Content automatic generation method, computer device, medium and system

Abstract

This application relates to the field of computer technology and provides a method, computer device, medium, and system for automatically generating content. The method includes: obtaining hot topics through a hotspot screening module; inputting the hot topics into multiple generative artificial intelligence models in a hotspot analysis module to obtain content associated with the hot topics generated by each of the multiple generative artificial intelligence models; and transmitting the content associated with the hot topics generated by each of the multiple generative artificial intelligence models to a style conversion module; and generating stylized content for the hot topics through the style conversion module based on system prompts that at least include style prompts and using the content associated with the hot topics generated by each of the multiple generative artificial intelligence models. This improves overall efficiency and meets personalized needs.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, computer device, medium, and system for automatically generating content. Background Art

[0002] Generative Artificial Intelligence (GAI) refers to the process of generating logical and coherent new content based on user input. For example, it uses algorithms and models to generate text, images, audio, and video. By employing various large-scale model technologies, multimodal techniques, and generative algorithms, AI-generated content (AIGC) has been widely applied in various fields. It typically generates matching text, images, and audio / video content based on a prompt. In the field of hot news processing, it is often necessary to quickly generate relevant news reports and analytical results based on hot topics. This requires considering the timeliness of hot topics, meaning their popularity can fade quickly, and the complex and ever-changing needs of audiences. To this end, it is necessary to be able to quickly generate relevant news processing results for market trends. This requires not only covering audiences and related information across different themes and industries, but also providing results with varying levels of content richness and analytical perspectives to specifically meet the needs of different audiences. These challenges pose challenges to the application of GAI and AIGC in this field. In the prior art, one content generation solution relies on manual retrieval to manage content classification and manual review to overcome the uncontrollability and instability of GAI in content generation. However, this results in high time costs and low execution efficiency, making it difficult to capture hot spots in a timely manner and provide targeted services for different customer needs. Another content generation solution uses stylized prompt words to match GAIs of corresponding style types from multiple candidate GAIs, and then generates content based on the matched GAIs. In other words, using prompt words to overcome the uncontrollability and instability of GAI in content generation. An example of this can be found in Chinese patent application number CN117115303A, "Method, system, computing device and storage medium for content generation." However, this content generation solution that uses stylized prompt words and matching GAIs requires that each time content is generated, a matching GAI must be found first and then content generation can only rely on the matching GAI. This results in low model resource utilization. Moreover, with the rapid changes in hot topics and the rapid changes in the needs of the audience, it may be necessary to frequently generate stylized prompt words, match models, and regenerate content, making it difficult to achieve rapid response and large-scale coverage, and is not conducive to improving overall efficiency.

[0003] To this end, the present application provides a method, computer device, medium and system for automatically generating content to address the technical difficulties in the prior art. Summary of the Invention

[0004] In a first aspect, the present application provides a method for automatically generating content. The method comprises: obtaining hot topics through a hotspot screening module; inputting the hot topics into multiple generative artificial intelligence models in a hotspot analysis module, thereby obtaining content associated with the hot topics generated by each of the multiple generative artificial intelligence models; and transmitting the content associated with the hot topics generated by each of the multiple generative artificial intelligence models to a style conversion module, wherein the optimization strategies of the multiple generative artificial intelligence models each have different optimization objectives for the model input, and the multiple generative artificial intelligence models are optimized independently of each other; and generating stylized content for the hot topics through the style conversion module based on system prompts including at least style prompts and using the content associated with the hot topics generated by each of the multiple generative artificial intelligence models, wherein the stylized content for the hot topics has a style corresponding to the style prompts.

[0005] Through the first aspect of the present application, through the separate design of content generation and stylization processing, multiple generative artificial intelligence models are used to generate content related to the hot topics. This helps to give full play to the design advantages of multi-model parallel processing, and can improve the content generation speed by increasing the parallelism of model operation. It can also expand the overall coverage by increasing the differences between different generative artificial intelligence models. By stylizing the content related to the hot topics generated by the multiple generative artificial intelligence models, it can support the free combination of core scene content and content style conversion, and support the generation of content of different styles to meet the personalized needs of different audience groups. In this way, it can not only cover the audience groups and related information within the scope of different subject industries, but also provide results with different content richness and different analysis angles, which is conducive to targeted satisfaction of the needs of different audience groups, with low time cost, high execution efficiency, and the ability to capture hot spots in a timely manner and provide targeted services for different customer needs. In addition, through the parallel operation of multiple models, the utilization rate of model resources is improved, and with the rapid changes in hot topics and the rapid changes in the needs of the audience groups, rapid response and wide coverage can be achieved through content recombination, which is conducive to improving overall efficiency.

[0006] In a possible implementation of the first aspect of the present application, the hot topic is used to indicate the content characteristics of the stylized content for the hot topic, the style prompt word is used to indicate the style characteristics of the stylized content for the hot topic, and the system prompt word also includes multiple task prompt words, and the multiple task prompt words correspond one-to-one to the multiple generative artificial intelligence models. Each of the multiple task prompt words is used for a model service call of the generative artificial intelligence model corresponding to the task prompt word among the multiple generative artificial intelligence models.

[0007] In a possible implementation of the first aspect of the present application, the system prompt words are provided to the hotspot analysis module and the style conversion module by the prompt word editing module, and the function of the prompt word editing module for generating the multiple task prompt words is based on content prompt words, content preferences corresponding to the content prompt words, and content associated with the content prompt words generated by each of the multiple generative artificial intelligence models.

[0008] In a possible implementation of the first aspect of the present application, the multiple task prompt words include settings of the multiple generative artificial intelligence models regarding artificial intelligence agents, domain common sense, special variables, processing tasks, and response formats.

[0009] In a possible implementation of the first aspect of the present application, the style conversion module is used to generate the stylized content for the hot topic based on the system prompt words that at least include the style prompt words and the content associated with the hot topic generated by each of the multiple generative artificial intelligence models, including: integrating the content associated with the hot topic generated by each of the multiple generative artificial intelligence models according to a standard style, thereby obtaining standardized content for the hot topic with the standard style; based on the style prompt words, selectively performing content deletion operations and selective stylized element addition operations on the standardized content for the hot topic, thereby obtaining the stylized content for the hot topic with the style corresponding to the style prompt words.

[0010] In a possible implementation of the first aspect of the present application, the standard style is independent of the style corresponding to the style prompt word, and the standardized content for the hot topic with the standard style is independent of the style prompt word.

[0011] In a possible implementation of the first aspect of the present application, the stylized element adding operation includes adding at least one stylized element of a style corresponding to the style prompt word, where the at least one stylized element is selected from a plurality of stylized elements, and the plurality of stylized elements includes symbols, emoticons, and pictures.

[0012] In a possible implementation of the first aspect of the present application, the multiple generative artificial intelligence models provide content associated with the hot topics generated by each of the multiple generative artificial intelligence models through multi-model parallel operation, and the updates of the multiple generative artificial intelligence models are performed independently of each other.

[0013] In a possible implementation of the first aspect of the present application, the system prompt word also includes the weight ratio of each of the multiple generative artificial intelligence models, and the weight ratio of each of the multiple generative artificial intelligence models is used to indicate the importance of the content associated with the hot topic generated by each of the multiple generative artificial intelligence models relative to the stylized content for the hot topic, and when there is an updated generative artificial intelligence model among the multiple generative artificial intelligence models and the weight ratio of the updated generative artificial intelligence model is not higher than a preset threshold, the old model of the updated generative artificial intelligence model is used to ensure that the multi-model parallel operation of the multiple generative artificial intelligence models is not interrupted.

[0014] In a possible implementation of the first aspect of the present application, the style conversion module generates the stylized content for the hot topic based on the system prompt word that at least includes the style prompt word and the content associated with the hot topic generated by each of the multiple generative artificial intelligence models, including: integrating the content associated with the hot topic generated by each of the multiple generative artificial intelligence models according to a plurality of preset standard styles, thereby obtaining a plurality of standardized content for the hot topic, each of the plurality of standardized content for the hot topic having the plurality of preset standard styles; based on the style prompt word, selecting a preset standard style that matches the style prompt word from the plurality of preset standard styles, and determining the standardized content for the hot topic with the selected preset standard style from the plurality of standardized content for the hot topic as a reference standardized content; based on the style prompt word, selectively performing a content deletion operation and a stylized element addition operation on the reference standardized content, thereby obtaining the stylized content for the hot topic having the style corresponding to the style prompt word.

[0015] In a possible implementation of the first aspect of the present application, the multiple preset standard styles are key styles determined based on user demand prediction.

[0016] In a possible implementation of the first aspect of the present application, the hot topic is a key hot topic determined based on the user demand prediction.

[0017] In a possible implementation of the first aspect of the present application, the style prompt word is selected from a plurality of preset style prompt words, and the plurality of preset style prompt words include standard style, consulting style, advisory style, relaxed style, and concise style.

[0018] In a possible implementation of the first aspect of the present application, the multiple generative artificial intelligence models include a first generative artificial intelligence model, a second generative artificial intelligence model, a third generative artificial intelligence model, a fourth generative artificial intelligence model and a fifth generative artificial intelligence model, wherein the optimization strategy of the first generative artificial intelligence model has an optimization goal for the relevance and timeliness of the model input, the optimization strategy of the second generative artificial intelligence model has an optimization goal for information extraction and in-depth research of the model input, the third generative artificial intelligence model has an optimization goal for trend prediction of the model input, the optimization strategy of the fourth generative artificial intelligence model has an optimization goal for multi-industry analysis results of the model input, and the optimization strategy of the fifth generative artificial intelligence model has an optimization goal for event context analysis results of the model input.

[0019] In a possible implementation of the first aspect of the present application, the stylized content for the hot topic is a hot analysis report for the hot topic, the content associated with the hot topic generated by the first generative artificial intelligence model is the hot information scenario content associated with the hot topic, the content associated with the hot topic generated by the second generative artificial intelligence model is the research report analysis scenario content associated with the hot topic, the content associated with the hot topic generated by the third generative artificial intelligence model is the external market analysis scenario content associated with the hot topic, the content associated with the hot topic generated by the fourth generative artificial intelligence model is the multi-industry analysis scenario content associated with the hot topic, and the content associated with the hot topic generated by the fifth generative artificial intelligence model is the event development context scenario content associated with the hot topic.

