's architecture content based on the target document's architecture content to generate the target document.

[0109] By configuring the first agent and the second agent to jointly complete document generation, the accuracy of the generated content of the target document can be ensured and the quality of the generated document can be improved.

[0110] Figure 10 Shown is a structural block diagram of an agent-based document generation system 1000 according to another embodiment of the present disclosure.

[0111] like Figure 10 As shown, the system 1000 may include a first agent 1001, a second agent 1002, a third agent 1003 and a fourth agent 1004. The operations of the first agent 1001 and the second agent 1002 may be the same as those in FIG. Figure 9 The operations of the first agent 901 and the second agent 902 are the same.

[0112] In some embodiments, a document generation request may include raw data for generating a target document and a target category for the target document. Accordingly, third agent 1003 may be configured to plan a document generation task so as to generate a document list indicating at least one target document when invoked. The document list may be determined based on at least a portion of the raw data associated with the target category.

[0113] In this example, the third agent 1003 may be a dedicated agent for planning document generation tasks. When called, the third agent 1003 may determine a list of target documents based on the target category of the documents to be generated and relevant information in the original data, and display the names of the target documents to be generated in the form of a list.

[0114] By configuring the third agent 1003 to plan the document generation task, the entire document generation task can be planned in an overall manner and the number of target documents to be generated can be clarified.

[0115] In some embodiments, the fourth agent 1004 can be used to quantify the document generation effect to evaluate the target document when called to obtain an evaluation result.

[0116] In this example, the fourth agent 1004 may be a dedicated agent for quantifying the quality of document generation. When called, the fourth agent 1004 may evaluate or score the target document according to a preset scoring standard to obtain an evaluation result that quantifies the quality of target document generation.

[0117] By configuring the fourth agent to evaluate the target document, the quality, advantages and disadvantages of the current content of the target document can be learned from the evaluation results, providing a basis for improving the target document.

[0118] In some embodiments, the evaluation result may include an evaluation score and prompt information for improving the target document. Accordingly, the fourth agent 1004 may also be configured to: in response to the evaluation score not reaching a preset threshold, determine, based on the prompt information, a target agent associated with improvement among the first agent 1001, the second agent 1002, and the third agent 1003.

[0119] In this example, in addition to document evaluation, fourth agent 1004 may also have a document improvement function. By comparing the evaluation score in the evaluation result with a preset threshold score, if the evaluation score of the target document does not meet the preset threshold score requirement, fourth agent 1004 may determine a target agent for improving the target document based on the prompt information. The target agent may include one or more of first agent 1001, second agent 1002, and third agent 1003.

[0120] Therefore, by configuring the fourth agent 1004 to determine the target agent for improving the target document based on the prompt information, the document generation system can automatically improve the target document to enhance the intelligence of the document generation system.

[0121] Figure 11 A structural block diagram of an agent-based document generation device 1100 according to an embodiment of the present disclosure is shown.

[0122] like Figure 11As shown, the device 1100 includes an agent determination module 1101, a first calling module 1102 and a second calling module 1103. The agent determination module 1101 can be configured to determine a plurality of agents for processing the document generation request to generate a target document in response to receiving a document generation request, wherein the plurality of agents include a first agent for generating a document architecture and a second agent for generating document content. The first calling module 1102 can be configured to call the first agent to generate architecture content associated with a chapter of the target document. The second calling module 1103 can be configured to call the second agent to generate a target document based on the architecture content.

[0123] The operations of the agent determination module 1101, the first calling module 1102 and the second calling module 1103 can respectively correspond to the following: Figure 2 The operations of steps S201, S202 and S203 are shown in FIG. Therefore, details of each aspect thereof will not be repeated here.

[0124] Figure 12 A structural block diagram of an agent-based document generation device 1200 according to another embodiment of the present disclosure is shown.

[0125] like Figure 12 As shown, the apparatus 1200 may include an agent determination module 1201, a first calling module 1202, and a second calling module 1203. The operations of the agent determination module 1201, the first calling module 1202, and the second calling module 1203 may be the same as those described above. Figure 11 The operations of the agent determination module 1101, the first calling module 1102 and the second calling module 1103 shown are the same.

[0126] In some embodiments, the plurality of agents may include a third agent for planning a document generation task. The document generation request may include raw data for generating a target document and a target category for the target document. Apparatus 1200 may further include a third invoking module 1204 configured to invoke the third agent to generate a document list indicating at least one target document, wherein the document list is determined based on at least a portion of the raw data associated with the target category.

[0127] In some embodiments, the first calling module 1202 may include a document list transmission module 1202a that may be configured to transmit the document list to the first agent, so that the first agent generates the respective schema content of at least one target document in response to receiving the document list.

