Content acquisition method and device and electronic equipment

Abstract

The invention provides a content acquisition method. The method comprises the following steps: in response to a content acquisition request from target content to a first intelligent engine, determining a target temperature parameter for the target content based on the target content; and configuring the target temperature parameter to the first intelligent engine, so that the first intelligent engine processes the target content according to the target temperature parameter. Wherein the first intelligent engine is used for outputting the to-be-acquired content, and the target temperature parameter is used for regulating and controlling the predictable degree of the to-be-acquired content.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, and in particular to a content acquisition method, device and electronic device. Background Art

[0002] Large Language Model (LLM) is a deep learning model trained with a large amount of text data, designed to generate and understand natural language text. These models play an important role in the field of natural language processing (NLP) and are widely used in a variety of tasks such as text generation, machine translation, sentiment analysis, question-answering systems, etc.

[0003] For a pre-trained large language model, its text generation effect is controlled by the model's inference parameters by default. In existing text generation scenarios, users are allowed to manually change parameters, but this method is neither intuitive nor accurate. Most of them even use fixed temperature parameters, and the text generation effect needs to be improved. Summary of the invention

[0004] One aspect of the present disclosure provides a content acquisition method, comprising: in response to a content acquisition request to a first intelligent engine with target content, determining a target temperature parameter for the target content based on the target content. The target temperature parameter is configured to the first intelligent engine so that the first intelligent engine processes the target content with the target temperature parameter. The first intelligent engine is used to output the content to be acquired, and the target temperature parameter is used to adjust the predictability of the content to be acquired.

[0005] Optionally, based on the target content, determining the target temperature parameter for the target content includes: obtaining a content acquisition requirement corresponding to the target content. Based on the content acquisition requirement, determining the target temperature parameter for the target content. The content acquisition requirement is a content acquisition intention, and the content acquisition intention includes a content generation intention and a content rewriting intention. The content generation intention represents content generation based on the target content, and the content rewriting intention represents rewriting the target content.

[0006] Optionally, obtaining the content acquisition requirement corresponding to the target content includes: inputting the target content into a second intelligent engine to obtain the content acquisition requirement corresponding to the target content. The second intelligent engine is used to infer the content acquisition intention based on the target content to determine that the content acquisition requirement is one of the content generation intention and the content rewriting intention.

[0007] Optionally, based on the target content, determining the target temperature parameter for the target content includes: obtaining a content type corresponding to the target content. Based on the content type, determining the target temperature parameter for the target content. The content type is a text type of the content to be obtained, and the text type of the content to be obtained includes a first text type and a second text type, and the predictability of the content corresponding to the first text type is greater than the predictability of the content corresponding to the second text type.

[0008] Optionally, obtaining the content type corresponding to the target content includes: inputting the target content into a third intelligent engine to obtain the content type corresponding to the target content. The third intelligent engine is used to infer the text type of the content to be obtained based on the target content to determine that the content type is one of the first text type and the second text type.

[0009] Optionally, based on the target content, determining the target temperature parameters for the target content includes: obtaining the content acquisition requirements and content type corresponding to the target content. The content acquisition requirements and content type are input into the fourth intelligent engine to obtain the target temperature parameters corresponding to the target content. Among them, the fourth intelligent engine is used to predict or adjust the target temperature parameters according to the content acquisition requirements and content type. The content acquisition requirements are content acquisition intentions, and the content acquisition intentions include content generation intentions and content rewriting intentions. The content generation intention represents content generation based on the target content, and the content rewriting intention represents rewriting the target content. The content type is the text type of the content to be acquired, and the text type of the content to be acquired includes a first text type and a second text type. The predictability of the content corresponding to the first text type is greater than the predictability of the content corresponding to the second text type.

[0010] Optionally, the content acquisition method further includes: in response to the user selecting a target output content corresponding to the target content from the plurality of output contents, determining a first temperature parameter corresponding to the target content based on the similarity between the target output content and the target content, and using the target content and the first temperature parameter as training samples for the fourth intelligent engine.

[0011] Optionally, training the fourth intelligent engine includes: in response to the user adjusting the initial temperature parameter of the first intelligent engine to a custom temperature parameter, inputting the difference between the custom temperature parameter and the initial temperature parameter to the fourth intelligent engine to obtain a second temperature parameter. In response to the user repeatedly adjusting the temperature parameter until the difference between the second temperature parameters obtained by two consecutive adjustments is less than a preset threshold.

[0012] Another aspect of the present disclosure provides a content acquisition device, including: a determination module, for responding to a content acquisition request with target content to a first intelligent engine, and determining a target temperature parameter for the target content based on the target content. A processing module, for configuring the target temperature parameter to the first intelligent engine, so that the first intelligent engine processes the target content with the target temperature parameter. The first intelligent engine is used to output the content to be acquired, and the target temperature parameter is used to adjust the predictability of the content to be acquired.

[0013] Another aspect of the present disclosure provides an electronic device, comprising: one or more processors. An input component for receiving target content. The processor is used to execute a content acquisition method, and the content acquisition method includes: in response to a content acquisition request with the target content to a first intelligent engine, based on the target content, determining a target temperature parameter for the target content. The target temperature parameter is configured to the first intelligent engine so that the first intelligent engine processes the target content with the target temperature parameter. The first intelligent engine is used to output the content to be acquired, and the target temperature parameter is used to regulate the predictability of the content to be acquired. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The above and other objects, features and advantages of the present disclosure will become more apparent through the following description of the embodiments of the present disclosure with reference to the accompanying drawings, in which:

[0015] Figure 1 The application scenario diagram of the content acquisition method according to the embodiment of the present disclosure is schematically shown;

[0016] Figure 2 A flowchart of a content acquisition method according to an embodiment of the present disclosure is schematically shown;

[0017] Figure 3 A flow chart of a method for determining a target temperature parameter according to an embodiment of the present disclosure is schematically shown;