[0020] In a possible implementation of the first aspect of the present application, the style prompt words are used to set the combination between the scene contents generated by each of the multiple generative artificial intelligence models.

[0021] In a possible implementation of the first aspect of the present application, the hot information scenario content associated with the hot topic is based on the hot information information associated with the hot topic after vectorization processing, the event development context scenario content associated with the hot topic is the event development context output result obtained by sorting out the hot topic according to the timeline associated with the hot topic through a large language model, and the external market analysis scenario content associated with the hot topic and the multi-industry analysis scenario content associated with the hot topic are obtained by hot spot analysis through artificial intelligence agents and analysis frameworks.

[0022] In a possible implementation of the first aspect of the present application, the automatic content generation method also includes: through the style conversion module, based on another style prompt word different from the style prompt word, using the content associated with the hot topic generated by each of the multiple generative artificial intelligence models, generating another stylized content for the hot topic, wherein the another stylized content for the hot topic has a style corresponding to the another style prompt word.

[0023] In a possible implementation of the first aspect of the present application, the hot topic is user-defined, or the hot topic is obtained by processing public domain hot information data through the hot spot screening module based on a hot spot capture algorithm, a hot spot trend analysis algorithm, and a hot word analysis algorithm.

[0024] In a second aspect, an embodiment of the present application further provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements a method according to any one of the implementation methods of any of the above aspects when executing the computer program.

[0025] In a third aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores computer instructions. When the computer instructions are executed on a computer device, the computer device executes a method according to any one of the implementation methods of any of the above aspects.

[0026] In a fourth aspect, an embodiment of the present application further provides a computer program product, which includes instructions stored on a computer-readable storage medium, and when the instructions are executed on a computer device, the computer device executes a method according to any one of the implementation methods of any of the above aspects.

[0027] In a fifth aspect, the present application provides a content automatic generation system. The content automatic generation system includes: a hot spot screening module for obtaining hot topics; a hot spot analysis module, the hot spot analysis module including multiple generative artificial intelligence models, the hot spot analysis module for: inputting the hot topics into the multiple generative artificial intelligence models, thereby obtaining the content associated with the hot topics generated by each of the multiple generative artificial intelligence models, and transmitting the content associated with the hot topics generated by each of the multiple generative artificial intelligence models to a style conversion module, wherein the optimization strategies of the multiple generative artificial intelligence models each have different optimization objectives for the model input, and the multiple generative artificial intelligence models are optimized independently of each other; the style conversion module for: based on system prompts including at least style prompts, using the content associated with the hot topics generated by each of the multiple generative artificial intelligence models, generating stylized content for the hot topics, wherein the stylized content for the hot topics has a style corresponding to the style prompts.

[0028] According to the fifth aspect of the present application, through the separate design of content generation and stylization processing, multiple generative artificial intelligence models are used to generate content related to the hot topics. This helps to give full play to the design advantages of multi-model parallel processing, and can improve the content generation speed by increasing the parallelism of model operation. It can also expand the overall coverage by increasing the differences between different generative artificial intelligence models. By stylizing the content related to the hot topics generated by the multiple generative artificial intelligence models, it can support the free combination of core scene content and content style conversion, and support the generation of content of different styles to meet the personalized needs of different audience groups. In this way, it can not only cover the audience groups and related information within the scope of different subject industries, but also provide results with different content richness and different analysis angles, which is conducive to targeted satisfaction of the needs of different audience groups, with low time cost, high execution efficiency, and the ability to capture hot spots in a timely manner and provide targeted services for different customer needs. In addition, through the parallel operation of multiple models, the utilization rate of model resources is improved, and with the rapid changes in hot topics and the rapid changes in the needs of the audience groups, rapid response and wide coverage can be achieved through content recombination, which is conducive to improving overall efficiency.

[0029] In a possible implementation of the fifth aspect of the present application, the hot topic is used to indicate the content characteristics of the stylized content for the hot topic, the style prompt word is used to indicate the style characteristics of the stylized content for the hot topic, and the system prompt word also includes multiple task prompt words, and the multiple task prompt words correspond one-to-one to the multiple generative artificial intelligence models. Each of the multiple task prompt words is used for a model service call of the generative artificial intelligence model corresponding to the task prompt word among the multiple generative artificial intelligence models.

[0030] In a possible implementation of the fifth aspect of the present application, the style conversion module is used to: generate the stylized content for the hot topic based on the system prompt words that at least include the style prompt words, using the content associated with the hot topic generated by each of the multiple generative artificial intelligence models, including: integrating the content associated with the hot topic generated by each of the multiple generative artificial intelligence models according to a standard style, thereby obtaining standardized content for the hot topic with the standard style; based on the style prompt words, selectively performing content deletion operations and selective stylized element addition operations on the standardized content for the hot topic, thereby obtaining the stylized content for the hot topic with the style corresponding to the style prompt words.

[0031] In a possible implementation of the fifth aspect of the present application, the multiple generative artificial intelligence models include a first generative artificial intelligence model, a second generative artificial intelligence model, a third generative artificial intelligence model, a fourth generative artificial intelligence model and a fifth generative artificial intelligence model, wherein the optimization strategy of the first generative artificial intelligence model has an optimization goal for the relevance and timeliness of the model input, the optimization strategy of the second generative artificial intelligence model has an optimization goal for information extraction and in-depth research of the model input, the third generative artificial intelligence model has an optimization goal for trend prediction of the model input, the optimization strategy of the fourth generative artificial intelligence model has an optimization goal for multi-industry analysis results of the model input, and the optimization strategy of the fifth generative artificial intelligence model has an optimization goal for event context analysis results of the model input. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0033] Figure 1 A flowchart of a method for automatically generating content provided in an embodiment of the present application;

[0034] Figure 2 The first embodiment of the present application provides a Figure 1 A schematic diagram of the business process flow of the content automatic generation method shown;

[0035] Figure 3 The second embodiment of the present application provides a Figure 1 A schematic diagram of the business process flow of the content automatic generation method shown;

[0036] Figure 4 A schematic diagram of a content automatic generation system provided in an embodiment of the present application;

[0037] Figure 5 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0038] The embodiments of the present application will be described in further detail below with reference to the accompanying drawings.

[0039] It should be understood that, in the description of this application, "at least one" means one or more, and "a plurality" means two or more. In addition, unless otherwise specified, the terms "first" and "second" are used only for descriptive purposes and are not to be construed as indicating or implying relative importance or order.

[0040] Figure 1 This is a flow chart of a method for automatically generating content provided by an embodiment of the present application. Figure 1 As shown, the content automatic generation method includes the following steps.

[0041] Step S101: Obtain hot topics through a hot topic screening module.

[0042] Step S103: Input the hot topic into multiple generative artificial intelligence models in the hot topic analysis module, thereby obtaining the content associated with the hot topic generated by each of the multiple generative artificial intelligence models, and transmitting the content associated with the hot topic generated by each of the multiple generative artificial intelligence models to the style conversion module, wherein the optimization strategy of each of the multiple generative artificial intelligence models has different optimization objectives for the model input, and the multiple generative artificial intelligence models are optimized independently of each other.

[0043] Step S105: Through the style conversion module, based on the system prompt words that at least include style prompt words, using the content associated with the hot topic generated by each of the multiple generative artificial intelligence models, stylized content for the hot topic is generated, wherein the stylized content for the hot topic has a style corresponding to the style prompt word.

[0044] Figure 1 The illustrated automatic content generation method is applied to the field of hot news information processing. For example, it can rapidly generate reports and various analytical results based on hot topics. Leveraging generative artificial intelligence (GAI) technology and artificial intelligence generated content (AIGC), it not only covers audiences and relevant information across diverse subject industries, but also provides results with varying levels of content richness and analytical perspectives, facilitating targeted responses to the needs of diverse audiences. It offers advantages such as low time cost, high execution efficiency, and the ability to capture hot topics in a timely manner and provide targeted services tailored to different customer needs. Furthermore, unlike existing content generation solutions, this method improves model resource utilization by operating multiple models in parallel. Furthermore, it can achieve rapid response and wide-ranging coverage through content reorganization as hot topics and audience needs rapidly change, thereby improving overall efficiency.

[0045] See Figure 1In step S101, a hot topic is obtained through a hot topic screening module. Here, the hot topic screening module is used to provide hot topics, which can be user-defined hot topics, such as specific international events. Hot topics can also be obtained through analysis and organization by the hot topic screening module, for example, based on market hot spots and independent hot topics, such as trade topics, or hot events. Next, in step S103, the hot topics are input into multiple generative artificial intelligence models in the hot topic analysis module, thereby obtaining content associated with the hot topics generated by each of the multiple generative artificial intelligence models. The content associated with the hot topics generated by each of the multiple generative artificial intelligence models is then transmitted to the style transfer module. The optimization strategies of each of the multiple generative artificial intelligence models have different optimization objectives for the model inputs, and the multiple generative artificial intelligence models are optimized independently of each other. This process is divided into a first stage of obtaining hot topics and a second stage of content generation. In the second stage of content generation, based on the hot topics, content associated with the hot topics generated by each of the multiple generative artificial intelligence models is obtained. It should be understood that the optimization strategies of each of the multiple generative artificial intelligence models have different optimization objectives for the model inputs, and the multiple generative artificial intelligence models are optimized independently of each other. This means that different generative AI models, for the same model input, such as the same prompt word representing a hot topic, will generate content with different focuses based on optimization strategies with different optimization objectives. By setting different optimization objectives and optimizing the multiple generative AI models independently, this ensures that the content generated by the multiple generative AI models has a greater overall content richness and can provide different analysis angles. Considering that when a single generative AI model generates content based on model input (such as a prompt word), the generated content has a certain degree of uncontrollability and instability. For example, the same generative AI model may generate different content in two content generation requests for the same prompt word. To this end, the hot topic is input into multiple generative AI models in the hot topic analysis module, thereby obtaining content related to the hot topic generated by each of the multiple generative AI models. In this way, utilizing the different optimization objectives of the optimization strategies of the multiple generative AI models helps to control the overall uncontrollability and instability of the content generated by the multiple generative AI models.In other words, compared with the content generated by a single generative AI model for a given prompt word (such as a hot event), the content generated by multiple generative AI models for the given prompt word is generally more stable and controllable overall. Moreover, by setting different optimization goals for multiple generative AI models as the optimization direction of their respective optimization strategies, the overall content coverage can also be increased, which helps avoid missing important content features or insufficient output of key content features.