[0128] In some embodiments, the second calling module 1203 may include a schema content transmission module 1203a. The schema content transmission module 1203a may be configured to transmit the schema content of at least one target document to the second agent, so that the second agent generates the target document by expanding the schema content in response to receiving the schema content.

[0129] In some embodiments, the plurality of agents may include a fourth agent for quantifying the document generation effect. The apparatus 1200 may further include a fourth calling module 1205 configured to call the fourth agent to evaluate the target document to obtain an evaluation result.

[0130] In some embodiments, the evaluation result may include an evaluation score. The fourth calling module 1205 may include a target document transmission module 1205a. The target document transmission module 1205a may be configured to transmit the target document to the fourth agent, so that the fourth agent, in response to receiving the target document, scores the target document based on a preset scoring standard to obtain an evaluation score.

[0131] In some embodiments, the evaluation result also includes prompt information for improving the target document. Apparatus 1200 may further include: a first determination module 1206 configured to, in response to the evaluation score not reaching a preset threshold, call a fourth agent to determine, based on the prompt information, a first target agent associated with improvement among the first to third agents; and a document improvement module 1207 configured to call the first target agent to improve the target document based on the prompt information.

[0132] In some embodiments, the device 1200 may also include: a second determination module 1208, configured to determine the second target agent associated with the update from the first agent to the third agent in response to receiving a request to update the target document; and a document update module 1209, configured to call the second target agent to update the target document based on the request.

[0133] In the technical solutions disclosed herein, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0134] According to an embodiment of the present disclosure, an electronic device is also provided, comprising at least one processor; and a memory communicatively connected to the at least one processor, wherein the memory stores instructions that can be executed 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 method described above.

[0135] According to an embodiment of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is further provided, wherein the computer instructions are used to enable a computer to execute the method described above.

[0136] According to an embodiment of the present disclosure, a computer program product is further provided, including a computer program, wherein the computer program implements the method described above when executed by a processor.

[0137] refer to Figure 13 , a block diagram of an electronic device 1300 that can serve as a server or client of the present disclosure will now be described, which is an example of a hardware device that can be applied to various aspects of the present disclosure. The electronic device is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0138] like Figure 13 As shown, electronic device 1300 includes a computing unit 1301, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1302 or a computer program loaded from a storage unit 1308 into a random access memory (RAM) 1303. Various programs and data required for the operation of electronic device 1300 can also be stored in RAM 1303. Computing unit 1301, ROM 1302, and RAM 1303 are connected to each other via a bus 1304. An input / output (I / O) interface 1305 is also connected to bus 1304.

[0139] Multiple components in the electronic device 1300 are connected to the I / O interface 1305, including: an input unit 1306, an output unit 1307, a storage unit 1308, and a communication unit 13013. The input unit 1306 can be any type of device that can input information to the electronic device 1300. The input unit 1306 can receive input digital or character information and generate key signal input related to user settings and / or function control of the electronic device, and can include but is not limited to a mouse, a keyboard, a touch screen, a trackpad, a trackball, a joystick, a microphone, and / or a remote control. The output unit 1307 can be any type of device that can present information, and can include but is not limited to a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. The storage unit 1308 can include but is not limited to a magnetic disk and an optical disk. The communication unit 1309 allows the electronic device 1300 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks, and may include but is not limited to a modem, a network card, an infrared communication device, a wireless communication transceiver and / or a chipset, such as a Bluetooth device, an 802.11 device, a WiFi device, a WiMax device, a cellular communication device and / or the like.

[0140] The computing unit 1301 can be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 1301 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units that run machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 1301 performs the various methods and processes described above. For example, in some embodiments, the method can be implemented as a computer software program that is tangibly contained in a machine-readable medium, such as a storage unit 1308. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 1300 via the ROM 1302 and / or the communication unit 1309. When the computer program is loaded into the RAM 1303 and executed by the computing unit 1301, one or more steps of the method described above can be performed. Alternatively, in other embodiments, the computing unit 1301 can be configured to perform the method in any other appropriate manner (e.g., by means of firmware).

[0141] Various embodiments of the systems and techniques described above can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0142] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0143] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0144] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0145] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), the Internet, and a blockchain network.

[0146] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises through computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.

[0147] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved. This is not limited herein.

[0148] Although the embodiments or examples of the present disclosure have been described with reference to the accompanying drawings, it should be understood that the above-mentioned methods, systems and devices are merely exemplary embodiments or examples, and the scope of the present invention is not limited by these embodiments or examples, but is only limited by the claims after authorization and their equivalents. Various elements in the embodiments or examples may be omitted or replaced by their equivalents. In addition, the steps may be performed in an order different from that described in this disclosure. Further, the various elements in the embodiments or examples may be combined in various ways. It is important that as technology evolves, many of the elements described herein may be replaced by equivalent elements that appear after this disclosure.