[0018] Figure 4 A flowchart of a method for obtaining content acquisition requirements according to an embodiment of the present disclosure is schematically shown;

[0019] Figure 5 A flow chart of a method for determining a target temperature parameter according to another embodiment of the present disclosure is schematically shown;

[0020] Figure 6 A flowchart of a method for obtaining a content type corresponding to a target content according to an embodiment of the present disclosure is schematically shown;

[0021] Figure 7 A flow chart of a method for determining a target temperature parameter according to another embodiment of the present disclosure is schematically shown;

[0022] Figure 8A flowchart of a method for training a fourth intelligent engine according to an embodiment of the present disclosure is schematically shown;

[0023] Figure 9A A flowchart of a method for training a fourth intelligent engine according to another embodiment of the present disclosure is schematically shown; Figure 9B The following schematically shows a logical architecture diagram of training the fourth intelligent engine according to an embodiment of the present disclosure;

[0024] Figure 10 The structure block diagram of the content acquisition device according to the embodiment of the present disclosure is schematically shown;

[0025] Figure 11 A block diagram of an electronic device suitable for implementing a content acquisition method according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0026] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present disclosure. In addition, in the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0027] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise", "include", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0028] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0029] Some block diagrams and / or flow charts are shown in the accompanying drawings. It should be understood that some blocks or combinations thereof in the block diagrams and / or flow charts may be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that these instructions, when executed by the processor, may create a device for implementing the functions / operations described in these block diagrams and / or flow charts.

[0030] Therefore, the technology of the present disclosure can be implemented in the form of hardware and / or software (including firmware, microcode, etc.). In addition, the technology of the present disclosure can take the form of a computer program product on a computer-readable medium storing instructions, which can be used by an instruction execution system or in combination with an instruction execution system. In the context of the present disclosure, a computer-readable medium can be any medium that can contain, store, transmit, propagate, or transmit instructions. For example, a computer-readable medium can include, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, device, or propagation medium. Specific examples of computer-readable media include: magnetic storage devices, such as magnetic tape or hard disk (HDD); optical storage devices, such as compact disk (CD-ROM); memory, such as random access memory (RAM) or flash memory; and / or wired / wireless communication links.

[0031] An embodiment of the present disclosure provides a method for adaptively adjusting the predictability of content generated by a large language model. The content acquisition method includes: in response to a content acquisition request with target content to a first intelligent engine, determining a target temperature parameter for the target content based on the target content. The target temperature parameter is configured to the first intelligent engine so that the first intelligent engine processes the target content with the target temperature parameter. Among them, the first intelligent engine is used to output the content to be acquired, and the target temperature parameter is used to regulate the predictability of the content to be acquired. By identifying the target content in the user's content acquisition request, the temperature parameters of the first intelligent engine can be adjusted so that the content generated by it matches the target content requested by the user, thereby improving the content generation efficiency and effect of the first intelligent engine (i.e., the large language model).

[0032] Figure 1 The application scenario diagram of the content acquisition method according to the embodiment of the present disclosure is schematically shown.

[0033] like Figure 1 As shown, the application scenario 100 according to this embodiment may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for a communication link between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0034] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, 103, such as security applications, shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only for example).

[0035] The terminal devices 101 , 102 , and 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers.

[0036] The server 105 may be a server that provides various services, such as a background management server (only an example) that provides support for websites browsed or applications logged in by users using the terminal devices 101, 102, and 103. The background management server may analyze and process the received data such as user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal device.

[0037] It should be noted that the content acquisition method provided in the embodiment of the present disclosure can generally be executed by the server 105 or by the terminal device. Accordingly, the content acquisition device provided in the embodiment of the present disclosure can generally be set in the server 105. The content acquisition method provided in the embodiment of the present disclosure can also be executed by a server or server cluster that is different from the server 105 and can communicate with the terminal devices 101, 102, 103 and / or the server 105. Accordingly, the content acquisition device provided in the embodiment of the present disclosure can also be set in a server or server cluster that is different from the server 105 and can communicate with the terminal devices 101, 102, 103 and / or the server 105.

[0038] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0039] The following will be based on Figure 1 The scene described by Figures 2 to 9B The content acquisition method of the disclosed embodiment is described in detail.

[0040] Figure 2 The flowchart of the content acquisition method according to the embodiment of the present disclosure is schematically shown.

[0041] According to the embodiments of the present disclosure, Figure 2 As shown, the content acquisition method of this embodiment includes, for example, operations S210 to S220.

[0042] In operation S210, in response to a content acquisition request to a first intelligent engine with target content, a target temperature parameter for the target content is determined based on the target content, wherein the first intelligent engine is used to output the content to be acquired, and the target temperature parameter is used to adjust the predictability of the content to be acquired.

[0043] The system first receives a content acquisition request for a specific target content, and sends the request to the first intelligent engine, requesting the engine to provide relevant information or content according to the target content.

[0044] After receiving the request, the system will calculate and determine a target temperature parameter for the target content through a series of predefined algorithms or models based on the specific properties of the target content (such as topic, type, context, etc.). This parameter is a quantitative value used to guide the predictability or diversity of the content in the subsequent processing process.

[0045] For example, the first intelligent engine can be a software system that integrates a variety of advanced artificial intelligence technologies and has powerful data processing, analysis, reasoning and generation capabilities. This intelligent engine can receive and process data from different sources, including text, images, audio, etc., and generate content or information that meets specific needs based on preset algorithms and models.

[0046] In some embodiments, when generating story content such as novels and scripts, the plot development, character behavior, and dialogue diversity of the story can be controlled by adjusting the temperature parameter. A low temperature parameter may produce a more linear and predictable story line, while a high temperature parameter may introduce more unexpected twists and surprises.

[0047] When writing news reports, you can also adjust the temperature parameters to balance the accuracy of information and the diversity of reports based on the importance and timeliness of the news. For emergencies, you may need to lower the temperature parameters to ensure the accuracy and stability of the report; while for in-depth investigative reports, you can appropriately increase the temperature parameters to explore more possibilities and perspectives.