[0046] Continue reading Figure 1In step S105, the style conversion module generates stylized content for the hot topic based on system prompts that include at least style prompts and the content associated with the hot topic generated by each of the multiple generative artificial intelligence models. The stylized content for the hot topic has a style corresponding to the style prompt. As described above, the process is divided into a first stage of acquiring the hot topic and a second stage of content generation. In the second stage of content generation, the hot topic is input into the multiple generative artificial intelligence models in the hot analysis module using the different optimization objectives of the optimization strategies of the multiple generative artificial intelligence models, thereby generating content associated with the hot topic generated by each of the multiple generative artificial intelligence models. This ensures that the content generated by the multiple generative artificial intelligence models has good overall controllability and stability, increases overall content coverage, and helps avoid missing important content features or insufficient output of key content features. Therefore, based on the content generation in step S103, the third stage of stylization processing is performed in step S105. That is, the third stage of stylization processing is performed after the first stage of acquiring the hot topic and the second stage of content generation. Specifically, in the third stage of stylization processing, step S105, based on system prompts that include at least style prompts, the multiple generative AI models generate stylized content related to the hot topic using content generated by the multiple generative AI models. This achieves a separate design for content generation and stylization processing. In the second stage of content generation, step S103, style prompts and styles corresponding to the style prompts are not involved. Instead, multiple generative AI models are used to generate content related to the hot topic. This helps fully leverage the design advantages of multi-model parallel processing, increasing the speed of content generation by increasing model execution parallelism and broadening overall coverage by increasing the diversity between different generative AI models. For example, the optimization objective of one generative AI model can be set to prioritize timeliness and relevance (i.e., current information and events), while the optimization objective of another generative AI model can be set to prioritize the organization of historical event context (i.e., retrospective information and historical events). Still another generative AI model can be set to prioritize trend prediction. On the basis of maximizing the richness of the generated content and expanding its coverage, the third stage of stylization processing, namely step S105, does not involve content generation, that is, no large model is called to regenerate content. Instead, the content related to the hot topic generated by the multiple generative artificial intelligence models that have been generated is used, and through stylization processing, it is transformed into stylized content for the hot topic.In this way, based on the full utilization of the model resources of multiple generative artificial intelligence models in the hot spot analysis module, by stylizing the content associated with the hot topics generated by each of the multiple generative artificial intelligence models, it is possible to support the free combination of core scene content and content style conversion, and support the generation of content of different styles to meet the personalized needs of different audience groups. It should be understood that when hot topics change rapidly, multiple generative artificial intelligence models can be used in parallel to generate content for new hot topics in a timely manner, achieving rapid response; when the needs of the audience group change rapidly, new style prompts can be used to match new personalized needs, thereby ensuring wide coverage. By using the above-mentioned separate design of content generation and stylization processing, the hot spot analysis module and the style conversion module can be selectively optimized. For example, new hot topics do not necessarily lead to new personalized needs, so new content can be generated by the hot spot analysis module based on the new hot topics while maintaining the style conversion module. For example, if new personalized needs arise for an existing hot topic, we can maintain the content generated by the hotspot analysis module and introduce new style cues to perform corresponding stylized processing, thereby generating stylized content with a new style. This helps improve overall efficiency. Furthermore, when updating and upgrading large models, the hotspot analysis module responsible for content generation and the style conversion module responsible for stylized processing can be upgraded separately, which helps improve upgrade efficiency.

[0047] In short, Figure 1The illustrated automatic content generation method utilizes separate designs for content generation and stylization, using multiple generative AI models to generate content related to the hot topics. This helps leverage the design advantages of multi-model parallel processing, increasing the speed of content generation by increasing the degree of model parallelism, and expanding overall coverage by increasing the diversity between different generative AI models. By stylizing the content related to the hot topics generated by each of the multiple generative AI models, it supports the free combination of core scenario content and content style conversion, supporting the generation of content with different styles to meet the personalized needs of different audiences. This not only covers audiences and related information within different subject industries, but also provides results with varying levels of content richness and analytical perspectives, facilitating targeted responses to the needs of different audiences. It has the advantages of low time cost, high execution efficiency, and the ability to capture hot topics in a timely manner and provide targeted services for different customer needs. Furthermore, the parallel operation of multiple models improves model resource utilization, and allows for rapid response and wide-scale coverage through content recombination as hot topics and audience needs change rapidly, thus improving overall efficiency.

[0048] Figure 2 The first embodiment of the present application provides a Figure 1 The diagram shows the business process flow of the automatic content generation method. Figure 2 The business process flow shown here refers to Figure 1The content automatic generation method shown combines public domain hot spots with internal information, research reports and other data, and uses prompt word arrangement and large model technology to generate hot spot analysis results. Specifically, starting from obtaining hot spots 201, the sequence is hot spot vectorization 202, knowledge base retrieval and sorting 203, extraction of related information 204, adding reference traceability information 205, obtaining prompt words 206, and large model analysis 207; in this way, a one-way business processing flow is formed through multiple nodes. In addition, in the process of this one-way business processing flow, there are other nodes for providing supporting functions. For example, hot spot synchronization 210 is used to support the acquisition of hot spots 201. For another example, knowledge base retrieval and sorting 203 can call the vector knowledge base 220, and the maintenance and update of the vector knowledge base 220 can be achieved through the information database 230, semantic segmentation service 231 and message queue 232. For another example, after obtaining prompt words 206, the prompt word service 241 can be used to better support the large model analysis 207. The prompt word arrangement 240 can also support the prompt word service 241. In addition, a hotspot list module can be provided to further support the acquisition of hotspots 201. For example, it can provide hotspots that aggregate specific fields and support switching between hourly, daily, and weekly rankings. In addition, each hotspot can support viewing details (such as hot trend, related information, and hot word analysis). A custom hotspot module can also be provided to support users to enter custom hotspot titles and generate analysis reports. For large model analysis 207, four content structures can be provided, such as title, self-developed large model research report analysis, market analysis, and industry analysis. Each structure supports functions such as re-retrieval, copying, and viewing historical versions. In addition, each hotspot also supports the analysis of the development of the event. In addition, a style conversion module can be provided for the final output analysis results to process the content after the analysis report is produced. To meet the personalized needs of different audience groups, it supports the generation of content with different styles to facilitate the needs of different users. In this way, the final output analysis results are achieved, and hot spot analysis reports are quickly generated for different audience groups based on market hot spots and customized hot spots, which reduces labor cost investment and content error rate, improves content richness, and meets customer demands in a targeted manner.

[0049] Figure 3 The second embodiment of the present application provides a Figure 1 The diagram shows the business process flow of the automatic content generation method. Figure 3 The business process flow shown here refers to Figure 1The content automatic generation method shown starts with logging into the content center 301, passing through hotspot screening 303 and artificial intelligence model content creation 305, and finally business scenario use 307. Among them, in the hotspot screening 303, the hotspots are first screened 310, and then the hotspot trends 311, hot word analysis 312, and related information 313 can be viewed synchronously, and finally the hotspots of interest 314 are determined. In the artificial intelligence model content creation 305, the artificial intelligence model is first called 320, and then the following can be carried out synchronously: using hotspots and market data to let the artificial intelligence model write analysis content 321, multi-module content combination 322, content style artificial intelligence model conversion 323, and finally publishing 324. In this way, based on the market hotspot capabilities of the content center, the hotspot data is vectorized and retrieved through the existing information in the library, and then the information is sliced ​​and associated and analyzed and the prompt word training is performed through the existing database, and finally the analysis results are output. In hotspot screening 303, optimization can be performed on hotspot acquisition. For example, hotspot insights first use external algorithmic capabilities to accurately capture market hotspots and related data, including analysis of popularity trends and hot words. Users can also create customized hot topics to ensure comprehensive and personalized hotspot information. Furthermore, when initializing the information knowledge base, content segmentation and vectorization techniques can be used to convert information from certain areas of the content system and implement it in the knowledge base system. When arranging system prompts, different prompts can be arranged for specific tasks, completing the configuration of information such as AI agent roles, domain knowledge, special variables, processing tasks, and response formats. Regarding vectorized matching of hotspot-related information, after hotspot acquisition, vectorization technology is used to efficiently match relevant information in the content library and intelligently sort the information based on its relevance and timeliness, ensuring that users can quickly access the most relevant and up-to-date information. When generating hotspot information through an AI model, key information from the relevant information is extracted through content slicing technology, and the prompts set in the previous step are obtained. These information is then assembled into the large model service interface parameters, and the large model service is called to complete the creation of hotspot content. Furthermore, in terms of setting AI model prompts, in some embodiments, prompt creation covers five core scenario capabilities: generating hot news headlines, self-developed large-scale model research report analysis, external model market analysis, industry analysis, and event context analysis. Prompts generate hot news headlines by using vectorized technology matching and AI large-scale model technology to generate hot and popular titles. Self-developed large-scale model prompts extract information from hot-related research reports, deeply explore the essence of the research reports, and extract key information and viewpoints. External market analysis and industry analysis prompts rely on the computing power of the large model itself and vector database information slice information to accurately predict market trends. Event context analysis prompts, by extracting relevant information and using large-scale model technology to clearly sort out the development context of the event, help users quickly understand the full picture of the event.Based on the above five core scenario capabilities, not only can the content of each scenario be regenerated, but the content of multiple scenarios can also be freely combined. By setting prompt words, the hot spot insight function also supports content style conversion, whether it is formal, humorous or concise style, it can be easily achieved to ensure that the content meets the needs of various users. This helps to achieve coverage of large-scale model research report analysis, external model market analysis, independent industry analysis, event context analysis, etc. In addition, the final generated results support style conversion, making the results more understandable to different audiences.

