A content generation method and device, electronic equipment and storage medium
By using multiple content generation models in the content generation platform and determining the target content generation result based on similarity, the problem of insufficient stability caused by a single model is solved, and more stable and accurate content generation is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- GEEKBANG TECH LTD
- Filing Date
- 2026-03-04
- Publication Date
- 2026-06-09
AI Technical Summary
Existing content generation platforms rely on a single large language model, resulting in insufficient stability of the generated results, especially in scenarios such as article summarization, content rewriting, and key point extraction, where significant errors are prone to occur.
By sending the target content generation requirements to several content generation models, candidate content generation results are generated, and the overall target content generation result is determined by similarity, thus avoiding over-reliance on a single model.
It improves the stability and accuracy of content generation results and reduces the error between the generated results and the requirements of the target object.
Smart Images
Figure CN122174890A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of artificial intelligence technology, and in particular to a content generation method, apparatus, electronic device, and storage medium. Background Technology
[0002] With the development of large language model technology, AI-based content generation technology has been widely applied in content platforms and knowledge service scenarios.
[0003] Current content generation platforms often rely on a single large language model to process input content and generation instructions, and directly output the generated results. However, the stability of the generated results is insufficient. In scenarios such as article summarization, content rewriting, and key point extraction, the generated results are prone to significant errors. Therefore, how to improve the stability of the generated results as much as possible has become an important issue. Summary of the Invention
[0004] This invention provides a content generation method, apparatus, electronic device, and storage medium to solve the problem of insufficient stability of the generated results.
[0005] According to one aspect of the present invention, a content generation method is provided, the method comprising: Define the target content generation requirements; The target content generation requirements are sent to several content generation models respectively, and the candidate content generation results generated by each content generation model are obtained respectively. The similarity of the generated results for each candidate content is determined, and the target content is generated based on the similarity.
[0006] According to another aspect of the present invention, a content generation apparatus is provided, the apparatus comprising: The requirement determination module is used to determine the target content generation requirements; The content determination module is used to send the target content generation requirements to several content generation models respectively, and obtain the candidate content generation results generated by each content generation model respectively. The result generation module is used to determine the similarity of the generated results of each candidate content and generate the target content based on the similarity.
[0007] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the content generation method described in any embodiment of the present invention.
[0008] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the content generation method described in any embodiment of the present invention.
[0009] The technical solution of this invention involves determining the target content generation requirements; sending the target content generation requirements to several content generation models to obtain candidate content generation results generated by each content generation model; determining the similarity of each candidate content generation result; and generating the target content generation result based on the similarity. This integrates the candidate content generation results of various content generation models, thereby avoiding the problem of insufficient stability of the generation results due to over-reliance on a single large language model.
[0010] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description
[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0012] Figure 1 This is a flowchart of a content generation method provided in Embodiment 1 of the present invention; Figure 2 This is a flowchart of another content generation method provided in Embodiment 2 of the present invention; Figure 3 This is a schematic diagram of the structure of a content generation device according to Embodiment 3 of the present invention; Figure 4 This is a schematic diagram of the structure of an electronic device that implements the content generation method of this invention. Detailed Implementation
[0013] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.
[0014] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.
[0015] Example 1 Figure 1 This is a flowchart of a content generation method provided in Embodiment 1 of the present invention. This embodiment is applicable to situations where the accuracy of the generated results is ensured when using a model to generate content. This method can be executed by a content generation device, which can be implemented in hardware and / or software and can be configured in an electronic device with data processing capabilities. Figure 1 As shown, the method includes: S110. Determine the target content generation requirements.
[0016] Content generation requirements can be specific requirements for generating target content, including but not limited to the relevant content, word count requirements, paragraphing requirements, and expression style. Content generation requirements can be in the form of questions. Content generation requirements can be in the format of audio or text. Content generation requirements can also be specific constraints used to guide the model in generating content that meets the requirements.
[0017] For example, content generation requirements could be "generate a text describing the taste of apples" or "retrieve information about apple cultivation methods," etc.
