Content generation method and device based on large model, equipment and medium

Abstract

The invention provides a content generation method and device based on a large model, equipment and a medium, and relates to the technical field of artificial intelligence, in particular to the technical fields of large models, man-machine interaction, intelligent creation and the like. According to the implementation scheme, in response to a received content generation instruction, the content generation instruction is split into multiple pieces of task description information; generating a content framework code corresponding to the content type information according to the content structure description information; obtaining a target material for content generation according to the content demand description information; according to the content framework code and a target material, generating a target code corresponding to the content type information; using a target engine corresponding to the content type information to render the target code into target content; and outputting the target content.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, in particular to technical fields such as large models, human-computer interaction, and intelligent creation, and specifically to a content generation method, device, electronic device, computer-readable storage medium, and computer program product based on large models. 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] In today's digital content creation field, with the vigorous development of AI technology, generative AI has been widely used in the creation of images, texts and other content. It is widely used in various fields, including but not limited to the production of product display pages in the e-commerce field, dynamic advertising creative generation in the advertising industry, interactive graphic reports in news media, interactive courseware production in the education field, and data visualization report generation within enterprises.

[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 a content generation method, apparatus, electronic device, computer-readable storage medium, and computer program product based on a large model.

[0006] According to one aspect of the present disclosure, a content generation method based on a large model is provided, comprising: in response to receiving a content generation instruction, splitting the content generation instruction into multiple task description information, wherein the multiple task description information includes content type information, content structure description information and content requirement description information; generating a content framework code corresponding to the content type information based on the content structure description information; acquiring a target material for content generation based on the content requirement description information; generating a target code corresponding to the content type information based on the content framework code and the target material; rendering the target code into target content using a target engine corresponding to the content type information; and outputting the target content.

[0007] According to another aspect of the present disclosure, a content generation device based on a large model is provided, including: a splitting unit, configured to split the content generation instruction into multiple task description information in response to receiving the content generation instruction, wherein the multiple task description information includes content type information, content structure description information and content requirement description information; a first generation unit, configured to generate a content framework code corresponding to the content type information based on the content structure description information; a first acquisition unit, configured to acquire a target material for content generation based on the content requirement description information; a second generation unit, configured to generate a target code corresponding to the content type information based on the content framework code and the target material; a rendering unit, configured to render the target code into target content using a target engine corresponding to the content type information; and an output unit, configured to output the target content.

[0008] 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; wherein the memory stores instructions executable 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 large model-based content generation method of the present disclosure.

[0009] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to cause a computer to execute the large model-based content generation method of the present disclosure.

[0010] According to another aspect of the present disclosure, a computer program product is provided, including a computer program, wherein when the computer program is executed by a processor, the computer program implements the large model-based content generation method of the present disclosure.

[0011] 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

[0012] 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.

[0013] 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;

[0014] Figure 2A flowchart of a method for generating content based on a large model according to an embodiment of the present disclosure is shown;

[0015] Figure 3 A schematic diagram illustrating target content of a static image type according to an exemplary embodiment of the present disclosure;

[0016] Figure 4 A flowchart of generating target code according to an embodiment of the present disclosure is shown;

[0017] Figure 5 A flowchart of a method for generating content based on a large model according to an embodiment of the present disclosure is shown;

[0018] Figure 6 A flowchart of re-executing target content generation according to an embodiment of the present disclosure is shown;

[0019] Figure 7 A flowchart showing a method for generating content based on a large model according to an exemplary embodiment of the present disclosure is shown.

[0020] Figure 8 FIG2 shows a structural block diagram of a content generation apparatus based on a large model according to an embodiment of the present disclosure;

[0021] Figure 9 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

[0022] 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.

[0023] 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.

[0024] 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.

[0025] The embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0026] 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.

[0027] In an embodiment of the present disclosure, the server 120 may run one or more services or software applications that enable the execution of the macro-model-based content generation method of the present disclosure.

[0028] 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.

[0029] 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.

[0030] The user may use client devices 101, 102, 103, 104, 105 and / or 106 to send content generation instructions. The client device may provide an interface that enables the user of the client device to interact with the client device. The client device may also output information to the user via the interface. 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.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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.

[0038] 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.

[0039] Figure 1 The system 100 may be configured and operated in various ways to enable the application of various methods and apparatuses described in accordance with the present disclosure.

[0040] According to the embodiments of the present disclosure, Figure 2 As shown, a content generation method based on a large model is provided, including: step S201, in response to receiving a content generation instruction, splitting the content generation instruction into multiple task description information, wherein the multiple task description information includes content type information, content structure description information and content requirement description information; step S202, generating a content framework code corresponding to the content type information based on the content structure description information; step S203, acquiring a target material for content generation based on the content requirement description information; step S204, generating a target code corresponding to the content type information based on the content framework code and the target material; step S205, rendering the target code into target content using a target engine corresponding to the content type information; and step S206, outputting the target content.