[0050] See Figure 1 、 Figure 2 and Figure 3 In one possible implementation, the hot topic is used to indicate the content features of the stylized content for the hot topic, the style prompt is used to indicate the style features of the stylized content for the hot topic, and the system prompt also includes multiple task prompts, the multiple task prompts correspond one-to-one to the multiple generative artificial intelligence models, and each of the multiple task prompts is used to call the model service of the generative artificial intelligence model corresponding to the task prompt in the multiple generative artificial intelligence models. As described above, by separating the content generation and stylization processing, multiple generative artificial intelligence models are used to generate content related to the hot topic. This helps to give full play to the design advantages of multi-model parallel processing, and can increase the content generation speed by increasing the parallelism of model operation. It can also expand the overall coverage by increasing the differences between different generative artificial intelligence models. By stylizing the content related to the hot topic generated by each of the multiple generative artificial intelligence models, it can support the free combination of core scene content and content style conversion, and support the generation of content of different styles to meet the personalized needs of different audience groups. Here, the hot topics are used to indicate the content features of the stylized content for the hot topics, and are therefore applicable to the second stage of content generation; the style prompts are used to indicate the stylistic features of the stylized content for the hot topics, and are therefore applicable to the third stage of stylization processing. Combined with multiple task prompts, model service calls for the generative AI models corresponding to the task prompts can be executed more efficiently, thus helping to improve the efficiency of model resource utilization.

[0051] In some embodiments, the system prompts are provided to the hotspot analysis module and the style conversion module by a prompt arrangement module. The prompt arrangement module generates the multiple task prompts based on content prompts, the content preferences corresponding to the content prompts, and the content associated with the content prompts generated by each of the multiple generative AI models. Here, the prompt arrangement module generates task prompts to assign tasks to each generative AI model. For example, assume that a generative AI model, such as a self-developed large model, extracts information, refines ideas, and generates research reports based on corresponding task prompts. Therefore, task prompts are related to content generation. As can be seen, the hotspot analysis module utilizes multiple generative AI models to generate content, and then the style conversion module performs specific style adjustments and style conversion. In other words, a standard draft is first generated, and then content is deleted and stylized. In some examples, a method for training the prompt arrangement module is to use content prompts as input training data and compare the content preferences corresponding to the content prompts with the content generated by the generative AI model. This can improve the quality of the content generated by the generative AI model. For example, a hot news model should generate content that is more consistent with the content preferences of hot news titles. Furthermore, the generation quality of task prompt words can be improved by combining debugging with specific business scenarios. Specifically, based on the hot spots selected by the user or customized hot spots, this is used as the "input hot spots", and combined with the information library, vector library, system prompt words, basic large models and other functions, the "input hot spots" are rewritten to make the hot spot information description more complete, which is convenient for the subsequent generation of standard version style content. Among them, the system prompt word arrangement and debugging link is as described above. In this way, by optimizing the design of the prompt word arrangement module for generating the multiple task prompt words, and designing the training loss function and model convergence strategy based on the content prompt words, the content preferences corresponding to the content prompt words, and the content associated with the content prompt words generated by the multiple generative artificial intelligence models, the performance of content generation can be enhanced, thereby making better use of the above-mentioned separate design of content generation and stylization processing, and using task prompt words to improve the effect of model service calls, which helps to improve the efficiency of model resource utilization.

[0052] In some embodiments, the multiple task prompts include the settings of the multiple generative AI models regarding AI agents, domain knowledge, special variables, processing tasks, and response formats. This helps to further improve the effectiveness of model service calls and improve the efficiency of model resource utilization through diversified settings.

[0053] In one possible embodiment, the style conversion module generates the stylized content for the hot topic based on the system prompt words that at least include the style prompt words and the content associated with the hot topic generated by each of the multiple generative artificial intelligence models, including: integrating the content associated with the hot topic generated by each of the multiple generative artificial intelligence models according to a standard style, thereby obtaining standardized content for the hot topic with the standard style; based on the style prompt words, selectively performing content deletion operations and selectively performing stylized element addition operations on the standardized content for the hot topic, thereby obtaining the stylized content for the hot topic with the style corresponding to the style prompt words. As described above, utilizing the different optimization objectives of the optimization strategies of each of the multiple generative artificial intelligence models helps to control the uncontrollability and instability of the content generated by the multiple generative artificial intelligence models as a whole. In other words, compared to content generated by a single generative artificial intelligence model for a given prompt word (such as a hot event), content generated by multiple generative artificial intelligence models for the given prompt word generally performs more stably and controllably. Furthermore, by setting different optimization objectives for the multiple generative artificial intelligence models as the optimization direction of their respective optimization strategies, the overall content coverage can be increased, helping to avoid missing important content features or insufficient output of key content features. Furthermore, in the second stage of content generation, the different optimization objectives of the optimization strategies of the multiple generative artificial intelligence models are utilized, and by inputting the hot topic into the multiple generative artificial intelligence models in the hot topic analysis module, content associated with the hot topic generated by each of the multiple generative artificial intelligence models is obtained. This ensures that the content generated by the multiple generative artificial intelligence models has better overall controllability and stability, increases overall content coverage, and helps avoid missing important content features or insufficient output of key content features. Here, the content associated with the hot topic generated by each of the multiple generative artificial intelligence models is integrated according to a standard style, thereby obtaining standardized content for the hot topic with the standard style. This means that the content generated by the multiple generative artificial intelligence models can be integrated in the form of a standard draft, which helps to provide standardized content for the hot topics, thereby achieving timely and efficient provision of targeted standardized content with a standard style for different hot topics, which helps with subsequent stylized processing.Moreover, because the content related to the hot topic generated by each of the multiple generative artificial intelligence models is integrated according to a standard style, standardized content for the hot topic with the standard style is obtained; therefore, the separate design of content generation and stylization processing is achieved. In the second stage of content generation, style prompts and the styles corresponding to the style prompts are not involved. Instead, multiple generative artificial intelligence models are used to generate content related to the hot topic. This helps to give full play to the design advantages of multi-model parallel processing, and can increase the content generation speed by increasing the parallelism of model operation, and can also expand the overall coverage by increasing the differences between different generative artificial intelligence models. Furthermore, on the basis of obtaining standardized content for the hot topic with the standard style, based on the style prompts, the standardized content for the hot topic is selectively deleted and stylized elements are selectively added, thereby obtaining stylized content for the hot topic with the style corresponding to the style prompts. In this way, in the third stage of stylization processing, content generation is not involved, that is, the large model is not called to regenerate content. Instead, the content related to the hot topic generated by each of the multiple generative artificial intelligence models that have been generated is used, and through stylization processing, stylized content for the hot topic is obtained. In this way, on the basis of fully utilizing the model resources of the multiple generative artificial intelligence models in the hot spot analysis module, by stylizing the content related to the hot topic generated by each of the multiple generative artificial intelligence models that have been generated, it is possible to support the free combination of core scene content and content style conversion, support the generation of content of different styles to meet the personalized needs of different audience groups. Moreover, because the content deletion operation and the stylized element addition operation are selectively performed, the overall complexity is simplified as much as possible, that is, only part of the content needs to be deleted to meet personalized needs, and the maximum coverage achieved in the content generation stage is fully utilized. Therefore, by using the content deletion operation to convert standardized content for the hot topic with the standard style into stylized content for the hot topic with the style corresponding to the style prompt, not only can the parallel operation of multiple models and the expanded overall coverage be used to increase the speed of content generation, but the deletion of content from the expanded overall coverage also increases the speed of stylization processing, thereby improving overall efficiency. Furthermore, by using the stylized element addition operation to add stylized elements, it is possible to respond to the personalized needs of different audience groups.In some examples, depending on the display platform of the final generated content, such as chat software on mobile phones, online video platforms, and any Internet media platforms, various restrictions may be placed on the overall data scale of the generated content, such as the total number of words, the total number of lines, the total data length, etc. These can be understood as part of the personalized needs. Therefore, in the stage of stylization processing, content deletion operations, that is, selective content deletion operations, can be performed to meet the content specifications on different platforms.

[0054] The following specific examples further illustrate how to integrate the content related to the hot topic generated by each of the multiple generative artificial intelligence models according to a standard style, thereby obtaining standardized content for the hot topic with the standard style, and how, based on the standardized content and based on the style prompt words, selectively delete content and selectively add stylized elements to the standardized content for the hot topic, thereby obtaining stylized content for the hot topic with the style corresponding to the style prompt words. As described above, hot topics are obtained through the hot topic screening module. Here, the hot topic screening module can provide a hot topic list, which can list multiple hot topics by hour, day, week, month, or any other sorting method, and can also allow users to enter custom hot topics. Here, hot topic insights can be generated by capturing public domain hot topics in real time and combining big data and artificial intelligence technology. For example, the hot topic list can list hot topics such as "Changes in container transportation volume between two countries after a certain event" and hot topics such as the start of cross-border railway construction and the opening of international road transport routes. Here, taking "Changes in container shipping volume between two countries after a certain event" as an exemplary hot topic, this hot topic is input into multiple generative AI models in the hot topic analysis module, thereby obtaining content associated with the hot topic generated by each of the multiple generative AI models. Here, each of the multiple generative AI models has a different optimization objective for the model input, and the multiple generative AI models are optimized independently of each other. For example, one generative AI model has an optimization objective for the relevance and timeliness of the model input. Therefore, for the hot topic "Changes in container shipping volume between two countries after a certain event," it can provide content such as export demand and capacity adjustments, such as a month-on-month increase in capacity and freight rates. Another generative AI model has an optimization objective for trend prediction of the model input. Therefore, for the hot topic "Changes in container shipping volume between two countries after a certain event," it can provide content such as future outlook, such as a forecast of high container shipping bookings. The content associated with the hot topic generated by each of the multiple generative AI models in this manner, based on different optimization objectives, helps increase the richness and coverage of the generated content. Next, the content related to the hot topic generated by each of the multiple generative AI models is integrated according to a standard style, resulting in standardized content for the hot topic, namely, "Changes in container shipping volume between two countries after a certain event." Next, based on style cues, the standardized content for the hot topic is selectively deleted and stylized elements are selectively added.For example, a standard style can be adopted, which means retaining the original standardized content; another example is an information style that highlights timeliness; another example is a consulting style that provides in-depth analysis and thinking; another example is a relaxed style that adds more fashionable stylized elements to increase readers' reading pleasure; another example is a concise style that streamlines content and compresses data specifications to suit readers on mobile terminals. Therefore, based on standardized content, the standardized content for the hot topic can be selectively deleted and stylized elements can be selectively added. This can support the free combination of core scene content and content style conversion, and support the generation of content of different styles to meet the personalized needs of different audience groups. Moreover, when hot topics change rapidly, for example, the current hot topic, "Changes in container shipping volume between two countries after a certain event," becomes "Changes in container shipping volume between two countries after a certain event and another sudden incident" due to a sudden international event, such new hot topics can be processed in parallel by multiple generative artificial intelligence models to generate content for the new hot topics in a timely manner, achieving rapid response; when the needs of the audience group change rapidly, new style prompts can be used to match new personalized needs, thereby ensuring wide coverage.