[0018] Optionally, define the target content generation requirements, including: Determine the target audio to be sent by the target object; The target audio is converted into text to generate the target content. To facilitate the input of target content generation requirements by the target object, the target object can be allowed to send target audio, and the target audio can be parsed to obtain the content in the target audio, thereby generating the target content generation requirements.
[0019] Optionally, define the target content generation requirements, including: Receive the target content sent by the target object to generate the requirement.
[0020] When determining the target content generation requirements, some target objects may directly upload the target content generation requirements in text form. In this case, you can directly receive the target content generation requirements sent by the target objects.
[0021] Optionally, define the target content generation requirements, including: Determine the initial requirements for sending to the target object; The initial requirements are segmented to generate several target content requirements.
[0022] In practical use, when the target object sends a request to generate target content, it often sends too much content, which may cause the model to misunderstand and thus affect the accuracy of the generated content.
[0023] In response, upon receiving the initial request from the target object, the initial request will be semantically segmented to generate several target content generation requests.
[0024] The initial requirements are divided into segments, including: The initial requirements are segmented based on preset segmentation marks.
[0025] When segmenting the initial requirements, punctuation marks in the initial requirements can be identified first. When a preset segmentation punctuation mark appears, the initial requirements can be segmented at the preset segmentation punctuation mark. The preset segmentation punctuation marks can be ",", ";", and ".", etc., and this application does not limit them.
[0026] The initial requirements are divided into segments, including: The initial requirements are segmented based on preset segmentation words.
[0027] When segmenting the initial requirement, the text in the initial requirement can be recognized first. When a preset segmentation word appears, the initial requirement can be segmented based on the preset segmentation word. The preset segmentation word can be "in addition," "furthermore," or "and," etc., and this application does not limit it.
[0028] S120. Send the target content generation requirements to several content generation models respectively, and obtain the candidate content generation results generated by each content generation model.
[0029] The content generation model can be a model used to generate content corresponding to the target content generation requirements. The content generation model can be a commonly used model in scenarios such as article summarization, content rewriting, and key point extraction, such as the Large Language Model (LLM), etc., and this application does not impose any restrictions on this. The content generation model can be an existing model or a model generated separately after training according to specific needs.
[0030] After obtaining the target content generation requirements, these requirements can be input into different content generation models so that each model performs content generation operations based on the target content generation requirements, thereby generating candidate content generation results from each model.
[0031] For example, there are content generation model A, content generation model B, and content generation model C; and a target content generation requirement D. The target content generation requirement D is input into content generation model A, content generation model B, and content generation model C respectively, yielding candidate content generation results a, b, and c respectively.
[0032] S130. Determine the similarity of the generated results for each candidate content, and generate the target content based on the similarity.
[0033] After obtaining the generated results of each candidate content, a similarity verification is performed on each candidate content generated result to determine whether the generated results of each candidate content are similar, and the target content generated result is generated based on the similarity.
[0034] By determining the similarity of the generated results of each candidate content and generating the target content based on the similarity, the candidate content generation results of each content generation model are integrated, thereby avoiding the problem of insufficient stability of the generated results due to over-reliance on a single large language model.
[0035] Optionally, the target content generation result can be generated based on similarity, including: If the similarity between two candidate content generation results is less than the first preset threshold, then the candidate content generation results are grouped to obtain several result groups; each result group contains at least one candidate content generation result; wherein the similarity between the candidate content generation results in any two result groups is less than the second preset threshold.
[0036] The candidate content generation results in the result group are merged to obtain the target content generation results.
[0037] After obtaining several candidate content generation results, differences may exist between them. Therefore, it is necessary to determine and judge the similarity of each candidate content generation result. If the similarity of two candidate content generation results is less than a first preset threshold, the candidate content generation results are grouped to obtain several result groups. The candidate content generation results in the same result group are then merged to obtain the target content generation results.
