Content input method and device, equipment and storage medium

Abstract

The invention discloses a content input method and device, equipment and a storage medium, and relates to the technical field of input methods.According to the method, after one-time content output, the content in a first display area corresponding to an input function can be automatically updated, so that the needed content can be continuously selected, and the user experience is improved. According to the method and the device, the effect of continuously outputting the content to the target application can be realized only by simple triggering operation without other input operation or updating operation, so that the content input efficiency is greatly improved. Moreover, the method can identify the intention of the user according to the page scene when the input function is called, actively generate the content according to the intention and automatically display the content to the user, so that the content can be obtained without user operation, the content input efficiency is improved, and meanwhile, an available instruction can be matched for the user according to the content selected by the user, and the user experience is improved. The user is assisted to optimize the content, the complexity of content optimization is reduced, and the efficiency of content input is improved.

Description

Technical Field

[0001] The present application relates to the technical field of input methods, and provides a content input method, apparatus, device, and storage medium. Background Art

[0002] Artificial intelligence (AI) is now widely used in daily life. For example, a language model is a deep learning model based on artificial intelligence (AI). It is trained to understand and generate human language. A large-scale language model (LLM), also known as a large model, is a language model with a large parameter scale and complexity, often reaching millions to billions of parameters. Through a complex training process, large models can learn more subtle patterns and regularities, and have stronger generalization and expressive capabilities. After the large model is trained, by inputting prompts into the model, the model can generate the desired content based on the prompts.

[0003] However, currently, when using a large model, one typically asks the model questions through a conversation, waits for the model to output content, and then copies it to the scene where the content generated by the large model is needed. For example, when chatting with a social partner on a social platform, one needs to switch to the large model's conversation interface to generate the content generated by the large model, copy the content generated by the large model, switch to the chat interface, enter the content into the box, and then send it to the social partner. This operation is relatively complex, which increases the complexity of using the content results of the large model and reduces the efficiency of content input. Summary of the Invention

[0004] Embodiments of the present application provide a content input method, apparatus, device, and storage medium for improving content input efficiency.

[0005] In one aspect, a content input method is provided, the method comprising:

[0006] Presenting a first interface, the first interface including a first display area corresponding to an input function and a second display area corresponding to a target application, the first display area presenting at least one first content generated by a content generation model;

[0007] In response to a triggering operation on a first target content among the at least one first content, presenting the first target content in the second display area, and updating and presenting at least one second content in the first display area, where the at least one second content is generated by the content generation model according to the content intent corresponding to the first target content;

[0008] In response to a triggering operation on a second target content among the at least one second content, the second target content is presented in the second display area.

[0009] In one aspect, a content input device is provided, the device comprising:

[0010] an interface presenting unit, configured to present a first interface, the first interface comprising a first display area corresponding to an input function and a second display area corresponding to a target application, the first display area presenting at least one first content generated by a content generation model;

[0011] a content updating unit, configured to, in response to a triggering operation on a first target content in the at least one first content, present the first target content in the second display area, and update and present at least one second content in the first display area, where the at least one second content is generated by the content generation model according to the content intent corresponding to the first target content;

[0012] The content updating unit is further configured to present the second target content in the second display area in response to a triggering operation on a second target content in the at least one second content.

[0013] In one aspect, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any one of the above methods when executing the computer program.

[0014] In one aspect, a computer storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps of any of the above methods are implemented.

[0015] In one aspect, a computer program product is provided, comprising a computer program stored in a computer-readable storage medium. A processor of a computer device reads the computer program from the computer-readable storage medium and executes the computer program, causing the computer device to perform the steps of any of the above methods.

[0016] In an embodiment of the present application, when the target application calls an input function, the first display area corresponding to the input function can present at least one first content generated by the content generation model, and when a trigger operation is performed on the first target content, the first target content can be directly output to the target application so that the first target content is presented on the second display area of ​​the target application. In addition, the trigger operation will also trigger automatic update of the content in the first display area, and the user can continue to perform trigger operations on the content of the first display area, so that the second target content can be directly output to the target application so that the second target content is presented on the second display area of ​​the target application.

[0017] Through this method, by directly using the content generation model in the input function to assist in content input, there is no need to frequently switch between the AI ​​interface and the input method interface when inputting content, which reduces the complexity of using the content results of the large model. In addition, this method can also automatically update the content displayed in the first display area after one content output, so that the required content can be continuously selected in the first display area. Without the need for other input operations or update operations, only a simple trigger operation is required to achieve the effect of continuously outputting content to the target application, greatly improving the efficiency of content input. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0019] Figure 1 Schematic diagram of application scenarios provided by embodiments of the present application;

[0020] Figure 2 This is an example diagram of combining input function with large model technology in related technology;

[0021] Figure 3 A flowchart of a content input method provided in an embodiment of the present application;

[0022] Figures 4A to 4C A schematic diagram of an interface provided in an embodiment of the present application;

[0023] Figure 5 A schematic diagram of a flow chart of a presentation style instruction provided in an embodiment of the present application;

[0024] Figure 6 A schematic diagram of the intent recognition process provided in an embodiment of the present application;

[0025] Figures 7A to 7D Another interface diagram provided in an embodiment of the present application;

[0026] Figures 8A and 8B Another interface diagram provided in an embodiment of the present application;

[0027] Figures 9A to 9C A schematic diagram showing a first display area provided in an embodiment of the present application;

[0028] Figure 10 Another flowchart of the content input method provided in an embodiment of the present application;

[0029] Figures 11A to 11C A schematic diagram of another interface provided in an embodiment of the present application;

[0030] Figure 12 A schematic diagram of a processing method provided in an embodiment of the present application;

[0031] Figure 13 Another example diagram of combining a large model with an input function in the related art;

[0032] Figure 14 A schematic diagram of another flow chart of the content input method provided in an embodiment of the present application;

[0033] Figures 15A to 15D A schematic diagram of an interface provided in an embodiment of the present application;

[0034] Figure 16 A flowchart of intelligent input assistance for content generation models provided in an embodiment of the present application;

[0035] Figure 17 This is an example diagram of intelligent input assistance provided by an embodiment of the present application;

[0036] Figure 18 This is another example of combining a large model with an input function in the related art;

[0037] Figure 19 A schematic diagram of a custom instruction provided in an embodiment of the present application;

[0038] Figure 20 A schematic diagram of the interaction logic between the user, input function, and content generation model provided in an embodiment of the present application;

[0039] Figure 21 A schematic diagram of the structure of a content input device provided in an embodiment of the present application;

[0040] Figure 22 A schematic diagram of the structure of a computer device provided in an embodiment of the present application;

[0041] Figure 23 A schematic diagram of the structure of another computer device to which the embodiments of the present application are applied. DETAILED DESCRIPTION

[0042] To make the objectives, technical solutions and advantages of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application. Unless there is a conflict, the embodiments and features in the embodiments of this application can be combined with each other in any way. Moreover, although the logical order is shown in the flowchart, in some cases, the steps shown or described can be performed in an order different from that here. In the description of the embodiments of this application, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" can explicitly or implicitly include one or more of such features.