[0048] For example, in search engines, temperature parameters can be used to adjust the relevance and diversity of search results. A low temperature parameter may be more likely to return results that are highly relevant to the user's query, while a high temperature parameter may introduce more relevant but slightly different results to broaden the user's horizon.

[0049] In personalized recommendation systems on e-commerce, social media and other platforms, temperature parameters can be adjusted based on user preferences and historical behaviors. Low temperature parameters may produce recommendations that are more in line with the user's current interests, while high temperature parameters may recommend new areas or products that the user has not tried but may be interested in.

[0050] For example, in a conversation with an intelligent assistant or chatbot, the temperature parameter can be used to adjust the formality, humor, and creativity of the response. A low temperature parameter may produce a more formal and conservative response, while a high temperature parameter may introduce more humorous elements and creative expressions to enhance the user's interactive experience.

[0051] In the voice navigation system, the temperature parameters can be adjusted according to the road conditions, weather and the user's emotional state. For example, in congested roads or bad weather, the temperature parameters can be lowered to provide more stable and accurate navigation information; while in a relaxed and enjoyable trip, the temperature parameters can be appropriately increased to increase the fun and interactivity of the voice navigation.

[0052] For example, in image style transfer or artistic creation, the temperature parameter can be used to adjust the similarity and innovation of the converted image. A low temperature parameter may produce a style image that is more similar to the original image but slightly changed, while a high temperature parameter may produce a new image that is very different from the original image but has a unique artistic style.

[0053] In operation S220 , the target temperature parameter is configured to the first smart engine, so that the first smart engine processes the target content with the target temperature parameter.

[0054] The system configures the calculated target temperature parameters to the first intelligent engine. This step is critical because it ensures that the intelligent engine operates according to a predetermined predictability or diversity level when processing the target content.

[0055] After the configuration is completed, the first intelligent engine uses the configured target temperature parameters to process and filter data or information related to the target content. This process may include retrieving, analyzing, and aggregating data from databases, networks, or other data sources, and adjusting the output form of the content (such as the depth, breadth, and innovation of the content) based on the target temperature parameters to ensure that the output content to be obtained meets both the user's original request and the predetermined predictability or diversity requirements.

[0056] Finally, the system presents the content to be acquired after being processed by the first intelligent engine to the user. The content may be screened, sorted, summarized or optimized in other forms to meet the specific needs and preferences of the user.

[0057] By receiving the user's content acquisition request, the system can intelligently determine and configure the target temperature parameters suitable for the target content, thereby guiding the first intelligent engine to process the content in a specific way, and finally output the content to be acquired that is both relevant and meets the specific predictability or diversity requirements. This method enhances the flexibility and personalization of content acquisition and improves the user experience.

[0058] Figure 3 A flow chart of a method for determining a target temperature parameter according to an embodiment of the present disclosure is schematically shown.

[0059] According to the embodiments of the present disclosure, Figure 3 In addition to the above references Figure 2In addition to the operations S210-S220 described above, the content acquisition method of this embodiment also includes operations S311-S312 to determine the target temperature parameter for the target content based on the target content. For the sake of brevity, the description of operations S210-S220 is omitted here, and the subsequent related method embodiments are similar and will not be repeated.

[0060] In operation S311, a content acquisition requirement corresponding to the target content is acquired, wherein the content acquisition requirement is a content acquisition intention, and the content acquisition intention includes a content generation intention and a content rewriting intention, wherein the content generation intention indicates that content is generated based on the target content, and the content rewriting intention indicates that the target content is rewritten.

[0061] Before responding to the request, the system further analyzes the content acquisition needs behind the request, that is, the user's content acquisition intention. This step may involve natural language processing (NLP) technology to parse keywords and semantic information in the user's request to determine whether the user wants to generate content or rewrite content.

[0062] In operation S312 , a target temperature parameter for target content is determined based on the content acquisition requirement.

[0063] If the user's intention is to generate new content based on the target content (such as creating an article or generating a description), the system determines an appropriate target temperature parameter based on the expected style, depth, and innovation of the generated content. A high temperature parameter may encourage the generation of more innovative and diverse content, while a low temperature parameter may produce more stable and predictable content.

[0064] If the user's intention is to rewrite the target content in some form (such as simplifying, expanding, changing the style, etc.), the system will determine a target temperature parameter that can balance retaining the original intention and introducing changes based on the specific requirements of the rewriting task and the characteristics of the target content.

[0065] Based on the above analysis and differentiation of content acquisition requirements, the system uses a preset algorithm or model, combined with the characteristics of the target content, to calculate and determine an optimal target temperature parameter for the current content acquisition requirements.

[0066] The determined target temperature parameters are configured to the first intelligent engine, and the engine processes the target content according to the parameters to generate or rewrite the content to be acquired that meets the user's needs.

[0067] By introducing the concept of content acquisition requirements (content acquisition intent), the process of determining the target temperature parameters is further refined. It enables the system to more accurately understand the real needs of users and adjust the temperature parameters accordingly to optimize the results of content generation or rewriting. Whether it is content generation or content rewriting, the system can flexibly adjust the temperature parameters according to the specific intentions of users and the characteristics of the target content to produce content that meets the needs and has an appropriate degree of predictability or diversity. This approach not only improves the intelligence level of content acquisition, but also enhances the personalization of user experience.

[0068] Figure 4 A flow chart of a method for obtaining content acquisition requirements according to an embodiment of the present disclosure is schematically shown.

[0069] According to the embodiments of the present disclosure, Figure 4 As shown, for example, the content acquisition requirement corresponding to the target content is acquired through operation S4111.

[0070] In operation S4111, the target content is input into the second intelligent engine to obtain a content acquisition requirement corresponding to the target content, wherein the second intelligent engine is used to infer the content acquisition intention based on the target content to determine that the content acquisition requirement is one of the content generation intention and the content rewriting intention.