[0055] In some embodiments, the standard style is independent of the style corresponding to the style prompt word, and the standardized content for the hot topic with the standard style is independent of the style prompt word. In this way, the separate design of content generation and stylization processing is achieved, and the largest possible coverage achieved in the content generation stage is fully utilized. The standardized content for the hot topic with the standard style is converted into the stylized content for the hot topic with the style corresponding to the style prompt word by using the content deletion operation. This not only improves the content generation speed by using the parallel operation of multiple models and the expanded overall coverage, but also improves the stylization processing speed by using the deletion of content in the expanded overall coverage, which is conducive to improving overall efficiency. In addition, the stylization element addition operation is used to add stylization elements, which is conducive to responding to the personalized needs of different audience groups.

[0056] In some embodiments, the stylized element addition operation includes adding at least one stylized element of a style corresponding to the style prompt word, wherein the at least one stylized element is selected from a plurality of stylized elements, wherein the plurality of stylized elements includes symbols, emoticons, and images. In this way, by adding stylized elements, it is possible to respond to the personalized needs of different audience groups.

[0057] In one possible implementation, the multiple generative artificial intelligence models provide the content associated with the hot topics generated by each of the multiple generative artificial intelligence models through multi-model parallel operation, and the updates of the multiple generative artificial intelligence models are performed independently of each other. In this way, by using multiple generative artificial intelligence models to generate content associated with the hot topics, it helps to give full play to the design advantages of multi-model parallel processing, and can increase the speed of content generation by increasing the parallelism of model operation, and can also expand the overall coverage by increasing the differences between different generative artificial intelligence models.

[0058] In some embodiments, the system prompt also includes a weight ratio of each of the multiple generative artificial intelligence models, and the weight ratio of each of the multiple generative artificial intelligence models is used to indicate the importance of the content associated with the hot topic generated by each of the multiple generative artificial intelligence models relative to the stylized content for the hot topic. In addition, when there is an updated generative artificial intelligence model among the multiple generative artificial intelligence models and the weight ratio of the updated generative artificial intelligence model is not higher than a preset threshold, the old model of the updated generative artificial intelligence model is used to ensure that the multi-model parallel operation of the multiple generative artificial intelligence models is not interrupted. As described above, the content generation and stylization processing are divided into two stages, so that the generative artificial intelligence model can be updated without interrupting business. For example, if the system prompt indicates that the main purpose is to provide hot information on a certain hot topic issue, then the weight ratio of the generative artificial intelligence model responsible for generating the event context may be lower at this time, which means that the generative artificial intelligence model of the event context can be updated while the analysis report is generated, and the use of the old model has little impact on the quality of the report. In this way, through the separate design of content generation and stylization processing, multiple generative artificial intelligence models are used to generate content related to the hot topics. This helps to give full play to the design advantages of multi-model parallel processing. The content generation speed can be improved by increasing the parallelism of model operation. The overall coverage can also be expanded by increasing the differences between different generative artificial intelligence models. By stylizing the content related to the hot topics generated by the multiple generative artificial intelligence models that have been generated, it can support the free combination of core scene content and content style conversion, and support the generation of content of different styles to meet the personalized needs of different audience groups. Furthermore, in the face of the situation where model updates may occur from time to time, the design characteristics of the multiple generative artificial intelligence models being optimized independently of each other are utilized. The old models of the models being updated with relatively low importance can be used for multi-model parallel operation, thereby ensuring uninterrupted business and limited impact on the quality of the final generated content, which is conducive to meeting business continuity and stability.

[0059] In one possible embodiment, the style conversion module generates the stylized content for the hot topic based on the system prompt word including at least the style prompt word and the content associated with the hot topic generated by each of the multiple generative artificial intelligence models, including: integrating the content associated with the hot topic generated by each of the multiple generative artificial intelligence models according to a plurality of preset standard styles, thereby obtaining a plurality of standardized content for the hot topic, each of the plurality of standardized content for the hot topic having the plurality of preset standard styles; based on the style prompt word, selecting a preset standard style that matches the style prompt word from the plurality of preset standard styles, and determining the standardized content for the hot topic with the selected preset standard style among the plurality of standardized content for the hot topic as a reference standardized content; based on the style prompt word, selectively performing a content deletion operation and a stylized element addition operation on the reference standardized content, thereby obtaining the stylized content for the hot topic having the style corresponding to the style prompt word. Here, by utilizing the separate design of content generation and stylization processing, and taking advantage of multi-model parallel processing, we can generate analysis reports for multiple key styles, and then select matching key styles as a reference based on the user's style preferences, and then further perform user customization. In this way, multiple batches of generated content can be generated for key hot topics and key styles, and then screened based on specific hot topics, which helps to improve overall efficiency and meet the personalized needs of different audiences. In addition, by performing stylization processing on the basis of determining reference standardized content, we can use a combination of core scene content to better and faster generate content that meets user customization needs.

[0060] In some embodiments, the plurality of preset standard styles are key styles determined based on user demand predictions, thereby helping to improve overall efficiency and meet the personalized needs of different audience groups.

[0061] In some embodiments, the hot topics are key hot topics determined based on the user demand prediction, which helps to improve overall efficiency and meet the personalized needs of different audience groups.

[0062] In one possible implementation, the style prompts are selected from a variety of preset style prompts, including standard style, consulting style, advisory style, relaxed style, and concise style. In some examples, depending on the platform where the final content is displayed, such as mobile chat apps, online video platforms, or any internet media platform, various restrictions may be placed on the overall data size of the generated content, such as the total number of words, total number of lines, and total data length. These restrictions can be understood as part of personalized requirements. Therefore, during the stylization process, content reduction operations, i.e., selective content deletion operations, can be performed to meet the content standards of different platforms. Here, different preset style prompts can take into account the content standards of different platforms, such as one preset style prompt corresponding to the content standards of one platform. For example, a consulting style can correspond to the content standards of platforms with a strong timeliness, such as short news publishing platforms. Another example is that a consulting style can correspond to the content standards of platforms with a strong professional focus, such as those requiring in-depth information analysis and a complete chronological analysis of events. Another example is that a relaxed style can correspond to the content standards of platforms with a strong entertainment focus, such as online video platforms and live streaming platforms. For example, a concise style can be used for mobile platforms, such as chat apps. Because the display area of ​​these platforms is limited, content generated must be as concise as possible to facilitate presentation to end users. Providing a variety of preset style prompts can help meet the personalized needs of different audiences.

[0063] In one possible embodiment, the multiple generative artificial intelligence models include a first generative artificial intelligence model, a second generative artificial intelligence model, a third generative artificial intelligence model, a fourth generative artificial intelligence model, and a fifth generative artificial intelligence model, wherein the optimization strategy of the first generative artificial intelligence model has an optimization goal for the relevance and timeliness of the model input, the optimization strategy of the second generative artificial intelligence model has an optimization goal for information extraction and in-depth research of the model input, the third generative artificial intelligence model has an optimization goal for trend prediction of the model input, the optimization strategy of the fourth generative artificial intelligence model has an optimization goal for the multi-industry analysis results of the model input, and the optimization strategy of the fifth generative artificial intelligence model has an optimization goal for the event context analysis results of the model input. The optimization strategies of the multiple generative artificial intelligence models each have different optimization goals for the model input, and the multiple generative artificial intelligence models are optimized independently of each other. This means that different generative artificial intelligence models, for the same model input, such as the same prompt word representing a hot topic, based on optimization strategies with different optimization goals, generate content with different focuses. By setting different optimization goals and optimizing the multiple generative artificial intelligence models independently of each other, it is possible to ensure that the content generated by the multiple generative artificial intelligence models has a better overall content richness and can provide different analysis angles. Considering that when a single generative artificial intelligence model generates content based on model input (such as a prompt word), the content it generates has a certain degree of uncontrollability and instability. For example, the same generative artificial intelligence model may generate content that is somewhat different for two content generation requests for the same prompt word. To this end, the hot topic is input into the multiple generative artificial intelligence models in the hot topic analysis module, thereby obtaining the content associated with the hot topic generated by each of the multiple generative artificial intelligence models. In this way, the different optimization goals of the optimization strategies of the multiple generative artificial intelligence models are used to help control the overall uncontrollability and instability of the content generated by the multiple generative artificial intelligence models. In other words, compared with the content generated by a single generative AI model for a given prompt word (such as hot events in two countries), the content generated by multiple generative AI models for the given prompt word is generally more stable and controllable overall. Moreover, by setting different optimization goals for multiple generative AI models as the optimization direction of their respective optimization strategies, the overall content coverage can also be increased, which helps to avoid missing important content features or insufficient output of key content features.In this way, by providing the above-mentioned first generative artificial intelligence model, second generative artificial intelligence model, third generative artificial intelligence model, fourth generative artificial intelligence model and fifth generative artificial intelligence model, and utilizing the optimization strategies of each of these generative artificial intelligence models and their different optimization goals, it is helpful to give full play to the design advantages of multi-model parallel processing, and the content generation speed can be improved by increasing the parallelism of model operation, and the overall coverage can be expanded by increasing the differences between different generative artificial intelligence models.

[0064] In some embodiments, the stylized content for the hot topic is a hot analysis report for the hot topic, the content associated with the hot topic generated by the first generative artificial intelligence model is the hot information scenario content associated with the hot topic, the content associated with the hot topic generated by the second generative artificial intelligence model is the research report analysis scenario content associated with the hot topic, the content associated with the hot topic generated by the third generative artificial intelligence model is the external market analysis scenario content associated with the hot topic, the content associated with the hot topic generated by the fourth generative artificial intelligence model is the multi-industry analysis scenario content associated with the hot topic, and the content associated with the hot topic generated by the fifth generative artificial intelligence model is the event development context scenario content associated with the hot topic. In this way, through the separate design of content generation and stylization processing, multiple generative artificial intelligence models are used to generate content related to the hot topics. This helps to fully utilize the design advantages of multi-model parallel processing. It can increase the speed of content generation by increasing the parallelism of model operations, and can also expand the overall coverage by increasing the differences between different generative artificial intelligence models. By stylizing the content related to the hot topics generated by the multiple generative artificial intelligence models, it can support the free combination of core scene content and content style conversion, supporting the generation of content with different styles to meet the personalized needs of different audience groups. In this way, not only can audience groups and related information within the scope of different subject industries be covered, but also results with different content richness and different analytical perspectives can be provided, which is conducive to targeted meeting the needs of different audience groups. It has the advantages of low time cost, high execution efficiency, timely capture of hot spots, and providing targeted services for different customer needs. In addition, through the parallel operation of multiple models, the utilization rate of model resources is improved. With the rapid changes in hot topics and the rapid changes in audience needs, through the re-combination of content, rapid response and wide coverage can be achieved, which is conducive to improving overall efficiency.