[0043] It should be understood that in the embodiments of the present application, "at least one" refers to one or more, and "plurality" refers to two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can mean: a, b, c, a and b, a and c, b and c, or a, b, and c, where a, b, and c can be single or multiple.

[0044] It is understandable that in the following specific implementation methods of this application, user-related data, such as user behavior data, social data, or historical interaction data with social objects, etc., are involved. When the various embodiments of this application are applied to specific products or technologies, relevant permissions or consents need to be obtained, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0045] To facilitate understanding of the technical solutions provided in the embodiments of the present application, some key terms used in the embodiments of the present application are explained here:

[0046] 1. Content Generation Model

[0047] The content generation model can also be called a large model, a generative model, a foundation model, or a large language model. The content generation model in the embodiment of the present application is a generative deep learning model. Such a model can generate the required content based on the prompt content. Its characteristics are that it provides powerful expression and learning capabilities based on the huge scale of model parameters, and the large model usually learns a variety of different natural language processing (NLP) tasks, so that the model can learn a wider and more generalized language understanding ability, and has excellent capabilities in various NLP tasks, such as text classification, sentiment analysis, summary generation and translation, etc. The large model can be used in multiple application fields such as writing, chat robots, virtual assistants, voice assistants, automatic translation, etc.

[0048] One possible content generation model could be the Generative Pre-trained Transformer (GPT) model. Transformer is a deep learning architecture, a neural network structure based on self-attention. The GPT model's training dataset is large, and the resulting model is large. In layman's terms, the GPT model is like a person who has read thousands of books and possesses a comprehensive understanding of everything from astronomy to geography.

[0049] 2. Type directives, adjustment directives, and style directives

[0050] Instructions refer to instructions input into the content generation model, which are used to assist the content generation model in understanding the user's needs for content, thereby generating content that better meets user needs. Different types of instructions correspond to needs of different dimensions.

[0051] The type instruction is used to indicate the required content type. The content type may include text, image, or translation content. The text content type may also be further divided into short sentences, paragraphs, or articles.

[0052] Adjustment instructions can also be called intention instructions, which are used to represent the need to adjust the content output by the content generation model. For example, it is the need to further adjust the currently generated content. Adjustment instructions can be a simple sentence that the user gives to the content generation model to describe the output task, such as adding expressions to the current content, modifying the style, layout or language of the current content, etc. Exemplary adjustment instructions are, for example, "euphemistic expression" (instructing to use a more euphemistic tone of expression) or "using English" (instructing to change the language), etc.

[0053] Style instructions, also known as pattern instructions, are used to indicate the style of the content generated by the content generation model. They can also be understood as a style pattern. A style instruction can be a task description in a specific output format (e.g., a continuous input pattern). The content generation model will generate the content required by the user based on the style pattern, such as a boasting mode, a snarky mode, or a flirting mode.

[0054] 3. Content presentation sub-area, content input sub-area, and content editing sub-area

[0055] The "area" referred to in the embodiments of the present application refers to different areas corresponding to different applications (such as input functions or target applications) in the current display interface, and the "sub-area" is a partial area within the area corresponding to an application. Different sub-areas are used to implement different functions depending on the configuration.

[0056] The content presentation subarea is used to present social content or published content sent to social contacts. "Sent" here means that the content has been triggered. This trigger can be explicit by the user, such as through a send control displayed on a page or a send button on a device, or implicit by the user, such as by calling an application's send interface, or other possible triggering methods. For example, the content presentation subarea can be the record presentation area in a chat interface or the comment presentation area on a social platform.

[0057] The content input subarea is used to display the content currently being edited by the input function. This subarea can also be referred to as the input box area. While the content in the content presentation subarea is already sent or published and thus uneditable, the content in the content input subarea is currently being edited and is therefore editable.

[0058] The content editing subarea is used for editing content, that is, it allows users to perform editing operations to input user content. For example, the content editing subarea may include a keyboard or handwriting input pad. For example, the content editing subarea may be used to display a keyboard for implementing input functions, through which users can enter their desired content.

[0059] 4. Input function

[0060] The input function refers to the ability to input content in the target application. The input function can be a capability of the device itself, for example, it can be an input capability implemented by the input method provided in the device's own operating system, or it can be a capability provided by an input method application installed in the device.

[0061] The embodiments of the present application relate to AI and machine learning technologies, and are designed based on machine learning (ML), NLP technology, and pre-training model technology in AI.

[0062] Artificial intelligence (AI) refers to the theories, methods, techniques, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, to perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that seeks to understand the essence of intelligence and produce new intelligent machines that can respond in a manner similar to human intelligence. AI also involves studying the design principles and implementation methods of various intelligent machines, enabling them to possess the capabilities of perception, reasoning, and decision-making.

[0063] Artificial intelligence (AI) technology is a comprehensive discipline encompassing a wide range of fields, encompassing both hardware and software technologies. Foundational AI technologies generally include sensors, specialized AI chips, cloud computing, distributed storage, big data processing, pre-trained models, operating / interaction systems, and mechatronics. Pre-trained models, also known as large models or basic models, can be fine-tuned and widely applied to downstream tasks across various AI disciplines. AI software technologies primarily encompass computer vision, speech processing, natural language processing, and machine learning / deep learning.

[0064] The content input method in this application involves some processing of the input content or the content selected by the user. For example, based on the user's input content, a pre-trained model (i.e., the content generation model involved later) is used to generate the corresponding content. For another example, the user's input content is used to identify the intent of the user to accurately generate the content required by the user based on the intent. These processing processes may involve machine learning, NLP, pre-trained model technology, etc.

[0065] Machine learning is a multidisciplinary field that encompasses probability theory, statistics, approximation theory, convex analysis, and algorithmic complexity theory. It specifically studies how computers can simulate or implement human learning behaviors to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is at the core of artificial intelligence and the fundamental way to make computers intelligent. Its applications span all areas of artificial intelligence. Machine learning and deep learning typically include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and self-learning. Pretrained models are the latest development in deep learning, integrating these techniques.

[0066] Natural language processing (NLP) is a key area of ​​research in computer science and artificial intelligence. It studies theories and methods that enable effective communication between humans and computers using natural language. Natural language processing (NLP) involves natural language, the language we use daily, and is closely related to linguistics. It also involves computer science and mathematics. Pre-trained models, a key technology for model training in artificial intelligence, are derived from large language models in the field of NLP. After fine-tuning, large language models can be widely applied to downstream tasks. Natural language processing technologies generally include text processing, semantic understanding, machine translation, robotic question answering, knowledge graphs, and other technologies.

[0067] As in the embodiments of the present application, machine learning technology and NLP technology (or pre-trained model technology) can be used to train a deep learning model that can be used to generate new content based on at least one of the page scene, input content or instructions, and the trained deep learning model can be applied to the input function to implement auxiliary writing, translation or generation of corresponding images in the input function.

[0068] With the research and advancement of artificial intelligence technology, artificial intelligence technology has been studied and applied in many fields, such as common smart homes, smart wearable devices, virtual assistants, smart speakers, smart marketing, unmanned driving, autonomous driving, drones, digital twins, virtual humans, robots, artificial intelligence generated content (AIGC), conversational interaction, smart medical care, smart customer service, game AI, etc. It is believed that with the development of technology, artificial intelligence technology will be applied in more fields and play an increasingly important role.