Claims

1. A document generation method based on an agent, comprising: In response to receiving a document generation request, determining a plurality of agents for processing the document generation request to generate a target document, wherein the plurality of agents include a first agent for generating a document structure and a second agent for generating document content; Invoking the first agent to generate schema content associated with a section of the target document; and The second agent is called to generate the target document based on the architecture content.

2. The method according to claim 1, wherein The plurality of agents include a third agent for planning a document generation task, the document generation request includes raw data for generating the target document and a target category of the target document, and the method further includes: The third agent is called to generate a document list indicating at least one target document, wherein the document list is determined based on at least a portion of the original data associated with the target category.

3. The method according to claim 2, wherein: The calling of the first agent to generate the schema content associated with the section of the target document includes: The document list is transmitted to the first agent, so that the first agent generates the schema content of each of the at least one target document in response to receiving the document list.

4. The method according to claim 3, wherein: The calling of the second agent to generate the target document based on the architecture content includes: The schema content of each of the at least one target document is transmitted to the second agent, so that the second agent generates the target document by expanding the schema content in response to receiving the schema content.

5. The method according to claim 4, wherein The architecture content includes a document reference tag, and the document reference tag is used to indicate references between different target documents.

6. The method according to claim 5, wherein: The document reference mark is expanded to display the associated chapter information of the referenced target document.

7. The method according to claim 4, wherein: The schema content includes an illustration tag, which is used to indicate a location where an illustration is to be embedded in the target document.

8. The method according to claim 7, wherein: The illustration markup is expanded into a graphic description language for representing the illustration, so that the second agent can call a drawing tool to generate the illustration based on the graphic description language.

9. The method according to any one of claims 2 to 8, wherein The plurality of agents include a fourth agent for quantifying document generation effects, and the method further includes: The fourth agent is called to evaluate the target document to obtain an evaluation result.

10. The method according to claim 9, wherein: The evaluation result includes an evaluation score, and calling the fourth agent to evaluate the target document to obtain the evaluation result includes: The target document is transmitted to the fourth agent, so that the fourth agent, in response to receiving the target document, scores the target document based on a preset scoring standard to obtain the evaluation score.

11. The method according to claim 10, wherein: The evaluation result also includes prompt information for improving the target document. The method further includes: In response to the evaluation score not reaching a preset threshold, calling the fourth agent to determine a first target agent associated with the improvement from the first agent to the third agent based on the prompt information; and The first target agent is called to improve the target document based on the prompt information.

12. The method according to any one of claims 2 to 8, wherein The method comprises: In response to receiving a request to update the target document, determining a second target agent among the first agent to the third agent that is associated with performing the update; and The second target agent is called to update the target document based on the request.

13. An agent-based document generation system, the system being configured to receive a document generation request to generate a target document, the system comprising: A first agent, configured to generate a document schema, for generating schema content associated with a section of the target document when called; as well as The second intelligent agent is used to generate document content so as to generate the target document based on the architecture content when called.

14. The system according to claim 13, wherein: The document generation request includes original data for generating the target document and a target category of the target document. The system further includes: A third agent is used to plan a document generation task to generate a document list indicating at least one target document when called, wherein the document list is determined based on at least a portion of the original data associated with the target category.

15. The system according to claim 14, wherein: The system further comprises: The fourth agent is used to quantify the document generation effect, so as to evaluate the target document when called to obtain an evaluation result.

16. The system according to claim 15, wherein: The evaluation result includes an evaluation score and prompt information for improving the target document. The fourth agent is also used to: in response to the evaluation score not reaching a preset threshold, determine the target agent associated with the improvement from the first agent to the third agent based on the prompt information.

17. An agent-based document generation device, comprising: an agent determination module configured to, in response to receiving a document generation request, determine a plurality of agents for processing the document generation request to generate a target document, wherein the plurality of agents include a first agent for generating a document structure and a second agent for generating document content; A first calling module is configured to call the first agent to generate a structure content associated with a section of the target document; and The second calling module is configured to call the second agent to generate the target document based on the architecture content.

18. An electronic device comprising: at least one processor; as well as a memory communicatively coupled to the at least one processor; in The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 12.

19. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause a computer to execute the method according to any one of claims 1-12.

20. A computer program product comprising a computer program, wherein The computer program implements the method according to any one of claims 1 to 12 when executed by a processor.

Images