[0071] The target content is input into the second intelligent engine, which is, for example, an intelligent system specially designed to understand and analyze content, and has advanced natural language processing (NLP) and machine learning capabilities.

[0072] For example, the second intelligent engine can be an intent recognition engine based on BERT (Bidirectional Encoder Representations from Transformers). The user inputs the target content, and the BERT model encodes the content to generate a context-related vector representation. These vectors are classified using an intent classifier (which can be a multi-layer perceptron, convolutional neural network, etc.) to determine whether they belong to content generation intent or content rewriting intent. The intent classifier has been trained with a large amount of annotated data and can accurately identify user intent in different situations.

[0073] For another example, the second intelligent engine can be a deep learning-driven dialogue management system. It can not only understand the user's natural language input, but also dynamically adjust its understanding of the user's intention during the dialogue process. It can support multiple rounds of dialogue interaction and more accurately grasp the user's content acquisition needs through continuous questioning and clarification.

[0074] For another example, the second intelligent engine can also be a reasoning engine based on the knowledge graph. A structured knowledge base that can represent the relationships and attributes between entities. The reasoning process is, for example, that the system first builds or accesses a knowledge graph that contains relevant knowledge and rules such as content generation and content rewriting. When the user enters the target content, the system determines the user's content acquisition needs through entity recognition and relationship reasoning in the knowledge graph. Combined with the domain knowledge and expert rules in the knowledge graph, the system can understand the user's intentions more accurately and recommend appropriate target temperature parameters.

[0075] The second intelligent engine conducts in-depth analysis of the target content, identifies key words, semantic structures, contextual information, etc., and makes inferences and judgments based on this information.

[0076] Based on the inference results, the second intelligent engine determines the content acquisition requirements corresponding to the target content, that is, whether the user wants to generate content or rewrite content. This judgment process may involve scoring and sorting multiple potential intents, and ultimately selecting the intent with the highest score as the content acquisition requirement.

[0077] Once the content acquisition requirement (content generation intention or content rewriting intention) is determined, the system can calculate and determine a suitable target temperature parameter based on the requirement and the characteristics of the target content, and configure the target temperature parameter to the first intelligent engine, which generates or rewrites the target content accordingly based on the parameter.

[0078] By introducing a second intelligent engine to automatically reason about content acquisition requirements, the user operation process is further simplified and the system's intelligence level is improved. The user only needs to provide the target content, and the system can automatically determine the content acquisition intention behind it (content generation or content rewriting), and determine the appropriate target temperature parameters based on this to optimize the content processing process. This method not only reduces the user's operating burden, but also improves the accuracy and efficiency of content acquisition, providing users with a more convenient and personalized content service experience.

[0079] Figure 5 A flow chart of a method for determining a target temperature parameter according to another embodiment of the present disclosure is schematically shown.

[0080] According to the embodiments of the present disclosure, Figure 5 As shown, for example, through operations S511 to S512 , a target temperature parameter for the target content is determined based on the target content.

[0081] In operation S511, the content type corresponding to the target content is obtained, wherein the content type is the text type of the content to be obtained, the text type of the content to be obtained includes a first text type and a second text type, and the predictability of the content corresponding to the first text type is greater than the predictability of the content corresponding to the second text type.

[0082] Based on the above embodiments, this embodiment introduces the content type as a key factor in determining the target temperature parameter. By identifying the text type of the target content, the system can more accurately adjust the temperature parameter to generate or rewrite content that better meets the characteristics of the text type.

[0083] The system parses the target content and identifies the text type it belongs to. These text types can be predefined, such as news reports, academic papers, novels, technical documents, etc. Each type has a unique language style, structural characteristics and predictability.

[0084] During this process, the system may use natural language processing (NLP) techniques, such as text classification and topic models, to assist in identifying content types.

[0085] In operation S512 , based on the content type, a target temperature parameter for the target content is determined.

[0086] Once the content type of the target content is determined, the system determines the target temperature parameters based on the characteristics of that type.

[0087] Assuming that the content corresponding to the first text type (such as news reports, academic papers, etc.) is highly predictable, the system may choose a lower temperature parameter to ensure that the generated or rewritten content maintains a high degree of accuracy and stability.

[0088] For the second text type (such as novel stories, creative copywriting, etc.), whose content is less predictable, the system may choose a higher temperature parameter to encourage the generation of more innovative and diverse content.

[0089] The determined target temperature parameters are configured to the first intelligent engine, and the engine generates or rewrites the target content accordingly according to the parameters.

[0090] By introducing content type as the basis for determining the target temperature parameter, the personalization of the content acquisition method is further refined. The system can flexibly adjust the temperature parameters according to the characteristics and requirements of different text types to generate or rewrite content that not only meets the type specifications but also has an appropriate degree of predictability or diversity. This method not only improves the intelligence level of content acquisition, but also enhances the pertinence and satisfaction of user experience.

[0091] Figure 6The flowchart of the method for obtaining the content type corresponding to the target content according to the embodiment of the present disclosure is schematically shown.

[0092] According to the embodiments of the present disclosure, Figure 6 As shown, for example, the content type corresponding to the target content is obtained through operation S6111.

[0093] In operation S6111, the target content is input into the third intelligent engine to obtain the content type corresponding to the target content. The third intelligent engine is used to infer the text type of the content to be obtained based on the target content to determine that the content type is one of the first text type and the second text type.

[0094] In some embodiments, the content acquisition method is further refined and optimized, and more intelligent and personalized elements are introduced to better meet user needs for different content types.

[0095] The third intelligent engine, for example, uses deep learning technology (such as BERT, GPT, etc.) and natural language processing (NLP) to achieve accuracy and efficiency in identifying the content type of the target content. This is not limited to distinguishing between the first text type (such as news, blogs) and the second text type (such as novels, essays), but can also be further subdivided into more specific subcategories (such as technology news, entertainment, science fiction, etc.).