[0069] The following is a brief introduction to the design concept of the embodiment of this application:

[0070] At present, the application of big model technology is becoming increasingly widespread. However, at present, big models are usually implemented in a dialogue mode, that is, you can enter the topic content in the dialogue interface provided by the big model, wait for the content results output by the big model, and then copy the content results to other usage scenarios. For example, when chatting with social objects on a social platform, you need to switch to the dialogue interface of the big model to generate the content results generated by the big model, copy the content results generated by the big model and switch to the chat interface, input the content results into the box and then send it to the social object. The operation is relatively complex. In order to solve the inconvenience caused by this usage method, some input method applications have begun to consider using it in combination with big models, but the current method still has the problem of low efficiency.

[0071] Based on this, an embodiment of the present application provides a content input method, in which content input is assisted by directly using a content generation model in the input function, so that when inputting content, there is no need to frequently switch between the AI ​​interface and the input method interface, thereby reducing the complexity of using the content results of the large model. In addition, the method can also automatically update the content displayed in the first display area after one content output, so that the required content can be continuously selected in the first display area. Without the need for other input operations or update operations, only a simple trigger operation is required to achieve the effect of continuously outputting content to the target application, greatly improving the efficiency of content input.

[0072] The method of the embodiment of the present application can also present multiple style instructions to provide users with multiple optional content generation styles, and can continuously generate the content required by the user based on the style instructions selected by the user, so that the content generated by the large model can be continuously obtained without issuing instructions each time, and the content can be sent to the screen for presentation through simple operations, thereby improving the efficiency of content input.

[0073] In addition, the method can also identify the user's intention to input content, such as posting comments, chatting on WeChat, or posting to Moments, based on the page scenario when the input function is called up. It can actively generate content based on the intention and automatically display it to the user, so that content can be obtained without user operation, thereby improving the efficiency of content input.

[0074] In addition, in this method, after the user selects content, adjustment instructions corresponding to the selected content can be presented to assist the user in optimizing the selected content, so that the input content better meets the user's needs and improves the accuracy of content generation.

[0075] The following briefly introduces the application scenarios to which the technical solutions of the embodiments of the present application can be applied. It should be noted that the application scenarios described below are only used to illustrate the embodiments of the present application and are not limiting. In the specific implementation process, the technical solutions provided by the embodiments of the present application can be flexibly applied according to actual needs.

[0076] The solution provided in the embodiment of the present application can be applied to the use scenario of the input function. Figure 1 As shown, it is a schematic diagram of an application scenario provided by an embodiment of the present application. In this scenario, it can include a terminal device 101, a server 102 and a server 103.

[0077] The terminal device 101 can be, for example, any device such as a mobile phone, a tablet computer (PAD), a laptop computer, a desktop computer, a smart TV, an intelligent vehicle-mounted device, and an intelligent wearable device. The terminal device 101 can be installed with a target application, and the terminal device has an input function, which can be a function provided by the terminal device when it leaves the factory (such as an input function integrated in the operating system), or it can be an input function provided by installing an input method application in the terminal device. The target application can call the input function to input content and present content, such as an instant messaging application or a social platform application, or any application that needs to call the input function. The input function can be called by the target application to input the content required by the user in the target application. The application involved in the embodiment of the present application can be a software client, or it can be a client such as a web page or a mini-program.

[0078] Server 102 and server 103 may be servers corresponding to the input function and target application, respectively. Server 102 or server 103 may be an independent physical server, or a server cluster or distributed system consisting of multiple physical servers. It may also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms, but is not limited thereto.

[0079] It should be noted that the content input method in the embodiment of the present application can be executed by the terminal device 101 alone, or can be executed jointly by the server 102 and the terminal device 101. For example, the terminal device 101 is used to implement the presentation of the front-end page, and the server 102 is used to provide back-end services for the presentation of the front-end page, such as intent recognition of the content selected by the user.

[0080] It should be noted that the content input method in each embodiment of the present application can be executed by an electronic device, which can be a terminal device 101 or a server 102. That is, the method can be executed by the terminal device 101 or the server 102 alone, or can be executed jointly by the terminal device 101 and the server 102. For example, when the method is executed jointly by the terminal device 101 or the server 120, the terminal device 101 can be used to implement the presentation of the front-end related page of the input function, and the server 102 is used to provide back-end services for the presentation of the front-end page, such as identifying the intent of the content selected by the user and calling the content generation model to generate the content to be presented.

[0081] In the embodiment of the present application, the terminal device 101, the server 102, and the server 103 can be directly or indirectly connected to each other through one or more networks. The network can be a wired network or a wireless network. For example, the wireless network can be a mobile cellular network or a Wireless Fidelity (WIFI) network. Of course, other possible networks are also possible, and the embodiment of the present application does not limit this.

[0082] It should be noted that Figure 1 The examples shown are just for illustration. In fact, the number of terminal devices and servers is not limited and is not specifically limited in the embodiments of this application.

[0083] In the embodiments of the present application, when there are multiple servers, the multiple servers can form a blockchain, with the servers being nodes on the blockchain. For example, in the content input method disclosed in the embodiments of the present application, the content involved can be stored on the blockchain. Furthermore, when there are multiple servers, the multiple servers can form a distributed system, with the servers being nodes in the distributed system. Each terminal device can be assigned a node to provide services based on its location and system pressure.

[0084] In addition, the embodiments of the present application can be applied to various scenarios, including but not limited to cloud technology, artificial intelligence, smart transportation, assisted driving and other scenarios.

[0085] The following describes the content input method provided by the exemplary embodiment of the present application in combination with the application scenarios described above and with reference to the accompanying drawings. It should be noted that the above application scenarios are only shown to facilitate understanding of the spirit and principles of the present application, and the implementation of the present application is not limited in this respect.

[0086] First, see Figure 2 The figure shows an example of combining input function with large model technology in related technology. Figure 2 In the example shown, the user is provided with the "AI writing assistance" function, that is, the large model can generate corresponding content results based on the content entered by the user. After the user selects one of the content results, the content result will be entered into the input box, and the user can edit and optimize it in the input box. However, after the content result is selected and entered into the input box, the writing assistance panel will be retracted and switched to the input keyboard. If the user wants to continue using the model's writing assistance function, they need to re-enter the interface, which is very complicated and inefficient in content input.

[0087] Based on this, considering that the user may have a need for continuous input, the content input method of the embodiment of the present application provides a function that can input continuously. Figure 3As shown, a flow chart of a content input method provided in an embodiment of the present application is shown. Taking a terminal device (or a client of an input function in a terminal device) as an example, the specific implementation process of the method is as follows S31 to S33:

[0088] S31: Presenting a first interface, the first interface including a first display area corresponding to an input function and a second display area corresponding to a target application, the first display area presenting at least one first content generated by a content generation model.