[0038] For example, several candidate content generation results exist, and these results are grouped into group A, group B, and group C. The similarity between any two candidate content generation results in any two groups is less than a second preset threshold; that is, the similarity between candidate content generation results in group A and group B, group B and group C, and group A and group C is less than the second preset threshold. Then, the candidate content generation results within group A are merged, the candidate content generation results within group B are merged, and the candidate content generation results within group C are merged, thereby generating three target content generation results.
[0039] If the similarity between two candidate content generation results is less than a first preset threshold, the candidate content generation results are grouped to obtain several result groups. The candidate content generation results in the result groups are then merged to obtain the target content generation results. This avoids using only one set of candidate content generation results as the target content generation results, which could lead to a large error between the generated results and the requirements of the target object, thus ensuring the accuracy of the generated content.
[0040] The similarity of the generated results for each candidate content is determined, including: Semantic recognition is performed on the generated results of each candidate content, and pairwise similarity is determined based on the recognition results.
[0041] When determining the similarity of the generated results of each candidate content, semantic recognition can be performed on each generated result to determine whether the semantics of each pair of generated results are similar, thereby obtaining the similarity of the generated results of each candidate content.
[0042] Optionally, the similarity of the generated results for each candidate content can be determined, including: The structure of each candidate content generation result is identified, and the pairwise similarity of each candidate content generation result is determined based on the identification results.
[0043] When determining the similarity of the generated results of each candidate content, the generated results of the candidate content can also be structurally identified, and the pairwise similarity of the generated results of each candidate content can be determined based on whether the structures are the same.
[0044] Optionally, the similarity of the generated results for each candidate content can be determined, including: Information is extracted from the generated results of each candidate content, and pairwise similarity is determined based on the extraction results.
[0045] When determining the similarity of the generated results of each candidate content, information can be extracted from each generated result to obtain key information of each generated result, and pairwise similarity can be determined based on the key information of each generated result.
[0046] The technical solution of this application involves determining the target content generation requirements; sending the target content generation requirements to several content generation models to obtain the candidate content generation results generated by each content generation model; determining the similarity of each candidate content generation result; and generating the target content generation result based on the similarity. This integrates the candidate content generation results of various content generation models, thereby avoiding the problem of insufficient stability of the generation results due to over-reliance on a single large language model.
[0047] Example 2 Figure 2 This invention provides a flowchart of another content generation method. This embodiment further optimizes the process of generating target content based on similarity in the aforementioned embodiments, building upon the above embodiments. This embodiment can be combined with various optional solutions in one or more of the above embodiments. Figure 2 As shown, the content generation method of this embodiment may include the following steps: S210. Determine the target content generation requirements.
[0048] S220. Send the target content generation requirements to several content generation models respectively, and obtain the candidate content generation results generated by each content generation model.
[0049] S230. Determine the similarity of the generated results for each candidate content.
[0050] S240. If the similarity between any two candidate content generation results is greater than the first preset threshold, then select one candidate content generation result from each candidate content generation result as the target content generation result.
[0051] After determining the similarity of the generated results of each candidate content, if the similarity of any two generated results of candidate content is greater than the first preset threshold, it can be determined that the generated results of each candidate content are sufficiently similar. At this point, one of the generated results of candidate content can be selected as the target content generated result.
[0052] The technical solution of this application determines the target content generation requirements; the target content generation requirements are sent to several content generation models respectively, and the candidate content generation results generated by each content generation model are obtained respectively; the similarity of each candidate content generation result is determined; if the similarity of any two candidate content generation results is greater than a first preset threshold, then one candidate content generation result is selected from each candidate content generation result as the target content generation result, thereby achieving accurate determination of the target content generation result.
[0053] Example 3 Figure 3 This invention provides a structural block diagram of a content generation device, applicable to situations where the accuracy of the generated results is ensured when using a model for content generation. This content generation device can be implemented in hardware and / or software and can be configured in an electronic device with data processing capabilities. Figure 3 As shown, the content generation apparatus of this embodiment may include: a generation requirement determination module 310, a generation content determination module 320, and a generation result construction module 330. Wherein: The requirement determination module 310 is used to determine the target content generation requirements; The content determination module 320 is used to send the target content generation requirements to several content generation models respectively, and obtain the candidate content generation results generated by each content generation model respectively. The result generation module 330 is used to determine the similarity of the generated results of each candidate content and generate the target content based on the similarity.