[0096] In order to further improve the accuracy of the model in identifying content types, during the content acquisition process, the system can also build user portraits by collecting data such as user browsing history, click behavior, comment feedback, etc. These data are used to analyze user interests, preferences and reading habits, and provide a basis for subsequent personalized content recommendations.

[0097] Combining the user portrait and the content type recognition results of the third intelligent engine, the system can recommend content that better suits the user's interests. For example, if the user prefers science and technology news, the system will prioritize the latest information in related fields.

[0098] A real-time feedback mechanism can also be introduced to allow users to instantly evaluate the recommended content (such as like, dislike, not interested, etc.). These feedbacks will be used as training data to further optimize the accuracy of the third intelligent engine and user portraits.

[0099] The system dynamically adjusts content recommendation strategies based on users' real-time feedback and long-term behavior patterns to ensure that recommended content is always consistent with user interests.

[0100] In addition to the original content sources, the system can also continuously explore new content sources, including social media, professional forums, personal blogs, etc., to enrich the content library and increase the diversity and timeliness of the content.

[0101] De-duplicate content from different sources to avoid duplicate recommendations. At the same time, intelligently sort recommended content based on factors such as content quality, timeliness, and user preferences to ensure that users see the most interesting and valuable content first.

[0102] By introducing a third intelligent engine to infer the type of target content, the intelligence level and user experience of the content acquisition method are further improved. This method can not only accurately identify the content type of the target content, but also make accurate recommendations based on the user's personalized needs, bringing users a more efficient, convenient and personalized content acquisition experience.

[0103] Figure 7 A flow chart of a method for determining a target temperature parameter according to yet another embodiment of the present disclosure is schematically shown.

[0104] According to the embodiments of the present disclosure, Figure 7 As shown, for example, through operations S711 to S712, a target temperature parameter for the target content is determined based on the target content.

[0105] In operation S711, content acquisition requirements and content types corresponding to target content are acquired.

[0106] In operation S712, the content acquisition requirements and content types are input into the fourth intelligent engine to obtain the target temperature parameters corresponding to the target content. The fourth intelligent engine is used to predict or adjust the target temperature parameters according to the content acquisition requirements and content types. The content acquisition requirements are content acquisition intentions, and the content acquisition intentions include content generation intentions and content rewriting intentions. The content generation intention represents content generation based on the target content, and the content rewriting intention represents rewriting the target content. The content type is the text type of the content to be acquired, and the text type of the content to be acquired includes a first text type and a second text type. The predictability of the content corresponding to the first text type is greater than the predictability of the content corresponding to the second text type.

[0107] In some embodiments, content acquisition requirements (i.e., content acquisition intent) and content types are further integrated as dual bases for determining target temperature parameters, and more refined and personalized content acquisition is achieved by introducing a fourth intelligent engine.

[0108] The system obtains the user's content acquisition needs through some means (such as user input, context analysis, second intelligent engine reasoning, etc.), that is, whether the user wants to generate content or rewrite content. These two needs correspond to content generation intention and content rewriting intention respectively.

[0109] The system uses the third intelligent engine to parse the target content and identify the text type to which it belongs. These text types include but are not limited to the first text type (such as code, technical documents, general knowledge questions and answers, news reports, academic papers, etc., which have a high degree of content predictability) and the second text type (such as travel notes, poems, novels, essays, etc., which have a low degree of content predictability).

[0110] The content acquisition requirements (content generation intention or content rewriting intention) and content type are sent as input to the fourth intelligent engine.

[0111] The fourth intelligent engine is an intelligent system that integrates machine learning, natural language processing (NLP) and other technologies. It predicts or adjusts the target temperature parameters based on the input content acquisition requirements and content types. In this process, the system may consider the special requirements of different content types for content generation or rewriting, as well as the specific needs of users (such as whether innovation is needed, accuracy is maintained, etc.) to determine a target temperature parameter that best suits the current situation.

[0112] For example, the fourth intelligent engine can follow a series of complex logic and algorithms when reasoning about temperature parameters, which incorporate machine learning, natural language processing (NLP), and possible domain-specific knowledge.

[0113] First, the fourth intelligent engine needs to parse the content acquisition requirements transmitted by the user or the system, that is, to determine whether the user wants to generate content or rewrite content. This can be achieved through technologies such as text classification and intent recognition. At the same time, the engine needs to understand the text type of the target content, which can be determined, for example, based on the output of the third intelligent engine or other text type recognition technologies. Different text types may correspond to different processing difficulties and expected effects.

[0114] Then, key features are extracted from the target content, such as themes, keywords, sentence structures, sentiment tendencies, etc. These features help the engine understand the nature and context of the content. Content acquisition requirements are converted into quantifiable features, such as whether it needs to be highly innovative, whether it needs to maintain the original meaning, whether it needs to follow a specific style, etc.

[0115] Then use the preset rule set to directly map to a specific temperature parameter range based on the content type and content acquisition requirements. This method is simple and direct, but may lack flexibility. Machine learning models, such as neural networks, decision trees, or random forests, can also be used to train large amounts of historical data to learn the mapping relationship from content features, demand features to temperature parameters. This method can capture complex nonlinear relationships and provide more accurate predictions. It is also possible to combine rule-based methods with machine learning models, using the determinism of rules to assist the prediction of machine learning models and improve the accuracy and stability of reasoning.

[0116] In addition, a user feedback mechanism can be introduced to allow users to evaluate the generated or rewritten content and adjust the temperature parameters based on the feedback. This can be achieved through online learning, so that the model can be continuously optimized. During the content generation or rewriting process, the temperature parameters are dynamically adjusted according to the intermediate results and context changes to ensure that the final output meets user needs and expectations.

[0117] For content in specific fields (such as law, medicine, etc.), the fourth intelligent engine may need to integrate specific domain knowledge for reasoning. This can be achieved by introducing domain expert rules, pre-trained language models or professional databases.