[0089] In an embodiment of the present application, the input function can be a function provided by the terminal device when it leaves the factory (such as an input function integrated in the operating system), or it can be an input function provided to the terminal device by installing an input method application in the terminal device. The target application can be any application that needs to call the input function, such as a social platform application, an instant messaging application, a shopping application, a writing application, a video application, etc. The first interface can specifically refer to the interface presented when the target application calls the input function, the first display area is the area corresponding to the input function, and is used to present content related to the input function, and the second display area is used to present the page content of the target application, which can be a page of any scenario that may call the input function, such as a chat scene page, a comment page of a social sharing scene (such as a circle of friends or a blog, etc.), a comment page of a video scene, a social content publishing page, etc. This application does not impose any restrictions on this.

[0090] In one possible implementation, the second display area may include a content presentation sub-area for presenting social content sent to a social object. "Send" may also be referred to as "publish," "submit," etc., and "send" may be considered to have triggered "send." This triggering may be explicit by the user, such as through a send control displayed on a page or a send button on a device, or implicit, such as through an input function invoking an application's send interface, or other possible triggering methods.

[0091] In some scenarios, "Send" doesn't matter whether the content is sent successfully; as long as the content triggers "Send," it can be displayed in the content presentation subarea. Furthermore, "Sending to a social object" can include sending to a specific social object, such as a private chat (i.e., a one-on-one chat), or it can include sending to any specific social object, such as a group chat or content publishing on a social platform (e.g., Moments).

[0092] Exemplarily, the content presentation sub-area may be, for example, a record presentation area in a chat interface, in which case social content refers to content sent to social objects; or, it may be a comment presentation area on a social platform, in which case social content refers to comments; or, it may refer to a dynamic sharing page on a social platform, in which case social content refers to published dynamic content, such as the latest updates posted on a circle of friends, blogs posted in a blog application, or news posted in a news application, etc., without specific limitation.

[0093] Among them, the input function provides a function for generating content based on a content generation model. Users can use this function to generate new content, making it easier for users to directly use the content generated by the content generation model in various usage scenarios. The content generation model can be a generative large language model. Users can input prompt content or give content types, content styles, or content requirements, and the content generation model can output the content required by the user. For example, the content generation model can be a GPT model, etc., and there is no specific limitation on this. It can also be understood that the function of the content generation model is embedded in the input function, and this function can be integrated into a form of a virtual personal assistant (VPA) and provided to the user. As the name suggests, the VPA can be understood as a well-read and knowledgeable virtual person. By giving certain prompt content or content requirements, the VPA can convert the prompt content into new content that meets the content requirements, helping users to quickly input content in various scenarios.

[0094] Specifically, the content generated by the content generation model can be presented in a content generation sub-area included in the first display area. The content generation sub-area is used to present the content generated by the content generation model and can also be referred to as a content panel or a VPA content panel. For example, the content generation sub-area can be used to present a page corresponding to the content generation model. The page can be, for example, a dedicated page for the content generation model, i.e., a new page added to the input function. Alternatively, the page can be another functional page within the input function (e.g., a page containing a keyboard), i.e., a presentation area for the content generation model can be added to an existing page of the input function.

[0095] See also Figure 4A The figure shows a schematic diagram of the first interface. The upper area of ​​the first interface is the second display area corresponding to the target application, and the lower area is the first display area for the input function. In the first display area, three first contents generated by the content generation model are presented, and each first content is presented in the form of a content card. It should be noted that Figure 4AThe first interface is for illustration only. In actual scenarios, the second display area and the first display area can be distributed in other ways, such as with the second display area at the bottom and the first display area at the top, or with the second display area at the left and the first display area at the right, or with the first display area in the form of a floating window located in the second display area, etc. Furthermore, the embodiment of the present application does not limit the number of contents in the first display area, and can be, for example, 1, 2, 3, or 10, etc., and can be configured according to actual conditions, or the user can set the number of contents generated or presented by the content generation model each time.

[0096] In the embodiment of the present application, the first interface generally refers to any interface presented by the terminal device, and the first display area corresponding to the input function presented in the first interface can be in any stage of the input function usage content generation model.

[0097] In an embodiment of the present application, at least one first content presented in the first display area is content generated by a content generation model. The content generation model can generate content based on at least one of the information such as the page scene of the second display area, the user's input content, and the user's instructions.

[0098] In one embodiment, the function of the content generation model can be deployed in the client corresponding to the input function. When a trigger operation of calling the content generation model to generate content is detected, the content generation model in the client can be called to generate at least one corresponding first content to be presented in the first display area.

[0099] In another embodiment, the content generation model functionality can be deployed on a server, which can be a backend server corresponding to the input functionality or a dedicated server for the content generation model. Upon detecting a trigger operation to invoke the content generation model to generate content, a service call request can be initiated to the server. The server then generates at least one corresponding first content according to the service call request and returns it to the client of the input functionality for presentation in the first display area.

[0100] S32: In response to a triggering operation on a first target content in at least one first content, the first target content is presented in the second display area, and at least one second content is updated and presented in the first display area, where the at least one second content is generated by a content generation model according to the content intent corresponding to the first target content.

[0101] The first target content may be one of the at least one first content, i.e., the content that the user wants to output to the target application. The first content may be any type of content, including but not limited to one or more combinations of text, images, emoticons, and translation results.

[0102] In a possible implementation, when the second display area includes a content presentation sub-area, presenting the first target content in the second display area may include presenting the first target content in the content presentation sub-area. Presenting the first target content in the content presentation sub-area may also be referred to as "on-screen."

[0103] For example, see Figure 4B The figure shows the interface changes after the first target content is triggered. The second display area includes a content input sub-area (i.e. Figure 4B The input box area of ​​the chat interface shown) and the content presentation sub-area (i.e. Figure 4B The first display area of ​​the input function includes three first contents generated by the content generation model, namely, "When we look at each other, will your heart also say welcome to me?" When a trigger operation is performed on the first target content, the first target content can be presented in the content presentation sub-area, such as Figure 4B In the record presentation area of ​​the chat interface shown. That is, in Figure 4B In the chat scenario, when the first target content is triggered, the first target content can be directly sent to the chat object. Compared with the method in the related art that requires entering the selected content into the input box and then clicking the send control to send it to the chat object, the method of directly sending the first target content after the operation in the embodiment of the present application can reduce the operation steps when the user sends the content, thereby increasing the rate of sending the content.

[0104] It should be noted that although Figure 4B The chat scenario is used as an example to illustrate, but the above-mentioned direct sending method is also applicable in other scenarios, that is, the triggering operation performed by the user on the first target content can be understood as a confirmation of the first target content. According to different usage scenarios, the response logic brought about by the confirmation of the first target content may be different. After the triggering operation on the first target content, the first target content can be processed according to the logic to be executed after the content confirmation of the current scenario. For example, for the chat scenario, confirming the first target content means sending the first target content to the chat object, for the comment scenario, confirming the first target content means publishing the first target content as comment content, for the status content publishing scenario, confirming the first target content means publishing the first target content as status content, and for the search scenario, confirming the first target content means using the first target content as a search index to perform a search operation. The specific examples are not listed one by one.