[0065] In some embodiments, the style prompts are used to set the combination between the scene content generated by each of the multiple generative artificial intelligence models. In this way, through the separate design of content generation and stylization processing, multiple generative artificial intelligence models are used to generate content related to the hot topics. This helps to give full play to the design advantages of multi-model parallel processing, and can increase the speed of content generation by increasing the parallelism of model operation. It can also expand the overall coverage by increasing the differences between different generative artificial intelligence models. By stylizing the content related to the hot topics generated by each of the multiple generative artificial intelligence models, it can support the free combination of core scene content and content style conversion, and support the generation of content of different styles to meet the personalized needs of different audience groups. In this way, it can not only cover audience groups and related information within the scope of different subject industries, but also provide results with different content richness and different analysis angles, which is conducive to targeted satisfaction of the needs of different audience groups, with low time cost, high execution efficiency, and the ability to capture hot spots in a timely manner and provide targeted services for different customer needs. In addition, the parallel operation of multiple models improves the utilization of model resources. As hot topics and the needs of the audience change rapidly, content can be reorganized to achieve rapid response and wide coverage, which is conducive to improving overall efficiency.

[0066] In some embodiments, the hot information scene content associated with the hot topic is based on the hot information information associated with the hot topic after vectorization processing, the event development context scene content associated with the hot topic is the event development context output result obtained by sorting the hot topic according to the timeline associated with the hot topic through a large language model, and the external market analysis scene content associated with the hot topic and the multi-industry analysis scene content associated with the hot topic are obtained by performing hot spot analysis through artificial intelligence agents and analysis frameworks. In this way, through the separate design of content generation and stylization processing, multiple generative artificial intelligence models are used to generate content associated with the hot topic, which helps to give full play to the design advantages of multi-model parallel processing, and can increase the content generation speed by increasing the parallelism of model operation, and can also expand the overall coverage by increasing the differences between different generative artificial intelligence models. By stylizing the content associated with the hot topic generated by each of the multiple generative artificial intelligence models, it can support the free combination of core scene content and content style conversion, and support the generation of content of different styles to meet the personalized needs of different audience groups. This not only covers audiences and relevant information across diverse themes and industries, but also provides results with varying levels of content richness and analytical perspectives, facilitating targeted responses to the needs of diverse audiences. This approach offers advantages such as low time costs, high execution efficiency, and the ability to capture hot topics and provide targeted services tailored to specific customer needs. Furthermore, by running multiple models in parallel, model resource utilization is improved. Furthermore, by recombining content to respond quickly and achieve wide-ranging coverage in response to the rapid changes in hot topics and audience needs, overall efficiency is enhanced.

[0067] In one possible embodiment, the automatic content generation method further includes: generating, by the style conversion module, another stylized content for the hot topic based on another style prompt word different from the style prompt word, using the content associated with the hot topic generated by each of the multiple generative artificial intelligence models, wherein the other stylized content for the hot topic has a style corresponding to the other style prompt word. In this way, by separately designing content generation and stylization processing, multiple generative artificial intelligence models are used to generate content associated with the hot topic, which helps to give full play to the design advantages of multi-model parallel processing, and can increase the content generation speed by increasing the parallelism of model operation, and can also expand the overall coverage by increasing the differences between different generative artificial intelligence models. By stylizing the content associated with the hot topic generated by each of the multiple generative artificial intelligence models, it is possible to support the free combination of core scene content and content style conversion, and support the generation of content of different styles to meet the personalized needs of different audience groups. This not only covers audiences and relevant information across diverse themes and industries, but also provides results with varying levels of content richness and analytical perspectives, facilitating targeted responses to the needs of diverse audiences. This approach offers advantages such as low time costs, high execution efficiency, and the ability to capture hot topics and provide targeted services tailored to specific customer needs. Furthermore, by running multiple models in parallel, model resource utilization is improved. Furthermore, by recombining content to respond quickly and achieve wide-ranging coverage in response to the rapid changes in hot topics and audience needs, overall efficiency is enhanced.

[0068] In one possible embodiment, the hot topics are user-defined, or the hot topics are obtained by processing public domain hot information data through the hot topic screening module based on a hot spot capture algorithm, a hot spot trend analysis algorithm, and a hot word analysis algorithm. In this way, by separating the content generation and stylization processing, multiple generative artificial intelligence models are used to generate content related to the hot topics. This helps to fully utilize the design advantages of multi-model parallel processing, and can increase the speed of content generation by increasing the parallelism of model operations. It can also expand the overall coverage by increasing the differences between different generative artificial intelligence models. By stylizing the content related to the hot topics generated by each of the multiple generative artificial intelligence models, it can support the free combination of core scene content and content style conversion, and support the generation of content with different styles to meet the personalized needs of different audience groups. In this way, it can not only cover audience groups and related information within the scope of different subject industries, but also provide results with different content richness and different analysis angles, which is conducive to targeted satisfaction of the needs of different audience groups. It has the advantages of low time cost, high execution efficiency, timely capture of hot spots, and providing targeted services for different customer needs. In addition, the parallel operation of multiple models improves the utilization of model resources. As hot topics and the needs of the audience change rapidly, content can be reorganized to achieve rapid response and wide coverage, which is conducive to improving overall efficiency.

[0069] Figure 4 This is a schematic diagram of a content automatic generation system provided in an embodiment of the present application. Figure 4As shown, the automatic content generation system 400 includes a hotspot screening module 401, a hotspot analysis module 403, and a style transfer module 405. The hotspot screening module 401 is used to obtain hot topics. The hotspot analysis module 403 includes multiple generative artificial intelligence models. The hotspot analysis module 403 is used to input the hot topics into the multiple generative artificial intelligence models, thereby obtaining content associated with the hot topics generated by each of the multiple generative artificial intelligence models, and transmit the content associated with the hot topics generated by each of the multiple generative artificial intelligence models to the style transfer module 405. The optimization strategies of the multiple generative artificial intelligence models each have different optimization objectives for the model input, and the multiple generative artificial intelligence models are optimized independently of each other. The style transfer module 405 is used to generate stylized content for the hot topics based on system prompts that include at least style prompts, using the content associated with the hot topics generated by each of the multiple generative artificial intelligence models, wherein the stylized content for the hot topics has a style corresponding to the style prompts. Figure 4 The figure exemplarily shows that the multiple generative artificial intelligence models in the hotspot analysis module 403 include a first generative artificial intelligence model 411, a second generative artificial intelligence model 412, a third generative artificial intelligence model 413, a fourth generative artificial intelligence model 414 and a fifth generative artificial intelligence model 415.

[0070] Figure 4The illustrated automatic content generation system utilizes separate designs for content generation and stylization, using multiple generative AI models to independently generate content related to the hot topics. This helps leverage the design advantages of multi-model parallel processing, increasing the speed of content generation by increasing the degree of model parallelism, and broadening overall coverage by increasing the diversity between different generative AI models. By stylizing the content related to the hot topics generated by each of the multiple generative AI models, it supports the free combination of core scenario content and content style conversion, enabling the generation of content with different styles to meet the personalized needs of different audiences. This not only covers audiences and related information within different subject industries, but also provides results with varying levels of content richness and analytical perspectives, facilitating targeted responses to the needs of different audiences. It offers advantages such as low time cost, high execution efficiency, and the ability to capture hot topics in a timely manner and provide targeted services for different customer needs. Furthermore, the parallel operation of multiple models improves model resource utilization, and allows for rapid response and wide-scale coverage through content recombination as hot topics and audience needs change rapidly, thus improving overall efficiency.

[0071] See Figure 4In one possible implementation, the hot topic is used to indicate the content features of the stylized content for the hot topic, the style prompt is used to indicate the style features of the stylized content for the hot topic, and the system prompt also includes multiple task prompts, the multiple task prompts correspond one-to-one to the multiple generative artificial intelligence models, and each of the multiple task prompts is used to call the model service of the generative artificial intelligence model corresponding to the task prompt in the multiple generative artificial intelligence models. As described above, by separating the content generation and stylization processing, multiple generative artificial intelligence models are used to generate content related to the hot topic. This helps to give full play to the design advantages of multi-model parallel processing, and can increase the content generation speed by increasing the parallelism of model operation. It can also expand the overall coverage by increasing the differences between different generative artificial intelligence models. By stylizing the content related to the hot topic generated by each of the multiple generative artificial intelligence models, it can support the free combination of core scene content and content style conversion, and support the generation of content of different styles to meet the personalized needs of different audience groups. Here, the hot topics are used to indicate the content features of the stylized content for the hot topics, and are therefore applicable to the second stage of content generation; the style prompts are used to indicate the stylistic features of the stylized content for the hot topics, and are therefore applicable to the third stage of stylization processing. Combined with multiple task prompts, model service calls for the generative AI models corresponding to the task prompts can be executed more efficiently, thus helping to improve the efficiency of model resource utilization.

[0072] In one possible embodiment, the style conversion module 405 is used to: generate the stylized content for the hot topic based on the system prompt words that at least include the style prompt words, using the content associated with the hot topic generated by each of the multiple generative artificial intelligence models, including: integrating the content associated with the hot topic generated by each of the multiple generative artificial intelligence models according to a standard style, thereby obtaining standardized content for the hot topic with the standard style; based on the style prompt words, selectively deleting content and selectively adding stylized elements to the standardized content for the hot topic, thereby obtaining stylized content for the hot topic with a style corresponding to the style prompt words. In this way, the separate design of content generation and stylization processing is achieved, making full use of the maximum possible coverage achieved in the content generation stage. Using content deletion to convert standardized content for the hot topic with the standard style into stylized content for the hot topic with the style corresponding to the style prompt, this not only increases content generation speed by leveraging the parallel operation of multiple models and expanded overall coverage, but also increases stylization processing speed by deleting content from the expanded overall coverage, thereby improving overall efficiency. Furthermore, using the stylized element addition operation to add stylized elements facilitates responding to the personalized needs of different audience groups.