[0118] The determined target temperature parameter is configured to the first intelligent engine, and the first intelligent engine processes (generates or rewrites) the target content accordingly according to the parameter.

[0119] By comprehensively considering the content acquisition requirements (content generation intention or content rewriting intention) and content type (first text type or second text type), and introducing the fourth intelligent engine to predict or adjust the target temperature parameters, a more refined and personalized content acquisition method is achieved. This method not only improves the accuracy and diversity of content generation, but also better meets the actual needs of users and improves the user experience. At the same time, it also shows how to achieve efficient and accurate content acquisition by comprehensively applying multiple intelligent technologies and methods in complex content processing tasks.

[0120] Figure 8 A flow chart of a method for training a fourth intelligent engine according to an embodiment of the present disclosure is schematically shown.

[0121] According to the embodiments of the present disclosure, Figure 8 As shown, for example, the fourth intelligent engine is trained by operations S810~S820.

[0122] In operation S810 , in response to a user selecting target output content corresponding to target content from among a plurality of output contents, a first temperature parameter corresponding to the target content is determined based on a similarity between the target output content and the target content.

[0123] In operation S820, the target content and the first temperature parameter are used as training samples of a fourth intelligent engine.

[0124] In some embodiments, a user feedback mechanism is introduced to infer the first temperature parameter preferred by the user through the user's selection of output content, and the fourth intelligent engine is trained using this information to improve the accuracy and personalization of its prediction of the target temperature parameter.

[0125] The system first determines the target temperature parameters through the fourth intelligent engine based on the target content, content acquisition requirements and content type, and configures the first intelligent engine to generate or rewrite the content. The generated or rewritten multiple candidate contents are displayed to the user for selection.

[0126] The user selects one or more output contents that he or she considers satisfactory from the displayed candidate contents. The system records the user's selection behavior and collects the output contents selected by the user and the corresponding target contents.

[0127] The system uses similarity calculation algorithms (such as cosine similarity, Jaccard similarity, etc.) to calculate the similarity between the output content selected by the user and the target content. This similarity index reflects the user's acceptance of the degree of content change, and can indirectly reflect the user's preference for temperature parameters.

[0128] Based on the similarity calculation results and the output content selected by the user, the system attempts to infer the first temperature parameter that the user may prefer. This can be done by analyzing the similarity distribution of the content generated under different temperature parameters and finding the temperature parameter range that is closest to the similarity of the content selected by the user.

[0129] The output content, target content, similarity calculation results and the first temperature parameter obtained by reverse deduction selected by the user are used as training samples and input into the training module of the fourth intelligent engine. The training module uses these samples to iteratively train the fourth intelligent engine to optimize its model parameters for predicting the target temperature parameter. During the training process, the system may use a variety of machine learning optimization algorithms (such as gradient descent, stochastic gradient descent, Adam, etc.) to improve the convergence speed and prediction accuracy of the model.

[0130] During the training process, the system can regularly evaluate the performance of the fourth intelligent engine, such as prediction accuracy, recall rate, F1 score and other indicators. Based on the evaluation results, the system decides whether to continue training or terminate training early to avoid problems such as overfitting. After the training is completed, the system updates the model parameters of the fourth intelligent engine so that the updated model can be used in future content acquisition tasks.

[0131] By introducing a user feedback mechanism, the user's selection behavior of the output content is used to infer the first temperature parameter preferred by the user, and the fourth intelligent engine is trained based on this information. This method not only improves the accuracy of the fourth intelligent engine in predicting the target temperature parameters, but also enhances its personalization, enabling it to better adapt to the needs and preferences of different users. Through continuous iterative training and optimization, the performance of the fourth intelligent engine will continue to improve, providing users with a better content acquisition experience.

[0132] Figure 9AA flowchart of a method for training a fourth intelligent engine according to another embodiment of the present disclosure is schematically shown. Figure 9B The logical architecture diagram of training the fourth intelligent engine according to an embodiment of the present disclosure is schematically shown.

[0133] According to the embodiments of the present disclosure, Figure 9A As shown, for example, the fourth intelligent engine is trained by operations S910~S920.

[0134] In operation S910, in response to the user adjusting the initial temperature parameter of the first smart engine to a custom temperature parameter, a difference between the custom temperature parameter and the initial temperature parameter is input to the fourth smart engine to obtain a second temperature parameter.

[0135] In some embodiments, a user-defined temperature parameter adjustment function is further introduced, and the fourth intelligent engine is trained accordingly to optimize its ability to predict target temperature parameters.

[0136] The system first generates initial candidate content by configuring initial temperature parameters to the first intelligent engine based on the target content, content acquisition requirements and content type.

[0137] The system provides an interface to allow users to view the generated candidate content and to customize the initial temperature parameters of the first intelligent engine. Users can gradually adjust the temperature parameters according to their preferences and desired effects until they achieve satisfactory content generation results.

[0138] For each adjustment made by the user, the system records the difference between the customized temperature parameter and the initial temperature parameter. This difference is input into the fourth intelligent engine, and combined with other relevant factors (such as content type, user history behavior, etc.), the second temperature parameter is calculated through the model. This second temperature parameter is the temperature parameter predicted by the fourth intelligent engine based on the user's current adjustment behavior and may be more in line with the user's preferences.

[0139] In operation S920, in response to the user repeatedly adjusting the temperature parameter, a difference between two consecutive adjustments of the second temperature parameter is less than a preset threshold.

[0140] The system allows users to repeatedly adjust the temperature parameters and displays the content generation effect after each adjustment in real time. During the user's repeated adjustments, the system continuously monitors the difference between the second temperature parameters obtained from two consecutive adjustments. When this difference is less than the preset threshold, the system considers that the user's adjustment behavior has stabilized, that is, the user has found a relatively satisfactory temperature parameter setting.