[0041] Therefore, after receiving the content generation instruction sent by the user, the content generation instruction is first split into multiple task description information, and the content framework code is generated according to the content structure description information. At the same time, the corresponding target material is obtained according to the content requirement description information. On this basis, the content framework code and the target material are integrated to generate a target code corresponding to the content type information. The corresponding target content can be obtained by rendering through the corresponding target engine, thereby improving the accuracy and efficiency of content generation of the large model, so that the generated target content is more in line with the content generation instruction.

[0042] In some embodiments, the content generation instruction may be sent by a user through a client device. In response to the client device receiving the content generation instruction, the content generation instruction may be input into a large model, so that the large model analyzes the content generation instruction and breaks the content generation instruction into multiple structured task description information.

[0043] In some examples, the content generation instruction may be "Generate a picture of the domestic new energy vehicle sales ranking list for Q3 2024. The picture style is concise and clear, highlighting the top 10 sales models, including model pictures, sales data and rankings, and the picture is interactive. Clicking on the model will pop up a detailed parameter introduction." Accordingly, the multiple structured task description information obtained by splitting can include "time range (Q3 2024), list type (domestic new energy vehicle sales ranking list), display requirements (top 10 models, model pictures, sales data, ranking, interactivity)" and so on.

[0044] In some embodiments, the plurality of task description information may include content type information, content structure description information, and content requirement description information of the target content to be generated. The content structure description information may include, for example, description information of the overall structure and style of the target content (e.g., image-type target content), the layout of the content modules, the content contained in each content module, and the like. The content requirement description information may include, for example, a description of the information required to be contained in the target content.

[0045] Continuing to refer to the above example, the content type information can indicate that the target content to be generated is of image type, the content structure description information can include "display requirements (top 10 car models, car model pictures, sales data, rankings, concise and clear)", and the content demand description information can include "time range (Q3 2024), list type (domestic new energy vehicle sales rankings), top 10 car models, car model pictures, sales data, rankings".

[0046] In some embodiments, the aforementioned large model may be a lightweight large language model deployed on a client device.

[0047] In some embodiments, the large model can be deployed on a cloud server. In some embodiments, the client can establish standardized communication with the external large model through the Model Context Protocol (MCP) gateway, encapsulate the request or response format, and achieve cross-platform compatibility.

[0048] In some embodiments, the corresponding target material can be obtained through a search engine based on the content requirement description information. For example, the required material image (such as the above-mentioned car model image) can be retrieved through a search engine.

[0049] In some embodiments, the content requirement description information may include knowledge acquisition requirements for knowledge materials in the target field. According to the content requirement description information, obtaining target materials for content generation may include: according to the knowledge acquisition requirements, calling the target service corresponding to the target field to obtain the knowledge materials in the target field based on the target service.

[0050] Therefore, by analyzing the knowledge acquisition requirements for the target field in the content requirement description information and calling the corresponding target service to obtain the corresponding knowledge material, the accuracy of the required target material can be improved, and then the accuracy of the subsequently generated target content can be improved, so that the generated target content is more in line with the content generation instructions.

[0051] In some embodiments, the target service may include various services on a platform based on the Model Context Protocol (MCP), such as a map service for generating routes, a weather service for obtaining a temperature line chart, and the like.

[0052] In some embodiments, based on the knowledge acquisition requirements, calling the target service corresponding to the target field to obtain the knowledge materials of the target field based on the target service may include: based on the knowledge acquisition requirements, obtaining the knowledge materials of the target field from the historical material library, wherein the historical material library stores the knowledge materials obtained in the execution of historical content generation; and in response to not obtaining the required knowledge materials in the historical material library, calling the target service to obtain the knowledge materials of the target field based on the target service.

[0053] Therefore, by prioritizing the recall of knowledge materials from the historical material library before calling the target service to obtain knowledge materials, the pressure of data transmission can be further reduced and the efficiency of content generation can be improved.

[0054] In some embodiments, the historical material library may also include the acquired historical material images. The above-mentioned various materials may be stored in association with the corresponding content requirement description information, thereby enabling the recall of materials from the historical material library through the content requirement description information.

[0055] In some embodiments, generating a content framework code corresponding to the content type information based on the content structure description information can be accomplished by inputting the content structure description information and prompt text into a large model, so that the large model analyzes the content structure description information under the guidance of the prompt text, and inputs a content framework code of the corresponding type based on the content type information and the content structure description information.

[0056] In some embodiments, generating content framework code corresponding to content type information based on content structure description information may include: generating content framework code based on content structure description information and first data, wherein the first data includes historical interaction data and at least one of feedback data for historical content generation results.

[0057] Therefore, by further introducing the first data in the process of generating the content framework, the accuracy of the content framework generated by the large model can be improved, so that the generated target content is more consistent with the content generation instructions, thereby improving the user experience.

[0058] In some embodiments, the first data may include at least one of historical interaction data and feedback data regarding historical content generation results. The historical interaction data may include behavioral data such as browsing history, historical conversation content, and historical clicks. The feedback data regarding historical content generation results may include, for example, user requests for content regeneration after content generation, user preferences for structure or color, and the like.