[0105] In the embodiment of the present application, the triggering operation on the first target content can be any one of a click operation, a long press operation, a drag operation, or a drag operation in a specified direction, or it can be an operation on a physical button included in the device, or it can be other gesture operations, and the embodiment of the present application does not limit this. Among them, the long press operation refers to the operation of pressing the first target content for a time period exceeding a certain time threshold, and the drag operation refers to the operation of dragging the first target content from the original position to a target position, for example Figure 4B Drag the first target content from its original location to the content presentation sub-area.

[0106] Generally speaking, in the related art, the triggering operation of the first target content will input the first target content into the input box. Therefore, the triggering operation of directly presenting the first target content in the content presentation sub-area can be the same as or different from the operation of inputting the first target content into the input box. For example, if the first target content is clicked, the first target content is input into the input box. Then, the triggering operation of directly presenting the first target content in the content presentation sub-area can also be a click operation, that is, different from the method in the related art, after the click operation, the first target content will be directly sent to the chat object and presented in the content presentation sub-area; or, the triggering operation of directly presenting the first target content in the content presentation sub-area can also be other operations different from the click operation, such as a long press operation or a drag operation, etc., so that different logic can be executed according to different user operations. The user can choose which operation to use according to their needs, thereby improving the richness of the interaction method.

[0107] In the embodiment of the present application, when a user triggers an action on the first target content, the content displayed in the first display area is also updated, that is, at least one second content is updated and presented in the first display area. The at least one second content is at least partially different from the at least one first content. It should be noted that the first content or the second content is only used for differentiation and does not limit the actual content. The at least one second content is generated by the content generation model based on the content intent corresponding to the first target content selected by the user.

[0108] Among them, the content intent corresponding to the first target content may refer to the content intent based on which the first target content is generated, and the content intent may be determined based on at least one of the user input content, instructions, or page scenes; or, the content intent corresponding to the first target content may also refer to the content intent obtained by intent recognition of the first target content; or, it may also be a combination of the aforementioned two types of content intent.

[0109] In one possible implementation, updated presentation may refer to covering at least one first content with at least one second content, that is, the first display area will no longer include at least one first content; or, updated presentation may also refer to including at least one second content and at least one first content in the first display area, but currently only at least one second content is presented, and at least one first content may be hidden and displayed. For example, at least one first content may be hidden at the top of at least one second content. By sliding the page, the at least one first content may be presented again in the content presentation page, so that the user can go back to the previous content to help the user make a decision.

[0110] One implementation method is that at least one second content may have been generated by the content generation model before the triggering operation, that is, the content generation model may generate N contents according to the original content intention, and at least one first content and at least one second content are both parts of these N contents. At least when the user selects the first target content, the content that has not been presented before may be presented in the first display area.

[0111] Another implementation is that at least one second content can be regenerated after the content generation model triggers the operation, that is, each trigger operation of the user on the first display area can be used to trigger the content generation model to regenerate content and present the regenerated content in the first display area.

[0112] In one possible embodiment, the first target content is directly presented in the content presentation sub-area, and the automatic update of the content of the first display area can be performed only in a special input mode, that is, step S32 can be performed only in this special input mode. For example, the special input mode is a continuous input mode. In the continuous input mode, each time content is output to the target application, an update of the content presented in the first display area is triggered, or each time a new instruction is issued, an update of the content presented in the first display area is triggered. Among them, the continuous input mode can also be called a burst mode, a continuous sending mode, or a quick input mode, etc., which is not limited in the embodiments of the present application.

[0113] Then, before or when executing step S32 , it is necessary to determine whether the activation condition of the continuous input mode is satisfied.

[0114] One situation is that before step S32, the conditions for starting the continuous input mode have been met, that is, the continuous input mode has been turned on, or it is currently in the continuous input mode. Then, when a trigger operation on the first target content is detected, the first target content will be directly presented in the content presentation sub-area, and the content of the first display area will be automatically updated.

[0115] Another situation is that before step S32, it is in normal input mode, that is, not in continuous input mode. Then when executing step S32, when the trigger operation on the first target content is detected, it can be determined whether the conditions for starting the continuous input mode are currently met. If so, the continuous input mode can be turned on, and the first target content will be directly presented in the content presentation sub-area, and the content of the first display area will be automatically updated.

[0116] The conditions for starting the continuous input mode may include at least one of the following conditions:

[0117] (1) A preset operation is received that triggers the content presented in the first display area. That is, when the triggering operation for the content presented in the first display area is a preset operation, it is determined that the activation condition is satisfied. Generally speaking, the preset operation can be an operation other than a conventional operation (such as a click operation). Because if a content is not selected by the user to enter the input box for editing by clicking, but is expected to be directly displayed on the screen through other operations, it can be determined that the user is currently in a scenario where the expression accuracy of the produced content is not strong, and there is a certain demand for the efficiency of displaying one or more content on the screen. At this time, it can be considered that the activation condition is satisfied and the continuous input mode is automatically entered.

[0118] For example, the preset operation may be a long press operation or a drag operation, and the drag operation may be, for example, an operation of pressing and dragging the content to the content presentation sub-area with a gesture, and there is no specific limitation on this.

[0119] (2) Receiving an activation instruction corresponding to the continuous input mode. That is, an activation instruction for entering the continuous input mode may be provided in the first display area, and the user triggers the activation instruction to enter the continuous input mode. Alternatively, the activation instruction may be triggered by a physical button of the device.

[0120] (3) Determining the content intention to enable the continuous input mode based on at least one input content. The at least one input content can be content input by the user through the input function. For example, when the user inputs "enable continuous input mode" or "I want to input faster" and other content that can have a continuous input requirement, the content intention to enable the continuous input mode is determined. In addition, in addition to the content intention reflected in the content itself, whether the user has the content intention to enable the continuous input mode can also be comprehensively determined based on information such as the input speed when the user inputs the content.

[0121] (4) Determine whether the user has the content intention to enable the continuous input mode based on at least one social content sent in the past. The at least one social content sent in the past is the content that the user has sent to the social object. By performing intent recognition on these contents, it is also possible to determine whether the user has the content intention to enable the continuous input mode. Similarly, in addition to the content intention reflected in the content itself, it is also possible to comprehensively determine whether the user has the content intention to enable the continuous input mode based on information such as the interval length or sending frequency when the user sends the content.

[0122] It should be noted that when determining the user's content intent, user-related behavioral data and chat data, etc. are involved. When the various embodiments of this application are applied to specific products or technologies, relevant permissions or consents need to be obtained, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0123] S33: In response to a triggering operation on a second target content in the at least one second content, present the second target content in the second display area.

[0124] Similar to step S32, the at least one second content item presented in the first display area can also be processed in a similar manner to the at least one first content item. That is, when an action is triggered on the second target content in the at least one second content item, it can be directly sent to the social partner and presented in the content presentation sub-area. Similarly, after the action is triggered on the second target content, the content in the first display area is also automatically updated. The content of this step is similar to the aforementioned step S22, so it will not be repeated here.