[0073] In one possible embodiment, the multiple generative artificial intelligence models include a first generative artificial intelligence model 411, a second generative artificial intelligence model 412, a third generative artificial intelligence model 413, a fourth generative artificial intelligence model 414, and a fifth generative artificial intelligence model 415, wherein the optimization strategy of the first generative artificial intelligence model 411 has an optimization goal for the relevance and timeliness of the model input, the optimization strategy of the second generative artificial intelligence model 412 has an optimization goal for information extraction and in-depth research of the model input, the third generative artificial intelligence model 413 has an optimization goal for trend prediction of the model input, the optimization strategy of the fourth generative artificial intelligence model 414 has an optimization goal for multi-industry analysis results of the model input, and the optimization strategy of the fifth generative artificial intelligence model 415 has an optimization goal for event context analysis results of the model input. In this way, utilizing the respective optimization strategies of these generative artificial intelligence models and their different optimization goals helps to give full play to the design advantages of multi-model parallel processing, can increase the content generation speed by increasing the parallelism of model operation, and can also expand the overall coverage by increasing the differences between different generative artificial intelligence models.

[0074] See Figure 1、 Figure 2 、 Figure 3 and Figure 4In some embodiments, the content automatic generation method, computer equipment, medium and system provided by the specific embodiments and implementation methods of this application include: providing a system function for generating analysis reports based on hot spots; providing a system function for generating the time context of hot events with the help of artificial intelligence big models for hot events; providing a system function for generating different content styles for different hot events and different user groups; the analysis report provided generally includes multiple modules, including but not limited to: relevant industry research reports, market analysis, industry analysis, event context sorting and content style changes. In some embodiments, the content automatic generation method, computer equipment, medium and system provided by this application can be applied to the processing of hot news information in the financial field, and provide a content generation solution that integrates public domain hot spots with internal information, research reports and other data into the event context of the financial field, and combines prompt word arrangement and big model technology to generate hot spot analysis. The content generation solution includes the following steps. (1) Combining public domain hot spots with internal information, research reports and other data, and combining prompt word arrangement and big model technology to generate hot spot analysis, which can specifically include: writing financial information segmentation and vectorization methods to build a financial field information knowledge base; writing system prompt word arrangement code to realize the setting of artificial intelligence model agent roles, field common sense, special variables, processing tasks, response formats and other information; writing public domain hot spot acquisition methods and text vectorization methods; writing methods for performing knowledge retrieval in the financial field information knowledge base based on public domain hot spot vectors to obtain relevant information data; writing methods for filtering and sorting relevant information to extract target knowledge; writing adapter codes for different big model service calls; writing methods for obtaining system prompt words for hot spot analysis tasks; and writing methods for docking big model adapters based on associated knowledge and relevant prompt words. (2) Integrate public domain hot spots with internal information, research reports and other data in combination with big model technology into financial field event context, which can specifically include: writing financial information segmentation and vectorization methods to build a financial field information knowledge base; writing system prompt word arrangement code to realize the setting of artificial intelligence model agent roles, domain common sense, special variables, processing tasks, response formats and other information; writing public domain hot spot acquisition methods and text vectorization methods; writing methods for performing knowledge retrieval in the financial field information knowledge base based on public domain hot spot vectors to obtain relevant information data; writing methods for filtering and sorting relevant information data to extract target knowledge; writing adapter codes for different big model service calls; writing system prompt words for obtaining event context tasks; and writing methods for connecting to big model adapters based on associated knowledge and relevant prompt words.This content generation solution offers the following advantages: It uses large-scale modeling and vectorization technologies to match news, research reports, and information corresponding to relevant hot topics; by building an intelligent analytical agent and inputting a unique analytical framework and relevant data, it can analyze various hot topics and output exclusive results such as market analysis and industry analysis; customizable styles: By training style conversion prompt words, it optimizes the output of diversified styles and styles set by users to meet the needs of audiences with different content styles; and automatically generates a timeline-based event context: Through a large language model, it understands relevant data and outputs results in a timeline-organized event context. Furthermore, this content generation solution quickly screens hot topics through hotspot aggregation and display, accurately outputs hotspot analysis through large-scale model prompt word training, and quickly retrieves relevant information through vectorized content library matching, reducing manual time costs.

[0075] Figure 5 : is a structural diagram of a computing device provided in an embodiment of the present application, and the computing device 500 includes: one or more processors 510, a communication interface 520 and a memory 530. The processor 510, the communication interface 520 and the memory 530 are interconnected via a bus 540. Optionally, the computing device 500 may also include an input / output interface 550, and the input / output interface 550 is connected to an input / output device for receiving parameters set by the user, etc. The computing device 500 can be used to implement part or all of the functions of the device embodiment or system embodiment in the above-mentioned embodiment of the present application; the processor 510 can also be used to implement part or all of the operating steps of the method embodiment in the above-mentioned embodiment of the present application. For example, the specific implementation of the various operations performed by the computing device 500 can refer to the specific details in the above-mentioned embodiments, such as the processor 510 is used to execute part or all of the steps in the above-mentioned method embodiment or part or all of the operations in the above-mentioned method embodiment. For another example, in an embodiment of the present application, the computing device 500 may be used to implement part or all of the functions of one or more components in the above-mentioned device embodiment. In addition, the communication interface 520 may be specifically used to perform the communication functions necessary to implement the functions of these devices and components, and the processor 510 may be specifically used to perform the processing functions necessary to implement the functions of these devices and components.

[0076] It should be understood that Figure 5The computing device 500 may include one or more processors 510, and the multiple processors 510 may be connected in parallel, serially, serially, or in parallel with each other, or in any other manner to collaboratively provide processing capabilities. Alternatively, the multiple processors 510 may form a processor sequence or a processor array. Alternatively, the multiple processors 510 may be divided into a main processor and an auxiliary processor. Alternatively, the multiple processors 510 may have different architectures, such as a heterogeneous computing architecture. Figure 5 The computing device 500 shown, and the related structural and functional descriptions are exemplary and non-limiting. In some exemplary embodiments, the computing device 500 may include Figure 5 More or fewer components may be shown, some components may be combined, some components may be separated, or there may be a different arrangement of components.

[0077] The processor 510 can have various specific implementation forms. For example, the processor 510 can include a central processing unit (CPU), a graphics processing unit (GPU), a neural-network processing unit (NPU), a tensor processing unit (TPU), or a data processing unit (DPU), among others, and this is not specifically limited in the embodiments of the present application. The processor 510 can also be a single-core processor or a multi-core processor. The processor 510 can be a combination of a CPU and a hardware chip. The hardware chip can be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The PLD can be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof. The processor 510 can also be implemented solely using a logic device with built-in processing logic, such as an FPGA or a digital signal processor (DSP). The communication interface 520 may be a wired interface or a wireless interface for communicating with other modules or devices. The wired interface may be an Ethernet interface, a local interconnect network (LIN), etc. The wireless interface may be a cellular network interface or a wireless local area network interface, etc.

[0078] Memory 530 may be a non-volatile memory, such as read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Memory 530 may also be a volatile memory, which may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM). The memory 530 may also be used to store program codes and data, so that the processor 510 can call the program codes stored in the memory 530 to execute part or all of the operation steps in the above method embodiment, or to execute the corresponding functions in the above device embodiment. Figure 5 Show more or fewer components, or configure components differently.

[0079] The bus 540 may be a Peripheral Component Interconnect Express (PCIe) bus, an Extended Industry Standard Architecture (EISA) bus, a Unified Bus (UBus or UB), a Compute Express Link (CXL), a Cache Coherent Interconnect for Accelerators (CCIX), etc. The bus 540 may be divided into an address bus, a data bus, a control bus, etc. In addition to the data bus, the bus 540 may also include a power bus, a control bus, and a status signal bus. However, for the sake of clarity, Figure 5 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0080] The methods and devices provided in the embodiments of the present application are based on the same inventive concept. Since the principles of the methods and devices for solving problems are similar, the embodiments, implementation methods, examples or implementation methods of the methods and devices can refer to each other, and the repeated parts will not be repeated. The embodiments of the present application also provide a system, which includes multiple computing devices, and the structure of each computing device can refer to the structure of the computing device described above. The functions or operations that can be implemented by the system can refer to the specific implementation steps in the above method embodiments and / or the specific functions described in the above device embodiments, and will not be repeated here.

[0081] The present application also provides a computer-readable storage medium storing computer instructions. When the computer instructions are executed on a computer device (e.g., one or more processors), the method steps described in the above method embodiments can be implemented. The specific implementation of the above method steps by the processor of the computer-readable storage medium can refer to the specific operations described in the above method embodiments and / or the specific functions described in the above device embodiments, and will not be further described here.

[0082] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. The present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. The embodiments of the present application may be implemented in whole or in part via software, hardware, firmware, or any other combination. When implemented using software, the above embodiments may be implemented in whole or in part as a computer program product. The present application may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The computer program product comprises one or more computer instructions. When loaded or executed on a computer, the computer program instructions fully or partially perform the processes or functions described in the embodiments of the present application. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. Computer-readable storage media can be any available medium that can be accessed by a computer, or a data storage device such as a server or data center that contains a collection of one or more available media. Available media can be magnetic media (such as floppy disks, hard disks, or magnetic tape), optical media, or semiconductor media. Semiconductor media can be solid-state drives, random access memory, flash memory, read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, registers, or any other suitable storage medium.

[0083] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. Each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including the instruction device, which implements the function specified in the process. Figure 1a process or multiple processes and / or boxes Figure 1 These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0084] In the above embodiments, the descriptions of each embodiment have different emphases. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments. Obviously, those skilled in the art can make various changes and modifications to the embodiments of the present application without departing from the spirit and scope of the embodiments of the present application. The steps in the method of the embodiment of the present application can be adjusted in sequence, merged or deleted according to actual needs; the modules in the system of the embodiment of the present application can be divided, merged or deleted according to actual needs. If these modifications and variations of the embodiments of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application is also intended to include these modifications and variations.