[0059] In some exemplary embodiments, the first data may also include user attribute information, such as age, gender, interests and hobbies.

[0060] It should be noted that the acquisition and use of the above-mentioned user attribute information, historical interaction data, and feedback data of historical content generation results have been clearly informed to users and have been explicitly authorized by users.

[0061] In some embodiments, the model context protocol can provide context chain management capabilities to maintain the consistency of user intent in multiple rounds of dialogue. For example, when generating HTML-based images, the image style or layout can be dynamically adjusted based on historical interaction data (such as the user's explicit preference for structure / color, etc.).

[0062] In some embodiments, generating a target code corresponding to the content type information based on the content framework code and the target material can be inputting the content framework code, the target material and multiple content description information into a large model, so that the large model fills each target material in the content framework according to the multiple content description information, thereby obtaining the target code for generating the target content.

[0063] In some embodiments, generating the target code corresponding to the content type information according to the content framework code and the target material may include: generating the target code according to the content framework code, the target material, and the first data.

[0064] Therefore, by further introducing the first data in the process of generating the target code, the accuracy of the target code generated by the large model can be improved, so that the generated target content is more consistent with the content generation instructions, thereby improving the user experience.

[0065] In some embodiments, generating a target code corresponding to the content type information based on the content framework code and the target material can be inputting the content framework code, the target material, the first data together with multiple content description information into a large model, so that the large model fills each target material in the content framework according to the multiple content description information and the user preferences reflected in the first data, thereby obtaining the target code for generating the target content.

[0066] In some embodiments, the content framework code, target material and first data are input into the big model so that the big model analyzes the first data and fills the target material into the content framework according to user preferences, thereby obtaining the target code for generating the target content.

[0067] In some embodiments, generating content framework code corresponding to content type information based on content structure description information may include: generating content framework code based on content structure description information and a reference template, wherein the reference template is obtained from a content template library based on multiple task description information, and the content template library includes code corresponding to historical content generation results as the content template.

[0068] Therefore, by further introducing reference templates in the process of generating content frameworks, it is possible to generate content frameworks by referring to historical content generation results with better performance, thereby further improving the accuracy and efficiency of content generation and enhancing user experience.

[0069] In some embodiments, during the historical content generation process, target content with good user feedback (e.g., high user click volume and long dwell time) can be saved in the content template library as a success case, thereby gradually forming a reusable template library. During the content generation process, the most matching content template can be recalled from the content template library based on the content description information, and used as part of the data for generating the content framework code and / or target code, and input into the macro model, so that the macro model can refer to the content template when generating the content framework code and / or target code, thereby improving the accuracy of content generation while improving the efficiency of content generation.

[0070] In some embodiments, the above-mentioned large model for generating the target code can be, for example, a multimodal large model. The multimodal large model can combine the text features and image features in the various input information, fully integrate the various target materials with the content framework, and input the corresponding target code.

[0071] In some embodiments, the content type information may indicate a static image type, the content frame code may be described by a page frame description language, and rendering the target code into target content using a target engine corresponding to the content type information may include: rendering the target code into a target page using a browser-driven engine; and obtaining the target content based on the target page.

[0072] Therefore, for the content to be generated whose content type information indicates that it is a static image type, by generating content framework code and content code based on the page framework description language, and rendering the page based on the browser driving engine, and then obtaining the target content (i.e., the target image) based on the target page, the convenience and accuracy of the target content generation can be improved.

[0073] In some embodiments, the page frame description language may be, for example, Hyper Text Markup Language (HTML) or Extensible Markup Language (XML), which is not limited here.

[0074] In some embodiments, the generated target code may be based on an HTML code framework and combined with CSS style sheets and SVG / Canvas graphic instructions.

[0075] In some embodiments, the generated code can be parsed in real time based on the browser-driven engine, and the target code can be rendered as a target page. Subsequently, the target page can be converted into a target image (i.e., target content) through a screenshot tool or a format conversion tool.

[0076] Figure 3 A schematic diagram illustrating target content of a static image type according to an exemplary embodiment of the present disclosure.

[0077] In some examples, for example, the content generation instruction can be "Generate a picture of the domestic new energy vehicle sales ranking list for the third quarter of 2024. The picture style is concise and clear, highlighting the top 10 sales models, including model pictures, sales data and rankings, and the picture is interactive, and clicking on the model will pop up a detailed parameter introduction."

[0078] The instructions are analyzed through a semantic parsing engine based on a large model to extract key information, such as the time range (July-October 2024), list type (domestic new energy vehicle sales ranking), display requirements (top 10 models, model images, sales data, ranking, interactivity), etc., and the relevant automobile data list service is called through the MCP protocol.

[0079] When interacting with the ranking service, you can obtain detailed car sales data, including model name, sales figures, etc., based on the system's pre-set rules and interface specifications. At the same time, the system uses the MCP protocol to match the corresponding model image from the image library based on the model information.