[0141] like Figure 9BAs shown in the figure, the system uses all customized temperature parameters, initial temperature parameters, differences, and the final stable second temperature parameters during the user's entire adjustment process as training samples. These training samples are input into the training module of the fourth intelligent engine to optimize the parameters and prediction logic of the model. Through continuous iterative training, the fourth intelligent engine will be able to more accurately predict the temperature parameters that users may prefer, thereby improving the personalization and satisfaction of content generation.

[0142] The fourth intelligent engine may also be synchronously trained in combination with the method of calculating the similarity between the output content selected by the user and the target content in the above embodiment to improve the training effect.

[0143] During and after training, the system will evaluate the performance of the fourth intelligent engine, including indicators such as prediction accuracy and response time. Based on the evaluation results, the system may further optimize and adjust the model to improve its overall performance and stability.

[0144] By introducing the function of user-defined temperature parameter adjustment and training the fourth intelligent engine based on the user's adjustment behavior, a more accurate and personalized prediction of the target temperature parameters is achieved. This method not only improves the flexibility and diversity of content generation, but also enhances the interactivity between users and the system, allowing users to customize the content generation effect according to their preferences and needs. Through continuous iterative training and optimization, the performance of the fourth intelligent engine will continue to improve, providing users with a more high-quality and personalized content generation experience.

[0145] Based on the above method, the present disclosure also provides a content acquisition device. Figure 10 The content acquisition device is described in detail.

[0146] Figure 10 The structural block diagram of the content acquisition device according to an embodiment of the present disclosure is schematically shown.

[0147] like Figure 10 As shown, the content acquisition device 1000 of this embodiment includes, for example, a determination module 1010 and a processing module 1020. The content acquisition device 1000 can execute the above reference Figures 2 to 9B The method described is used to achieve adaptive adjustment of content generation effects.

[0148] Specifically, the determination module 1010 is used to respond to the content acquisition request to the first intelligent engine with the target content and determine the target temperature parameter for the target content based on the target content. In one embodiment, the determination module 1010 can be used to perform the operation S210 described above, which will not be repeated here.

[0149] The processing module 1020 is used to configure the target temperature parameter to the first intelligent engine so that the first intelligent engine processes the target content with the target temperature parameter. The first intelligent engine is used to output the content to be acquired, and the target temperature parameter is used to adjust the predictability of the content to be acquired. In one embodiment, the processing module 1020 can be used to perform the operation S220 described above, which will not be repeated here.

[0150] The second adjustment module 1030 is used to: In one embodiment, the second adjustment module 1030 may be used to perform the operation S230 described above, which will not be described in detail herein.

[0151] It is understandable that the determination module 1010 and the processing module 1020 can be combined in one module for implementation, or any one of the modules can be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present disclosure, at least one of the determination module 1010 and the processing module 1020 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application specific integrated circuit (ASIC), or can be implemented in hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or in a proper combination of software, hardware and firmware. Alternatively, at least one of the determination module 1010 and the processing module 1020 can be at least partially implemented as a computer program module, and when the program is run by a computer, the function of the corresponding module can be executed.

[0152] Figure 11 A block diagram of an electronic device suitable for implementing a content acquisition method according to an embodiment of the present disclosure is schematically shown.

[0153] like Figure 11 As shown, the electronic device 1100 according to an embodiment of the present disclosure includes a processor 1101, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1102 or a program loaded from a storage part 1108 to a random access memory (RAM) 1103. The processor 1101 may include, for example, a general-purpose microprocessor (such as a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (for example, an application-specific integrated circuit (ASIC)), etc. The processor 1101 may also include an onboard memory for caching purposes. The processor 1101 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0154] For example, the processor 1101 is used to execute a content acquisition method, which includes: in response to a content acquisition request to the first intelligent engine with target content, determining a target temperature parameter for the target content based on the target content. The target temperature parameter is configured to the first intelligent engine so that the first intelligent engine processes the target content with the target temperature parameter. The first intelligent engine is used to output the content to be acquired, and the target temperature parameter is used to adjust the predictability of the content to be acquired.

[0155] In RAM 1103, various programs and data required for the operation of electronic device 1100 are stored. Processor 1101, ROM 1102 and RAM 1103 are connected to each other through bus 1104. Processor 1101 performs various operations of the method flow according to the embodiment of the present disclosure by executing the program in ROM 1102 and / or RAM 1103. It should be noted that the program can also be stored in one or more memories other than ROM 1102 and RAM 1103. Processor 1101 can also perform various operations of the method flow according to the embodiment of the present disclosure by executing the program stored in the one or more memories.

[0156] According to an embodiment of the present disclosure, the electronic device 1100 may further include an input / output (I / O) interface 1105, which is also connected to the bus 1104. The electronic device 1100 may further include one or more of the following components connected to the I / O interface 1105: an input portion 1106 including a keyboard, a mouse, etc., the input portion 1106 is, for example, an input component for receiving target content; an output portion 1107 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker; a storage portion 1108 including a hard disk, etc.; and a communication portion 1109 including a network interface card such as a LAN card, a modem, etc. The communication portion 1109 performs communication processing via a network such as the Internet. A drive 1110 is also connected to the I / O interface 1105 as needed. A removable medium 1111, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1110 as needed, so that a computer program read therefrom is installed into the storage portion 1108 as needed.

[0157] The present disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the content acquisition method according to the embodiment of the present disclosure is implemented.

[0158] According to an embodiment of the present disclosure, a computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an apparatus, or a device. For example, according to an embodiment of the present disclosure, a computer-readable storage medium may include the ROM 1102 and / or RAM 1103 described above and / or one or more memories other than ROM 1102 and RAM 1103.

[0159] The embodiment of the present disclosure also includes a computer program product, which includes a computer program, and the computer program contains program code for executing the method shown in the flowchart. When the computer program product is run in a computer system, the program code is used to enable the computer system to implement the content acquisition method provided by the embodiment of the present disclosure.