[0125] For example, Figure 4C The following is an example of a continuous input mode interface diagram. Figure 4C As shown in (1), the user long presses the second content and drags it to the chat interface, then the content can be directly sent to the social object C, and the content will be presented in the record presentation area, see Figure 4C (2) As shown, at the same time, the content presented in the first display area is automatically updated. Then, the user long presses the second content and drags it to the chat interface, and can directly send the content to the social object C, and present the content in the record presentation area, see Figure 4C (3) As shown, at the same time, the content presented in the first display area is automatically updated. Then the user repeats the previous operation, long presses the first content and drags it to the chat interface, then the content can be directly sent to the social object C, and the content is presented in the record presentation area. In actual application scenarios, by continuously repeating this operation, the content can be quickly sent to the screen. The chat interface finally presented is as follows Figure 4CAs shown in (4), each piece of continuously sent content will be presented in the record presentation area in the order in which it was sent.

[0126] It can be seen that compared with Figure 2 In the manner of the related technology shown, the content input method provided in the embodiment of the present application automatically updates the content displayed in the first display area after one content output, so that the required content can be continuously selected in the first display area. Without the need for other input operations or update operations, only a simple trigger operation is required to achieve the effect of continuously outputting content to the target application, which greatly improves the efficiency of content input.

[0127] In an embodiment of the present application, considering that the user may need to further optimize the content generated by the content generation model, in order to improve the accuracy of the content generated by the content generation model and reduce the user's operations for further optimization of the content, in the continuous input mode, a variety of style instructions can also be provided to the user. Each style instruction is used to indicate a content style of the content generated by the content generation model. The style instruction can also be called a mode instruction. In some scenarios, the style instruction can be in a special mode, such as the task description in the aforementioned continuous input mode, which is used to guide the content generation model to perform content generation in a certain style. In other words, the style instruction can also be considered as an intention guidance in the continuous input mode, which is used to guide the content generation model's content intention in this continuous input mode, so that the content generation model can continuously generate new content around this content intention.

[0128] See also Figure 5 FIG. 5 is a flow chart of presenting a style instruction according to an embodiment of the present invention, which includes steps S51 to S53:

[0129] S51: Presenting a first interface, the first interface including a first display area corresponding to an input function and a second display area corresponding to a target application, the first display area presenting at least one first content generated by a content generation model.

[0130] The introduction of step S51 can refer to the above Figure 3 The introduction of S31 part is not detailed here.

[0131] S52: In response to a triggering operation on a first target content in the at least one first content, present at least one candidate style instruction associated with the first target content in the first display area, wherein each candidate style instruction is used to indicate a content style of content generated by the content generation model.

[0132] In an embodiment of the present application, in response to a triggering operation on a first target content in at least one first content, in addition to presenting the first target content directly on the screen and updating the content of the first display area, at least one candidate style instruction associated with the first target content will be determined and presented in the first display area.

[0133] At least one candidate style instruction can be determined based on the content intent corresponding to the first target content. For example, if the first target content is about "I miss you" or "That's great," it can be determined that the user has a clear intention to express longing, love, appreciation, and praise, and may be unsatisfied. Then, a related continuous mode instruction, such as "flirting mode," "boasting mode," "complaining mode," "arguing mode," "debating mode," and so on, can be matched.

[0134] The content intent corresponding to the first target content may refer to the content intent based on which the first target content is generated, and the content intent may be determined based on at least one of the user input content, instructions, or page scenarios; or, the content intent corresponding to the first target content may also refer to the content intent obtained by identifying the intent of the first target content; or, it may also be a combination of the aforementioned two types of content intent.

[0135] In one possible implementation, see Figure 6 As shown, when obtaining the content intent corresponding to the first target content, the first intent recognition result can be obtained by performing intent recognition on the first target content that triggers the operation, or the second intent recognition result can be obtained by performing intent recognition on at least one input content, wherein each input content is the content obtained by receiving the input operation performed by the user using the input function, that is, the content edited by the user through the content editing sub-area of ​​the input function, and the aforementioned at least one first content is a content generation model that can be generated based on at least one input content; or, the page scene of the current second display area can also be collected, and then the content intent corresponding to the first target content can be determined based on at least one of the first intent recognition result, the second intent recognition result, and the page scene and other information, thereby obtaining accurate content intent.

[0136] After obtaining the content intent, at least one candidate style instruction matching the content intent corresponding to the first target content may be obtained, and the at least one candidate style instruction may be presented in the first display area.

[0137] In one possible implementation, presenting at least one candidate style instruction in the first display area may be performed only in a special input mode, that is, only when the continuous input mode is determined to be in place. Regarding whether the conditions for enabling the continuous input mode are met, please refer to the description of step S32 above and will not be elaborated on here.

[0138] See also Figure 7A The figure shows a schematic diagram of an interface displaying style instructions. The style instructions are displayed in the instruction row sub-area, which is the area where the instructions are displayed. The present embodiment of the application does not impose any specific restrictions on the number and position of the instruction row sub-areas in the first display area. These can be configured based on specific implementation circumstances, or the user can configure their number and position.

[0139] In one possible implementation, before presenting at least one style instruction, at least one candidate adjustment instruction is presented in the first display area. Each candidate adjustment instruction is used to represent the need to adjust the content output by the content generation model. It should be noted that adjustment instructions and style instructions are used to distinguish instructions in different situations. For example, a style instruction can be an instruction in a continuous input mode, which is more inclined to the main style of the content during continuous input. An adjustment instruction can be an instruction in a normal input mode. The names of the adjustment instruction and style instruction are not used to limit the content of the instruction.

[0140] Then, when at least one candidate style instruction is presented in the first display area, see Figure 7B As shown, a plurality of candidate adjustment instructions are originally presented on the first display area. After the trigger operation is performed on the first target content, at least one candidate adjustment instruction can be updated to at least one candidate style instruction. Figure 7C As shown, at least one candidate adjustment instruction and at least one candidate style instruction may also be presented in the first display area, that is, the two instructions may be presented simultaneously. The two instructions may be presented in different instruction row sub-areas or in the same instruction row sub-area, and there is no specific limitation on this. Figure 7C Specifically, the example in which the candidate style instructions are presented below the candidate adjustment instructions is taken, but in actual application, there is no restriction on the positions of the two.

[0141] In an embodiment of the present application, when it is recognized that the conditions for entering the continuous input mode are met (for example, a preset operation is performed on the presented content), at least one candidate style instruction that matches the content input in the user input box or the content intent of the upper-screen content can be presented in the content presentation page. The candidate style instructions can be provided to the user to select the main style of the content, assist in guiding the content generation model to generate content that the user needs more, improve the accuracy of the content, thereby reducing the steps for the user to further optimize the content generated by the content generation model, and improving the efficiency of content input.

[0142] S53: In response to a triggering operation on a target style instruction in at least one candidate style instruction, updating and presenting at least one third content in the first display area, where the at least one third content is generated by the content generation model according to the content style corresponding to the target style instruction.

[0143] In an embodiment of the present application, if a user triggers an operation in response to a target style instruction, indicating that the user desires to input content in this style, multiple third contents may be generated for the user based on the content style of the target style instruction selected by the user, and at least one third content may be updated and presented in the first display area. The at least one third content may be part or all of the multiple third contents.