[0080] Subsequently, the multimodal model can be combined with the acquired data and images to generate content consisting of an HTML code framework, CSS style sheets, and JavaScript interactive logic. For example, the generated HTML code builds the basic structure of the ranking list, the CSS style sheet defines the layout, color, font, and other styles of the images and text, and the JavaScript code implements the interactive function of clicking on the model image to pop up detailed parameter information.

[0081] On this basis, the browser-driven engine can be used to render the generated code and convert the HTML content into a visual image. During the rendering process, the engine will accurately present the layout, style and interactive effects of the ranking list based on CSS style and JavaScript logic, and generate the final "2024 Q3 Domestic New Energy Vehicle Sales Ranking" image (such as Figure 3 shown).

[0082] The generated images can be directly embedded into the HTML code of marketing articles or downloaded for posting on social media platforms. When users browse articles on the web, they will see intuitive, beautiful, and interactive car sales ranking images. Clicking on the model in the image provides more detailed information, enriching the information display and interactive methods, and improving the user experience.

[0083] Traditionally, creating chart images requires marketers to manually collect data, design the layout, and then use graphic design software. This entire process can take hours or even days, resulting in low efficiency. However, the disclosed method, from inputting instructions to obtaining images, takes only minutes, significantly shortening the production cycle and improving the efficiency of marketing content production.

[0084] Using the MCP protocol to call data ranking services and image material libraries ensures the accuracy and timeliness of data, as well as the richness and adaptability of image materials. At the same time, the system's standardized processes and automated processing reduce human errors and ensure the consistent quality of each generated image.

[0085] In another example, the content generation instruction may be, for example, "Based on the 2025 climate change report, generate an interactive data visualization page, including a temperature change curve over the past decade and a distribution map of extreme weather events."

[0086] The system connects to external databases via the MCP protocol to obtain climate data, and combines this with large-scale models to generate HTML code and SVG chart descriptions for data visualization. A dynamic rendering engine combines this data with chart templates to generate interactive temperature curves (supporting time range filtering and data point details). Furthermore, the system automatically generates relevant illustrations based on data characteristics (e.g., glacier melting, rainstorms, and flooding) and embeds them into HTML pages.

[0087] Traditional data visualization requires the collaborative work of data analysts, designers, and developers, taking 1-3 days. However, the automated "data input - page output" process achieved with the disclosed embodiments only takes 15-30 minutes. Furthermore, the generated charts support real-time data updates, automatically refreshing the page content when the data source changes.

[0088] In some embodiments, the content type information may indicate a dynamic image type, the target code may include a target interactive element corresponding to the target interactive event, and rendering the target code into target content using a target engine corresponding to the content type information may include: using a dynamic image rendering engine to render the target code into target content, wherein the target content includes a target interactive element, and wherein the dynamic image rendering engine is also used to listen to the trigger signal of the target interactive event, and in response to listening to the trigger signal, render the display content corresponding to the target interactive event to replace at least part of the target content.

[0089] Therefore, by binding the target interactive elements and the target interactive events to each other, after rendering the target content, the display content can be transformed in real time according to the target interactive events by monitoring the trigger information of the target interactive events, thereby realizing the interactive function of dynamic images, enriching the interactive forms and scenes, and improving the user experience.

[0090] In some exemplary embodiments, real-time communication between HTML DOM elements and the underlying image rendering engine can be established, allowing, for example, image transformations to be automatically triggered when a user clicks a button. Through a custom event system, image elements can respond to various interactions within the HTML page (such as scrolling, hovering, and form submissions). This breaks through the limitations of traditional static image display, achieving a "code-as-animation" effect, enriching interactive forms and scenarios and significantly improving the user experience.

[0091] In some exemplary embodiments, a custom Web component can be created to represent a dynamic image in the DOM; within the component, communication is established between DOM events (such as clicks, inputs, etc.) and the image rendering engine; after rendering through an image rendering engine such as Canvas or WebGL, various custom target interaction events can be monitored, and when a trigger signal of a target interaction event is monitored, the corresponding rendering operation of the image rendering engine is triggered, thereby achieving a dynamic effect in which the information displayed in the image changes in response to the user's interactive behavior, enriching the interaction forms and scenarios, and improving the user interaction experience.

[0092] In some embodiments, the content type information may be a video type, and the target material may include text material and image material, such as Figure 4 As shown, generating a target code corresponding to the content type information based on the content framework code and the target material may include: step S401, generating video subtitles based on the text material; step S402, converting the video subtitles into speech audio; and step S403, filling the video subtitles, speech audio and image materials into the video frame corresponding to the content framework code according to multiple task description information to generate the target code.

[0093] In some embodiments, for target content of the video type, its content framework code may apply, for example, a professional-level video framework markup language (Synchronized Multimedia Integration Language, SMIL) or LottieJSON language, which is not limited here.