[0160] The above functions defined in the system / device of the embodiment of the present disclosure are performed when the computer program is executed by the processor 1101. According to the embodiment of the present disclosure, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0161] In one embodiment, the computer program may rely on tangible storage media such as optical storage devices, magnetic storage devices, etc. In another embodiment, the computer program may also be transmitted and distributed in the form of signals on a network medium, and downloaded and installed through the communication part 1109, and / or installed from a removable medium 1111. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0162] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 1109, and / or installed from the removable medium 1111. When the computer program is executed by the processor 1101, the above functions defined in the system of the embodiment of the present disclosure are performed. According to the embodiment of the present disclosure, the system, device, means, module, unit, etc. described above can be implemented by a computer program module.

[0163] According to an embodiment of the present disclosure, the program code for executing the computer program provided by the embodiment of the present disclosure can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level process and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, Java, C++, python, "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect through the Internet).

[0164] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0165] It will be appreciated by those skilled in the art that the features described in the various embodiments and / or claims of the present disclosure may be combined and / or combined in a variety of ways, even if such combinations and / or combinations are not explicitly described in the present disclosure. In particular, the features described in the various embodiments and / or claims of the present disclosure may be combined and / or combined in a variety of ways without departing from the spirit and teachings of the present disclosure. All of these combinations and / or combinations fall within the scope of the present disclosure.

[0166] The embodiments of the present disclosure are described above. However, these embodiments are only for illustrative purposes and are not intended to limit the scope of the present disclosure. Although the embodiments are described above separately, this does not mean that the measures in the various embodiments cannot be used in combination to advantage. The scope of the present disclosure is defined by the attached claims and their equivalents. Without departing from the scope of the present disclosure, those skilled in the art may make a variety of substitutions and modifications, which should all fall within the scope of the present disclosure.

Claims

1. A content acquisition method, comprising: In response to a content acquisition request to the first intelligent engine with target content, determining a target temperature parameter for the target content based on the target content; configuring the target temperature parameter to the first intelligent engine so that the first intelligent engine processes the target content with the target temperature parameter; The first intelligent engine is used to output the content to be acquired, and the target temperature parameter is used to adjust the predictability of the content to be acquired.

2. The method according to claim 1, wherein: The step of determining a target temperature parameter for the target content based on the target content includes: Acquire content acquisition requirements corresponding to the target content; Based on the content acquisition requirement, determining a target temperature parameter for the target content; The content acquisition requirement is a content acquisition intention, and the content acquisition intention includes a content generation intention and a content rewriting intention. The content generation intention represents content generation based on the target content, and the content rewriting intention represents rewriting the target content.

3. The method according to claim 2, wherein: The obtaining of the content acquisition requirement corresponding to the target content includes: Inputting the target content into a second intelligent engine to obtain a content acquisition requirement corresponding to the target content; Among them, the second intelligent engine is used to infer the content acquisition intention based on the target content to determine that the content acquisition requirement is one of the content generation intention and the content rewriting intention.

4. The method according to claim 1, wherein: The step of determining a target temperature parameter for the target content based on the target content includes: Obtaining the content type corresponding to the target content; Based on the content type, determining a target temperature parameter for the target content; The content type is a text type of the content to be acquired, and the text type of the content to be acquired includes a first text type and a second text type, and the predictability of the content corresponding to the first text type is greater than the predictability of the content corresponding to the second text type.

5. The method according to claim 4, wherein: Obtaining the content type corresponding to the target content includes: Inputting the target content into a third intelligent engine to obtain a content type corresponding to the target content; The third intelligent engine is used to infer the text type of the content to be acquired based on the target content to determine that the content type is one of the first text type and the second text type.

6. The method according to claim 1, wherein: The step of determining a target temperature parameter for the target content based on the target content includes: Obtaining content acquisition requirements and content types corresponding to the target content; Inputting the content acquisition requirement and the content type into a fourth intelligent engine to obtain a target temperature parameter corresponding to the target content; Wherein, the fourth intelligent engine is used to predict or adjust the target temperature parameter according to the content acquisition requirement and the content type; The content acquisition requirement is a content acquisition intention, and the content acquisition intention includes a content generation intention and a content rewriting intention. The content generation intention represents content generation based on the target content, and the content rewriting intention represents rewriting the target content. The content type is a text type of the content to be acquired, and the text type of the content to be acquired includes a first text type and a second text type. The predictability of the content corresponding to the first text type is greater than the predictability of the content corresponding to the second text type.

7. The method according to claim 6, wherein: Also includes: In response to a user selecting a target output content corresponding to the target content from a plurality of output contents, determining a first temperature parameter corresponding to the target content based on a similarity between the target output content and the target content; The target content and the first temperature parameter are used as training samples of the fourth intelligent engine.

8. The method according to claim 6, wherein: Training the fourth intelligent engine includes: In response to the user adjusting the initial temperature parameter of the first intelligent engine to a custom temperature parameter, inputting a difference between the custom temperature parameter and the initial temperature parameter to the fourth intelligent engine to obtain a second temperature parameter; In response to the user repeatedly adjusting the temperature parameter until a difference between two adjacent adjustments of the second temperature parameter is less than a preset threshold.

9. A content acquisition device, comprising: a determination module, configured to respond to a content acquisition request to the first intelligent engine with target content and determine a target temperature parameter for the target content based on the target content; a processing module, configured to configure the target temperature parameter to the first intelligent engine, so that the first intelligent engine processes the target content with the target temperature parameter; The first intelligent engine is used to output the content to be acquired, and the target temperature parameter is used to adjust the predictability of the content to be acquired.

10. An electronic device, comprising: one or more processors; An input component for receiving target content; The processor is used to execute a content acquisition method, and the content acquisition method includes: In response to a content acquisition request to the first intelligent engine with the target content, determining a target temperature parameter for the target content based on the target content; configuring the target temperature parameter to the first intelligent engine so that the first intelligent engine processes the target content with the target temperature parameter; The first intelligent engine is used to output the content to be acquired, and the target temperature parameter is used to adjust the predictability of the content to be acquired.

Images