[0054] Based on the above embodiments, optionally, generating target content generation results based on similarity includes: If the similarity between any two candidate content generation results is greater than the first preset threshold, then one candidate content generation result is selected from all candidate content generation results as the target content generation result.
[0055] Based on the above embodiments, optionally, generating target content generation results based on similarity includes: If the similarity between two candidate content generation results is less than the first preset threshold, then the candidate content generation results are grouped to obtain several result groups; each result group contains at least one candidate content generation result; wherein the similarity between the candidate content generation results in any two result groups is less than the second preset threshold. The candidate content generation results in the result group are merged to obtain the target content generation results.
[0056] Based on the above embodiments, optionally, similarity determination is performed on the generation results of each candidate content, including: Semantic recognition is performed on the generated results of each candidate content, and pairwise similarity is determined based on the recognition results.
[0057] Based on the above embodiments, optionally, similarity determination is performed on the generation results of each candidate content, including: The structure of each candidate content generation result is identified, and the pairwise similarity of each candidate content generation result is determined based on the identification results.
[0058] Based on the above embodiments, optionally, similarity determination is performed on the generation results of each candidate content, including: Information is extracted from the generated results of each candidate content, and pairwise similarity is determined based on the extraction results.
[0059] The content generation apparatus provided in the embodiments of the present invention can execute the content generation method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of executing the method.
[0060] Example 4 Figure 4 A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.
[0061] like Figure 4As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.
[0062] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.
[0063] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as content generation methods.
[0064] In some embodiments, the content generation method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or mounted on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the content generation method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the content generation method by any other suitable means (e.g., by means of firmware).
[0065] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.
[0066] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.
[0067] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.
[0068] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).
[0069] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.
[0070] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.
[0071] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.
[0072] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.
Claims
1. A content generation method, characterized in that, include: Define the target content generation requirements; The target content generation requirements are sent to several content generation models respectively, and the candidate content generation results generated by each content generation model are obtained respectively. The similarity of the generated results for each candidate content is determined, and the target content is generated based on the similarity.
2. The method according to claim 1, characterized in that, The generated target content based on similarity includes: If the similarity between any two candidate content generation results is greater than the first preset threshold, then one candidate content generation result is selected from all candidate content generation results as the target content generation result.
3. The method according to claim 1, characterized in that, The generated target content based on similarity includes: If the similarity between two candidate content generation results is less than the first preset threshold, then the candidate content generation results are grouped to obtain several result groups; each result group contains at least one candidate content generation result; wherein the similarity between the candidate content generation results in any two result groups is less than the second preset threshold. The candidate content generation results in the result group are merged to obtain the target content generation results.
4. The method according to claim 1, characterized in that, The similarity of the generated results for each candidate content is determined, including: Semantic recognition is performed on the generated results of each candidate content, and pairwise similarity is determined based on the recognition results.
5. The method according to claim 1, characterized in that, The similarity of the generated results for each candidate content is determined, including: The structure of each candidate content generation result is identified, and the pairwise similarity of each candidate content generation result is determined based on the identification results.
6. The method according to claim 1, characterized in that, The similarity of the generated results for each candidate content is determined, including: Information is extracted from the generated results of each candidate content, and pairwise similarity is determined based on the extraction results.
7. A content generation device, characterized in that, include: The requirement determination module is used to determine the target content generation requirements; The content determination module is used to send the target content generation requirements to several content generation models respectively, and obtain the candidate content generation results generated by each content generation model respectively. The result generation module is used to determine the similarity of the generated results of each candidate content and generate the target content based on the similarity.
8. The apparatus according to claim 7, characterized in that, The generated result construction module is specifically used for: If the similarity between any two candidate content generation results is greater than the first preset threshold, then one candidate content generation result is selected from all candidate content generation results as the target content generation result.
9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the content generation method according to any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the content generation method of any one of claims 1-7.