[0094] In some embodiments, based on multiple task description information, filling video subtitles, voice audio and image materials into the video frame corresponding to the content frame code to generate the target code can be achieved through a multimodal large model.

[0095] Therefore, through the preset workflow, after obtaining the target material, the video subtitles, voice audio and target code for generating the video are obtained in sequence, so that the automatic generation of the video can be realized, which further improves the accuracy of the content generation of the large model, makes the generated target content more consistent with the content generation instructions, and further enriches the interaction form and improves the user experience.

[0096] It is understandable that the target content of the present disclosure can also be any other type of content. Relevant technical personnel can define a workflow by themselves so that each node in the workflow is used to control the large model to complete a specific task, thereby gradually realizing the generation of target content. There is no limitation here.

[0097] In some embodiments, the above-mentioned large model-based content generation method may further include: obtaining display parameters of a target device for displaying target content; and wherein, rendering the target code into target content using a target engine corresponding to the content type information may include: adjusting the target code based on the display parameters; and rendering the adjusted target code into target content using the target engine, wherein the display size of the target content matches the content layout and the display parameters.

[0098] Therefore, by obtaining the display parameters of the target device and adjusting the target code according to the display parameters, the display size and content layout of the rendered target content can be adapted to the display parameters of the target device, thereby achieving adaptive adjustment of the target content, further improving the matching of the generated target content with the target device, and further improving the user experience by displaying the target content with matching size and layout on the target device.

[0099] In some embodiments, the target code can be adjusted based on the display parameters by inputting the above-mentioned display parameters, target code and corresponding prompt text into a large model, so that the large model, under the guidance of the prompt text, optimizes the resources and layout of the target code based on the display parameters, and outputs the target code for generating target content adapted to the target device.

[0100] In some embodiments, as Figure 5 As shown, the above-mentioned content generation method based on the big model may also include: step S501, performing an evaluation operation on the target content to obtain content evaluation information; step S502, judging whether it is necessary to regenerate the target content based on the content evaluation information; and step S503, in response to judging that it is necessary to regenerate the target content, re-executing the generation of the target content based on the content evaluation information and the content generation instruction.

[0101] Therefore, by evaluating the target content and determining whether the target content needs to be regenerated based on the content evaluation information, adaptive optimization of the target content can be achieved, so that the large model can generate more accurate target content and improve the user experience.

[0102] In some embodiments, the above-mentioned evaluation operation may include at least one of the following: performing accuracy verification on the target content to obtain an accuracy verification result, wherein the accuracy verification result is used to indicate whether the target content contains erroneous content; scoring the content quality of the target content to obtain a content quality score of the target content; and obtaining user feedback information on the target content.

[0103] Therefore, by evaluating the target content from different dimensions, more comprehensive evaluation information can be obtained, which can make subsequent optimization of the target content more accurate, and the regenerated target content can be more accurate and better match the content generation instructions, thereby improving the user experience.

[0104] In some embodiments, the target content can be intelligently verified for accuracy based on the large model to determine whether there are errors in the target content, such as knowledge materials. In some embodiments, manual review can also be performed based on the verification performed on the large model to further improve the accuracy of the verification.

[0105] In some embodiments, content quality scoring dimensions may include, but are not limited to, content completeness, content layout, color style, video coherence, and other quality dimensions. In some embodiments, the target content, user-entered content generation instructions, and corresponding prompt text can be input into the macro model, allowing the macro model to conduct a comprehensive evaluation and scoring of each of the aforementioned quality dimensions under the guidance of the prompt text.

[0106] In some embodiments, the feedback information may include, for example, various pieces of feedback information input by the user with respect to the target content.

[0107] In some embodiments, determining whether the target content needs to be regenerated based on the content evaluation information may include determining whether the target content needs to be regenerated based on at least one of an accuracy check result, a content quality score, and feedback information.

[0108] Therefore, by comprehensively considering at least one of the above evaluation information to determine whether to regenerate the target content, the accuracy of determining whether to regenerate the content can be further improved, thereby avoiding regenerating target content with higher accuracy and affecting content generation efficiency.

[0109] In some embodiments, based on at least one of the accuracy verification results, content quality score and feedback information, determining whether it is necessary to regenerate the target content can be done by inputting all of the above information and the corresponding prompt text into a large model, so that the large model can conduct a comprehensive analysis of the various information under the guidance of the prompt text to determine whether it is necessary to regenerate the target content.

[0110] In some embodiments, the content evaluation information may include accuracy check results, content quality scores, and feedback information, such as Figure 6As shown, based on the content evaluation information and the content generation instructions, re-executing the generation of the target content may include: step S601, adjusting the content framework code according to the feedback information and the content quality score; step S602, re-acquiring the corresponding material for the erroneous content according to the accuracy verification result to update the target material; step S603, regenerating the target code according to the adjusted content framework code and the updated target material; and step S604, using the target engine to re-render the regenerated target code as the target content.