Claims

1. A method for automatically generating content, characterized in that: The content automatic generation method includes: Obtain hot topics through the hot spot screening module; Inputting the hot topic into multiple generative artificial intelligence models in a hot topic analysis module, thereby obtaining content associated with the hot topic generated by each of the multiple generative artificial intelligence models, and transmitting the content associated with the hot topic generated by each of the multiple generative artificial intelligence models to a style transfer module, wherein each optimization strategy of the multiple generative artificial intelligence models has a different optimization objective for the model input, and the multiple generative artificial intelligence models are optimized independently of each other; Generating stylized content for the hot topic by the style conversion module based on system prompt words including at least style prompt words and using the content associated with the hot topic generated by each of the multiple generative artificial intelligence models, including: Integrating the content related to the hot topic generated by each of the multiple generative artificial intelligence models according to a standard style, thereby obtaining standardized content for the hot topic in the standard style; Based on the style prompt word, selectively performing a content deletion operation and a stylized element addition operation on the standardized content for the hot topic, thereby obtaining the stylized content for the hot topic having a style corresponding to the style prompt word, The stylized content for the hot topic has a style corresponding to the style prompt word, the standard style is independent of the style corresponding to the style prompt word, and the standardized content for the hot topic with the standard style is independent of the style prompt word.

2. The content automatic generation method according to claim 1, characterized in that: The hot topic is used to indicate the content characteristics of the stylized content for the hot topic, and the style prompt word is used to indicate the style characteristics of the stylized content for the hot topic. The system prompt word also includes multiple task prompt words, and the multiple task prompt words correspond one-to-one to the multiple generative artificial intelligence models. Each of the multiple task prompt words is used to call the model service of the generative artificial intelligence model corresponding to the task prompt word among the multiple generative artificial intelligence models.

3. The content automatic generation method according to claim 2, characterized in that: The system prompt words are provided to the hotspot analysis module and the style conversion module by the prompt word arrangement module. The function of the prompt word arrangement module for generating the multiple task prompt words is based on the content prompt words, the content preferences corresponding to the content prompt words, and the content associated with the content prompt words generated by each of the multiple generative artificial intelligence models.

4. The content automatic generation method according to claim 2, characterized in that: The multiple task prompts include settings for the multiple generative artificial intelligence models regarding artificial intelligence agents, domain common sense, special variables, processing tasks, and response formats.

5. The content automatic generation method according to claim 1, characterized in that: The stylized element adding operation includes adding at least one stylized element of a style corresponding to the style prompt word, wherein the at least one stylized element is selected from a plurality of stylized elements, and the plurality of stylized elements includes symbols, emoticons, and pictures.

6. The content automatic generation method according to claim 1, characterized in that: The multiple generative artificial intelligence models provide content associated with the hot topics generated by each of the multiple generative artificial intelligence models through multi-model parallel operation, and the updates of the multiple generative artificial intelligence models are performed independently of each other.

7. The content automatic generation method according to claim 6, characterized in that: The system prompt also includes the weight ratio of each of the multiple generative artificial intelligence models, and the weight ratio of each of the multiple generative artificial intelligence models is used to indicate the importance of the content associated with the hot topic generated by each of the multiple generative artificial intelligence models relative to the stylized content for the hot topic. In addition, when there is an updated generative artificial intelligence model among the multiple generative artificial intelligence models and the weight ratio of the updated generative artificial intelligence model is not higher than a preset threshold, the old model of the updated generative artificial intelligence model is used to ensure that the multi-model parallel operation of the multiple generative artificial intelligence models is not interrupted.

8. The content automatic generation method according to claim 1, characterized in that: Generating, by the style conversion module, the stylized content for the hot topic based on the system prompt word including at least the style prompt word and using the content associated with the hot topic generated by each of the multiple generative artificial intelligence models, includes: Integrating the content associated with the hot topics generated by each of the multiple generative artificial intelligence models according to multiple preset standard styles, thereby obtaining multiple standardized contents for the hot topics, each of the multiple standardized contents for the hot topics having the multiple preset standard styles; Based on the style prompt word, selecting a preset standard style that matches the style prompt word from the plurality of preset standard styles, and determining, among the plurality of standardized contents for the hot topic, standardized content for the hot topic having the selected preset standard style as reference standardized content; Based on the style prompt word, a content deletion operation and a stylized element addition operation are selectively performed on the reference standardized content, thereby obtaining the stylized content for the hot topic with a style corresponding to the style prompt word.

9. The content automatic generation method according to claim 8, characterized in that: The plurality of preset standard styles are key styles determined based on user demand prediction.

10. The content automatic generation method according to claim 9, characterized in that: The hot topics are key hot topics determined based on the user demand prediction.

11. The method for automatically generating content according to claim 1, wherein: The style prompt words are selected from a plurality of preset style prompt words, and the plurality of preset style prompt words include standard style, consulting style, advisory style, relaxed style, and concise style.

12. The content automatic generation method according to claim 1, characterized in that: The multiple generative artificial intelligence models include a first generative artificial intelligence model, a second generative artificial intelligence model, a third generative artificial intelligence model, a fourth generative artificial intelligence model and a fifth generative artificial intelligence model, wherein the optimization strategy of the first generative artificial intelligence model has an optimization goal for the relevance and timeliness of the model input, the optimization strategy of the second generative artificial intelligence model has an optimization goal for information extraction and in-depth research of the model input, the third generative artificial intelligence model has an optimization goal for trend prediction of the model input, the optimization strategy of the fourth generative artificial intelligence model has an optimization goal for multi-industry analysis results of the model input, and the optimization strategy of the fifth generative artificial intelligence model has an optimization goal for the event context analysis results of the model input.

13. The content automatic generation method according to claim 12, characterized in that: The stylized content for the hot topic is a hot analysis report for the hot topic, the content associated with the hot topic generated by the first generative artificial intelligence model is the hot information scenario content associated with the hot topic, the content associated with the hot topic generated by the second generative artificial intelligence model is the research report analysis scenario content associated with the hot topic, the content associated with the hot topic generated by the third generative artificial intelligence model is the external market analysis scenario content associated with the hot topic, the content associated with the hot topic generated by the fourth generative artificial intelligence model is the multi-industry analysis scenario content associated with the hot topic, and the content associated with the hot topic generated by the fifth generative artificial intelligence model is the event development context scenario content associated with the hot topic.

14. The content automatic generation method according to claim 13, characterized in that: The style prompt words are used to set the combination between the scene contents generated by the multiple generative artificial intelligence models.

15. The content automatic generation method according to claim 13, characterized in that: The hot information scenario content associated with the hot topic is based on the hot information information associated with the hot topic after vectorization processing. The event development context scenario content associated with the hot topic is the event development context output result obtained by sorting out the hot topic according to the timeline associated with the hot topic through a large language model. The external market analysis scenario content associated with the hot topic and the multi-industry analysis scenario content associated with the hot topic are obtained by performing hot spot analysis through artificial intelligence agents and analysis frameworks.

16. The content automatic generation method according to claim 1, characterized in that: The content automatic generation method further includes: Through the style conversion module, based on another style prompt word different from the style prompt word, another stylized content for the hot topic is generated using the content associated with the hot topic generated by each of the multiple generative artificial intelligence models, wherein the another stylized content for the hot topic has a style corresponding to the another style prompt word.

17. The content automatic generation method according to claim 1, characterized in that: The hot topics are user-defined, or the hot topics are obtained by processing public domain hot information data through the hot spot screening module based on a hot spot capture algorithm, a hot spot trend analysis algorithm, and a hot word analysis algorithm.

18. A computer device, characterized in that: The computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the method according to any one of claims 1 to 17 when executing the computer program.

19. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, which, when executed on a computer device, cause the computer device to perform the method according to any one of claims 1 to 17.

20. A content automatic generation system, characterized in that: The automatic content generation system includes: Hotspot screening module, used to obtain hot topics; A hotspot analysis module, comprising a plurality of generative artificial intelligence models, configured to: input the hot topics into the plurality of generative artificial intelligence models, thereby obtaining content associated with the hot topics generated by each of the plurality of generative artificial intelligence models; and transmit the content associated with the hot topics generated by each of the plurality of generative artificial intelligence models to a style transfer module, wherein each optimization strategy of the plurality of generative artificial intelligence models has a different optimization objective for the model input, and the plurality of generative artificial intelligence models are optimized independently of one another; The style conversion module is configured to generate stylized content for the hot topic based on system prompt words including at least style prompt words and using the content associated with the hot topic generated by each of the multiple generative artificial intelligence models, including: Integrating the content related to the hot topic generated by each of the multiple generative artificial intelligence models according to a standard style, thereby obtaining standardized content for the hot topic in the standard style; Based on the style prompt word, selectively performing a content deletion operation and a stylized element addition operation on the standardized content for the hot topic, thereby obtaining the stylized content for the hot topic having a style corresponding to the style prompt word, The stylized content for the hot topic has a style corresponding to the style prompt word, the standard style is independent of the style corresponding to the style prompt word, and the standardized content for the hot topic with the standard style is independent of the style prompt word.

21. The content automatic generation system according to claim 20, characterized in that: The hot topic is used to indicate the content characteristics of the stylized content for the hot topic, and the style prompt word is used to indicate the style characteristics of the stylized content for the hot topic. The system prompt word also includes multiple task prompt words, and the multiple task prompt words correspond one-to-one to the multiple generative artificial intelligence models. Each of the multiple task prompt words is used to call the model service of the generative artificial intelligence model corresponding to the task prompt word among the multiple generative artificial intelligence models.

22. The content automatic generation system according to claim 20, characterized in that: The multiple generative artificial intelligence models include a first generative artificial intelligence model, a second generative artificial intelligence model, a third generative artificial intelligence model, a fourth generative artificial intelligence model and a fifth generative artificial intelligence model, wherein the optimization strategy of the first generative artificial intelligence model has an optimization goal for the relevance and timeliness of the model input, the optimization strategy of the second generative artificial intelligence model has an optimization goal for information extraction and in-depth research of the model input, the third generative artificial intelligence model has an optimization goal for trend prediction of the model input, the optimization strategy of the fourth generative artificial intelligence model has an optimization goal for multi-industry analysis results of the model input, and the optimization strategy of the fifth generative artificial intelligence model has an optimization goal for the event context analysis results of the model input.

Images