[0144] In one possible implementation, the content generation model can generate more content per session in continuous input mode than in regular input mode. This is because the focus of continuous input mode shifts from precise content to a large, continuous amount of content. Therefore, the output of the content generation model has been upgraded to meet user demands for efficient content delivery. For example, regular input mode generates three pieces of content per session, while continuous input mode generates ten pieces of content per session.

[0145] In one possible implementation, in continuous input mode, the amount of content presented in the first display area at a time can be greater than in regular input mode. This is also because the focus of the demand in continuous input mode has shifted from precise content to a large amount of continuous content. Therefore, the amount of content presented in the first display area has been increased to improve user selectivity, allowing users to more quickly obtain content that meets their needs and satisfying their desire for efficient content delivery. For example, while the number of content presented in the first display area at a time in regular input mode is three, the number of content presented in the first display area at a time in continuous input mode can be ten.

[0146] Among them, considering that the amount of content presented in the first display area is limited, when all the content cannot be presented, part of the content can be presented in the first display area first, and the user can slide up and down to make the first display area display the remaining content.

[0147] In an embodiment of the present application, if the functionality of the content generation model can be deployed in the client corresponding to the input function, then when a user triggers a target style instruction, the target style instruction is sent to the content generation model in the client, instructing the content generation model to generate at least one corresponding third content according to the target style instruction for presentation in the first display area. If the functionality of the content generation model is deployed in a server, then when a triggering operation is detected for the target style instruction, a service call request can be initiated to the server, carrying the target style instruction. The server then generates at least one corresponding third content according to the target style instruction according to the instruction of the service call request and returns the generated content to the client of the input function for presentation in the first display area.

[0148] See also Figure 7D As shown in FIG, another interface diagram of continuous input mode is shown. Figure 7D As shown in (1), after multiple pieces of content generated by the content generation model are presented in the first display area, if a piece of content is not selected by the user to enter the input box and customized editing, but is directly pressed and dragged onto the screen by gesture, it can be determined that the user is currently in a scenario where the user does not have a strong demand for accurate expression of the produced content, but has a certain demand for the efficiency of outputting one or more pieces of content onto the screen. At this time, it can be determined that the conditions for opening the continuous input mode are met and the continuous input mode is automatically entered. Of course, the continuous input mode can also be entered in other ways.

[0149] See also Figure 7D As shown in (2), after entering the continuous input mode, style instructions that match the content intention of the content input in the user input box or the content on the upper screen selected by the user can be filtered out and displayed in the instruction bar. The user can click on one of the style instructions to trigger the content generation model to generate multiple related new content and display them in the first display area.

[0150] Among them, after entering the continuous input mode, the number of contents presented in the first display area is not limited to the three contents in the normal input mode. When the user selects a content on the screen or switches the style instruction, a new instruction can be implicitly generated and sent to the content generation model to instruct the content generation model to add multiple contents, such as adding three similar contents. Then, the content generation model can generate new content again based on the previous style instruction. If the style instruction is not triggered, the content intent can be determined based on the content entered in the user input box or the content on the screen, and new content can be generated based on this. Or, when the style instruction is triggered, new content can be generated based on the content style indicated by the style instruction, or based on the content style indicated by the style instruction combined with the identified content intent. Thus, three new contents consistent with the style of the last style instruction or the previous style instruction can be obtained and displayed in the first display area.

[0151] In one embodiment, the user can still use the above-mentioned gesture dragging method to send content directly to the screen. Sending to the screen will trigger the content generation model to generate new content. During the operation of the mode command, the user does not need to enter any custom instructions. As long as the user operates the screen, the above steps can be looped continuously. In this way, the gesture interaction of the dragging and sending steps can be looped, and the user can achieve continuous quick burst function. In another embodiment, the user can use the same method as the normal input mode, that is, the user can click on the content to enter it into the input box and then click Send to the screen.

[0152] See also Figure 7D As shown in (3), when the content generated by the content generation model cannot be fully presented, the user can slide in the first display area to present the unpresented content in the first display area, such as Figure 7D In (3), the content “My love for you will never stop, and I will give you everything without regrets” is not presented in the form of a content card, but in a different form from the conventional content card. Figure 7D As shown in (4), the content will be presented in the first display area in a conventional presentation format, and the next content will be presented in the same format as the content in the first display area. By sliding up and down to present different content, more content can be presented in a single presentation, thereby helping users make decisions.

[0153] In the embodiment of the present application, after entering the continuous input mode, it is also possible to detect whether the termination condition of the continuous input mode is met. If the termination condition is determined to be met, the continuous input mode is closed. After the input mode is closed, the operation logic of the normal input mode is executed. The termination of the continuous input mode can also be understood as the user's input intention being satisfied, that is, the input intention has been achieved, and the continuous input mode can be automatically exited and the normal input process can be returned.

[0154] In a possible implementation, the termination condition may include but is not limited to at least one of the following conditions:

[0155] (1) A trigger operation for presenting an input function in a content editing sub-area in the first display area is received, where the content editing sub-area is used to edit content. Generally speaking, when a user pulls up the content editing sub-area, it indicates that the user has a need to input or optimize the content, that is, the user has a need for content accuracy, and conversely, the need for content input efficiency is secondary to accuracy. In this case, it can be determined that the end condition of the continuous input mode is met.

[0156] For example, the triggering operation of the content editing sub-area for presenting the input function can be an operation on the keyboard control in the first display area, an operation in which the user clicks the input box to invoke the keyboard, an operation in which the basic keyboard is called, a new style instruction is entered, or a custom instruction is clicked to enter a new instruction. When it is determined that the user has completed the input of the current intention, the continuous input mode is automatically exited and the normal typing process is returned to.

[0157] It should be noted that each sub-area in the embodiment of the present application can be considered as a content display window within the application (or input function), which is used to display the corresponding content. Each sub-area can be presented simultaneously in the display area corresponding to the application (or input function), or some sub-areas can be hidden according to actual needs.

[0158] (2) Receiving a close instruction corresponding to the continuous input mode. That is, a close instruction for ending the continuous input mode may be provided in the first display area, and the user triggers the close instruction to enter the continuous input mode. Alternatively, the close instruction may be triggered by a physical button of the device.

[0159] (3) Determining the content intention of closing the continuous input mode based on at least one input content. The at least one input content can be content input by the user through the input function. For example, when the user inputs "end continuous input mode" or "input accuracy is too low" and other content indicating that the user has the need to end continuous input, it is determined that the content intention of closing the continuous input mode exists. In addition, in addition to the content intention reflected by the content itself, it is also possible to comprehensively determine whether the user has the content intention of opening the continuous input mode based on information such as the input speed when the user inputs the content. For example, if the user has not performed any operation for a long time, it indicates that the user may not need higher input efficiency at present, and the continuous input mode can be ended.

[0160] (4) Determine whether the user has the content intention to end the continuous input mode based on at least one social content sent in the past. The at least one social content sent in the past is the content that the user has sent to the social object. By performing intent recognition on these contents, it is also possible to determine whether the user has the content intention to end the continuous input mode. Similarly, in addition to the content intention reflected in the content itself, it is also possible to comprehensively determine whether the user has the content intention to end the continuous input mode based on information such as the interval length or sending frequency when the user sends the content.