[0111] In some embodiments, adjusting the content framework code according to the feedback information and the content quality score may be adjusting and optimizing the content framework code through a large model according to the feedback information and the content quality scores of various quality dimensions.

[0112] In some embodiments, adjusting the content framework code based on feedback information and content quality scores can be first optimizing the prompt text used to guide the big model to generate content framework code based on the feedback information and content quality scores of each quality dimension through the big model (for example, supplementing the prompt text with information such as the user's preference for the target content, the part of the content framework to be optimized, etc.), and adjusting and optimizing the content framework code through the big model based on the optimized prompt text.

[0113] In some embodiments, according to the accuracy check result, reacquiring corresponding materials for the erroneous content to update the target materials may be reacquiring the corresponding materials through a networking function.

[0114] In some embodiments, the prompt text used to regenerate the target code can also be optimized based on the above method, and the adjusted content framework code, updated target material and optimized prompt text are input into the multimodal large model to obtain the target code regenerated by the large model, and then the regenerated target content is obtained.

[0115] Therefore, by utilizing various evaluation information to optimize the steps of generating each target content, the accuracy of the target content regenerated by the large model can be improved as a whole, and the subsequent optimization of the target content can be more in line with the content generation instructions, thereby improving the user experience.

[0116] In some embodiments, the aforementioned evaluation operation may be performed again on the regenerated target content until it is determined based on the evaluation information that the target content does not need to be regenerated.

[0117] In some embodiments, the interaction data between the terminal users and the generated page (such as click rate and dwell time) can also be collected to form a quality evaluation index; then, through the reinforcement learning algorithm, the large model parameters are fine-tuned according to the evaluation results to gradually improve the quality of the generated content.

[0118] Figure 7 A flowchart of a content generation method based on a large model according to an exemplary embodiment of the present disclosure is shown.

[0119] In some exemplary embodiments, a content generation method based on a large model may include: a server receiving a content generation instruction transmitted by a client. The server performs the following steps by calling the large model: splitting the content generation instruction into multiple task description information, including content type information, content structure description information, and content requirement description information; generating content framework code corresponding to the content type information based on the content structure description information, first data, content template, etc.; obtaining each target material corresponding to the content requirement description information from a material library and knowledge service; integrating the target material and the content framework code based on the first data and content template to obtain a target code; rendering the target code based on a target engine to obtain target content; evaluating the target content to obtain evaluation information of each dimension of the target content; and in response to determining that the target content needs to be regenerated based on the evaluation information, optimizing and adjusting the target content based on the evaluation information to generate an optimized target code.

[0120] In some embodiments, as Figure 8 As shown, a content generation device 800 based on a large model is provided, including: a splitting unit 810, configured to split the content generation instruction into multiple task description information in response to receiving the content generation instruction, wherein the multiple task description information includes content type information, content structure description information and content requirement description information; a first generation unit 820, configured to generate a content framework code corresponding to the content type information based on the content structure description information; a first acquisition unit 830, configured to acquire a target material for content generation based on the content requirement description information; a second generation unit 840, configured to generate a target code corresponding to the content type information based on the content framework code and the target material; a rendering unit 850, configured to render the target code into target content using a target engine corresponding to the content type information; and an output unit 860, configured to output the target content.

[0121] Among them, the operations performed by units 810 to 860 in the above-mentioned large model-based content generation device 800 and the effects that can be achieved are similar to steps S201 to S206 in the above-mentioned large model-based content generation method, and will not be repeated here.

[0122] In some embodiments, the content type information can indicate a static image type, the content frame code is described by a page frame description language, and the rendering unit is configured to: use a browser-driven engine to render the target code into a target page; and obtain target content based on the target page.

[0123] In some embodiments, the content type information can be indicated as a dynamic image type, the target code can include a target interactive element corresponding to the target interactive event, and the rendering unit can be configured to: use a dynamic image rendering engine to render the target code into target content, wherein the target content includes the target interactive element, and wherein the dynamic image rendering engine can also be used to listen to the trigger signal of the target interactive event, and in response to listening to the trigger signal, render the display content corresponding to the target interactive event to replace at least part of the target content.

[0124] In some embodiments, the content type information can indicate a video type, the target material can include text material and image material, and the second generation unit can be configured to: generate video subtitles based on the text material; convert the video subtitles into speech audio; and fill the video subtitles, speech audio and image materials into the video frame corresponding to the content frame code according to multiple task description information to generate the target code.

[0125] In some embodiments, the above-mentioned large model-based content generation device may further include: a third acquisition unit, configured to acquire display parameters of a target device for displaying target content; and wherein the rendering unit may be configured to: adjust the target code based on the display parameters; and render the adjusted target code into target content using a target engine, wherein the display size of the target content matches the content layout and the display parameters.

[0126] In some embodiments, the content requirement description information may include knowledge acquisition requirements for knowledge materials in the target field, and the first acquisition unit may be configured to: call the target service corresponding to the target field according to the knowledge acquisition requirements to acquire the knowledge materials in the target field based on the target service.

[0127] In some embodiments, based on the knowledge acquisition requirements, calling the target service corresponding to the target field to obtain the knowledge materials of the target field based on the target service may include: based on the knowledge acquisition requirements, obtaining the knowledge materials of the target field from the historical material library, wherein the historical material library stores the knowledge materials obtained in the execution of historical content generation; and in response to not obtaining the required knowledge materials in the historical material library, calling the target service to obtain the knowledge materials of the target field based on the target service.

[0128] In some embodiments, the first generating unit may be configured to generate content framework code according to the content structure description information and first data, wherein the first data includes at least one of historical interaction data and feedback data for historical content generation results.

[0129] In some embodiments, the second generating unit may be configured to generate a target code according to the content framework code, the target material and the first data.

[0130] In some embodiments, the first generation unit can also be configured to: generate content framework code based on content structure description information and a reference template, wherein the reference template is obtained from a content template library based on multiple task description information, and the content template library includes code corresponding to the historical content generation result as the content template.

[0131] In some embodiments, the above-mentioned large model-based content generation device may also include: an evaluation unit, configured to perform an evaluation operation on the target content to obtain content evaluation information; a judgment unit, configured to judge whether it is necessary to regenerate the target content based on the content evaluation information; and an execution unit, configured to re-execute the generation of the target content based on the content evaluation information and the content generation instruction in response to the judgment that it is necessary to regenerate the target content.

[0132] In some embodiments, performing an evaluation operation on the target content to obtain content evaluation information may include at least one of the following: performing an accuracy check on the target content to obtain an accuracy check result, wherein the accuracy check result is used to indicate whether the target content contains erroneous content; scoring the target content for content quality to obtain a content quality score for the target content; and obtaining user feedback information on the target content.

[0133] In some embodiments, the determination unit may be configured to determine whether the target content needs to be regenerated based on at least one of the accuracy check result, the content quality score, and the feedback information.

[0134] In some embodiments, the content evaluation information includes accuracy verification results, content quality scores and feedback information, and the execution unit can be further configured to: adjust the content framework code based on the feedback information and content quality score; re-acquire corresponding materials for the erroneous content based on the accuracy verification results to update the target materials; regenerate the target code based on the adjusted content framework code and the updated target materials; and use the target engine to re-render the regenerated target code as the target content.

[0135] 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.

[0136] According to an embodiment of the present disclosure, an electronic device, a readable storage medium, and a computer program product are also provided.

[0137] refer to Figure 9 , a block diagram of an electronic device 900 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 processing, 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 9 As shown, the electronic device 900 includes a computing unit 901, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 902 or a computer program loaded from a storage unit 908 into a random access memory (RAM) 903. In the RAM 903, various programs and data required for the operation of the electronic device 900 can also be stored. The computing unit 901, the ROM 902, and the RAM 903 are connected to each other via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0139] Multiple components within electronic device 900 are connected to I / O interface 905, including an input unit 906, an output unit 907, a storage unit 908, and a communication unit 909. Input unit 906 can be any type of device capable of inputting information into electronic device 900. Input unit 906 can receive input numeric or character information and generate key signal input related to user settings and / or function control of the electronic device. It may include, but is not limited to, a mouse, keyboard, touch screen, trackpad, trackball, joystick, microphone, and / or remote control. Output unit 907 can be any type of device capable of presenting information, and may include, but is not limited to, a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. Storage unit 908 may include, but is not limited to, a magnetic disk or an optical disk. Communication unit 909 allows electronic device 900 to exchange information / data with other devices via computer networks such as the Internet and / or various telecommunication networks. It may include, but is not limited to, a modem, a network card, an infrared communication device, a wireless communication transceiver, and / or 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 901 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the computing unit 901 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, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 901 performs the various methods and processes described above, such as the aforementioned large-model-based content generation method. For example, in some embodiments, the aforementioned large-model-based content generation method can be implemented as a computer software program that is tangibly contained in a machine-readable medium, such as the storage unit 908. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 900 via the ROM 902 and / or the communication unit 909. When the computer program is loaded into the RAM 903 and executed by the computing unit 901, one or more steps of the aforementioned large-model-based content generation method can be performed. Alternatively, in other embodiments, the computing unit 901 may be configured to execute the above-mentioned large model-based content generation method in any other appropriate manner (for example, 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 having 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), and the Internet.

[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 content generation method based on a large model, comprising: In response to receiving the content generation instruction, splitting the content generation instruction into a plurality of task description information, wherein the plurality of task description information includes content type information, content structure description information and content requirement description information; generating a content framework code corresponding to the content type information according to the content structure description information; Acquiring target materials for content generation according to the content requirement description information; generating a target code corresponding to the content type information according to the content framework code and the target material; Rendering the target code into target content using a target engine corresponding to the content type information; and The target content is output.

2. The method according to claim 1, wherein The content type information indicates a static image type, the content frame code is described in a page frame description language, and the rendering of the target code into target content using a target engine corresponding to the content type information includes: Utilizing a browser-driven engine to render the target code into a target page; and Based on the target page, the target content is acquired.

3. The method according to claim 1, wherein The content type information indicates a dynamic image type, the target code includes a target interactive element corresponding to a target interactive event, and the rendering of the target code into target content using a target engine corresponding to the content type information includes: The target code is rendered into the target content using a dynamic image rendering engine, wherein the target content includes the target interactive element, and wherein the dynamic image rendering engine is also used to monitor the trigger signal of the target interactive event, and in response to monitoring the trigger signal, render the display content corresponding to the target interactive event to replace at least a portion of the target content.

4. The method according to claim 1, wherein The content type information indicates a video type, the target material includes a text material and an image material, and generating a target code corresponding to the content type information according to the content framework code and the target material includes: Generate video subtitles based on the text material; Converting the video subtitles into speech audio; and According to the plurality of task description information, the video subtitles, the voice audio, and the image material are filled into the video frame corresponding to the content frame code to generate the target code.

5. The method according to any one of claims 1 to 4, further comprising: Acquire display parameters of a target device for displaying the target content; And among them, The rendering of the target code into target content by using a target engine corresponding to the content type information includes: Based on the display parameters, adjusting the target code; and The adjusted target code is rendered into the target content using the target engine, wherein the display size and content layout of the target content match the display parameters.

6. The method according to any one of claims 1 to 5, wherein The content requirement description information includes knowledge acquisition requirements for knowledge materials in a target field, and acquiring target materials for content generation according to the content requirement description information includes: According to the knowledge acquisition requirement, a target service corresponding to the target field is called to acquire knowledge materials in the target field based on the target service.

7. The method according to claim 6, wherein: The calling of a target service corresponding to the target domain according to the knowledge acquisition requirement to acquire knowledge materials of the target domain based on the target service includes: According to the knowledge acquisition requirement, acquiring knowledge materials in the target field from a historical material library, wherein the historical material library stores knowledge materials acquired during the execution of historical content generation; and In response to not obtaining the required knowledge material in the historical material library, the target service is called to obtain the knowledge material in the target field based on the target service.

8. The method according to any one of claims 1 to 7, wherein Generating a content framework code corresponding to the content type information according to the content structure description information includes: The content framework code is generated according to the content structure description information and first data, wherein the first data includes at least one of historical interaction data and feedback data for historical content generation results.

9. The method according to claim 8, wherein Generating a target code corresponding to the content type information according to the content framework code and the target material includes: The target code is generated according to the content framework code, the target material and the first data.

10. The method according to any one of claims 1 to 9, wherein Generating a content framework code corresponding to the content type information according to the content structure description information includes: The content framework code is generated according to the content structure description information and a reference template, wherein the reference template is obtained from a content template library according to the plurality of task description information, and the content template library includes codes corresponding to historical content generation results as content templates.

11. The method according to claim 10, further comprising: Performing an evaluation operation on the target content to obtain content evaluation information; Determining whether to regenerate the target content based on the content evaluation information; as well as In response to a determination that regeneration of the target content needs to be performed, generation of the target content is re-executed based on the content evaluation information and the content generation instruction.

12. The method according to claim 11, wherein The evaluation operation includes at least one of the following: Performing an accuracy check on the target content to obtain an accuracy check result, wherein the accuracy check result is used to indicate whether the target content contains erroneous content; Scoring the target content for content quality to obtain a content quality score for the target content; and Obtain user feedback information regarding the target content.

13. The method according to claim 12, wherein: The determining whether it is necessary to regenerate the target content according to the content evaluation information includes: Whether it is necessary to regenerate the target content is determined according to at least one of the accuracy check result, the content quality score, and the feedback information.

14. The method according to claim 13, wherein The content evaluation information includes the accuracy check result, the content quality score, and the feedback information. The re-executing the generation of the target content based on the content evaluation information and the content generation instruction includes: Adjusting the content framework code according to the feedback information and the content quality score; Reacquiring corresponding materials for the erroneous content according to the accuracy check result to update the target material; regenerating the target code according to the adjusted content framework code and the updated target material; and The target engine is used to re-render the regenerated target code into target content.

15. A content generation device based on a large model, comprising: a splitting unit configured to, in response to receiving a content generation instruction, split the content generation instruction into a plurality of task description information, wherein the plurality of task description information includes content type information, content structure description information and content requirement description information; A first generating unit is configured to generate a content framework code corresponding to the content type information according to the content structure description information; A first acquisition unit is configured to acquire target materials for content generation according to the content requirement description information; a second generating unit configured to generate a target code corresponding to the content type information according to the content framework code and the target material; a rendering unit configured to render the target code into target content using a target engine corresponding to the content type information; and The output unit is configured to output the target content.

16. 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 14.

17. 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 1-14.

18. A computer program product comprising a computer program, wherein When the computer program is executed by a processor, the method according to any one of claims 1 to 14 is implemented.

Images