[0161] In one embodiment, after the continuous input mode is turned off, a content editing sub-area may be presented, and the user may perform input operations through the content editing sub-area to input content. Figure 8A As shown, after the control for waking up the keyboard in the first display area is operated, the following may be presented: Figure 8A The keyboard shown in the middle right picture is an area where users can input content.

[0162] In one embodiment, after the continuous input mode is turned off, a content generation sub-area and a content editing sub-area may be presented in the first display area. The content generation sub-area is used to present content generated by the content generation model. That is, the first display area corresponding to the updated input function may include both a content editing sub-area and a content generation sub-area. The user may input content through the content editing sub-area, or the user may select content from the content generation sub-area and input it into the content input sub-area to edit the content.

[0163] Taking the current content generation page presenting at least one second content as an example, after closing the continuous input mode, if the user triggers an operation on a third target content in at least one second content, the third target content can be presented in the content input sub-area in response to the trigger operation, and the third target content presented in the content input sub-area can be sent. In response to the sending operation, it is sent to the social object and the third target content is presented in the content presentation sub-area.

[0164] See also Figure 8B As shown, taking the content input sub-area as an input box as an example, when the user clicks the keyboard wake-up control, the keyboard area will be presented on the page corresponding to the input function, and the content generated by the content generation model will be presented above the keyboard. The user can click on any content, and the content will be entered into the input box. The user further clicks the send control to send the content to the chat object and present it in the record presentation area. It should be noted that the chat scenario is used as an example here, but the embodiment of this application does not limit the usage scenario of this solution. For example, in the social platform status content publishing scenario, the send control can be replaced by the publish control, and the record presentation area can be replaced by the status content presentation area, that is, the above solution is still applicable.

[0165] In the embodiment of the present application, the manner of triggering the presentation of the first interface, or triggering the presentation of the first display area including the content generation sub-area in the first interface, may include but is not limited to the following manners:

[0166] (1) When the input function is called, a first display area including a content generation sub-area is presented in the first interface.

[0167] That is, when an operation of calling an input function is detected in the second display area, the input function is called in response to the operation, and the first display area including the content generation sub-area is presented in the first interface. In other words, when the input function is called to input content in the target application, the corresponding triggering operation can be performed, such as clicking a comment or clicking an input box, etc., to pull up the input function, thereby presenting the first display area including the content generation sub-area in the first interface.

[0168] See also Figure 9A As shown in FIG, it is an example of presenting the first display area in the first interface. Figure 9A As shown, before the first display area is presented in the first interface, the first interface only includes the second display area corresponding to the target application. When input is required, the input box can be clicked to pull up the input function, thereby presenting the following: Figure 9A The first interface of the first display area including the input function is shown. By directly presenting the first display area including the content generation sub-area when the input function is pulled up, the speed of generating content for users can be improved, helping users to use the content generated by the large model more quickly. It should be noted that Figure 9A Specifically, the chat scene page is used as an example. When in other types of pages, as long as content input is required, such as clicking on a comment or clicking on editing an article, the input function can be triggered, thereby presenting the first display area containing the content generation sub-area in the first interface.

[0169] (2) When the prompt information prompting the use of the content generation model is operated, a first display area including a content generation sub-area is presented in the first interface.

[0170] When the input function has been called by the target application, the first display area of ​​the input function is presented, but the first display area does not contain the content generation sub-area, that is, the first display area does not display the page corresponding to the content generation model. For example, when the current first display area contains the content editing sub-area, the operation of displaying the page corresponding to the content generation model can be triggered in the current first display area to present the first display area containing the content generation sub-area. Among them, the content editing sub-area is used to edit content, that is, editing operations can be performed under the user's operation to input user content. For example, the content editing sub-area includes a keyboard or a handwriting input board. Taking the keyboard as an example, it can be used to present a keyboard for the input function, and the user can use the keyboard to input the content they need.

[0171] In one possible embodiment, before presenting the content generation sub-area, the first display area may include prompt information, which is used to prompt the user to use the content generation model to generate content. For example, the prompt information may display a VPA icon, prompt content, or prompt instructions. The prompt content may be, for example, content generated by the content generation model, and the prompt instructions may be, for example, "Click me to help you write a hit." Then, the user can perform a trigger operation based on the prompt information to present the first display area containing the content generation sub-area. The trigger operation performed on the prompt information may be, for example, a click operation or a selection operation performed by a button, etc., and there is no specific limitation on this.

[0172] See also Figure 9B As shown, another example of presenting a first display area including a content generation sub-area in a first interface. Figure 9B The first display area presented in the left picture includes a content editing sub-area for the input function, a VPA icon, and prompt instructions. By triggering the VPA icon or prompt instruction, you can enter the page corresponding to the content generation model of the input function, that is, presenting the first display area including the content generation sub-area.

[0173] (3) When using the function of the content generation model, a first display area including a content generation sub-area is presented in the first interface.

[0174] This method means that any step in the process of using the function of the content generation model can trigger the presentation of the first display area including the content generation sub-area in the first interface.

[0175] See also Figure 9CAs shown, it is another example of presenting the first display area in the first interface, wherein: Figure 9C The first display area in the middle left image includes a content generation sub-area and a content editing sub-area. The content generation sub-area is used to present the content generated by the content generation model, and the content editing sub-area is used to edit the content. When the control to collapse the content editing sub-area is operated, the content editing sub-area can be hidden and the content can be presented. Figure 9C The middle right figure shows the first display area that only includes the content generation sub-area; or, when it is detected that the content of the input box has been entered, for example Figure 9C When you click the "OK" control shown, you can present Figure 9C The first interface shown in the middle right figure; or, after the Figure 9C In the first interface shown in the right figure, by operating the content or instructions included therein, the content in the first display area can also be triggered to update, presenting the first interface including new content.

[0176] It should be noted that the presentation of the first interface is not limited to the above-mentioned methods, and this application does not impose any restrictions on this.

[0177] In an embodiment of the present application, in order to remind the user to use the content generation model to generate content, the user may be prompted when the target application calls the input function.

[0178] See also Figure 10 , which is another flow chart of the content input method provided in an embodiment of the present application, the method includes steps 101 to 102.

[0179] Step 101: In response to an input operation triggered on a target application, a first display area is presented, wherein the first display area includes a content editing sub-area for editing content.

[0180] When the user triggers an input operation to input content in the target application, indicating that the input function needs to be called, the first display area corresponding to the input function is presented in response to the input operation triggered in the second display area. The current first display area may include a content editing sub-area, such as a keyboard area.

[0181] Step 102: Prompt information is presented in other areas of the first display area except for the content editing sub-area. The prompt information is used to prompt the use of the content generation model to generate content. The prompt information includes at least one of the fifth content generated by the content generation model and content generation instructions. The prompt information is determined based on the page scene of the second display area and at least one of the content themes associated at the current moment. The content generation instructions are used to indicate the use of the content generation model to generate the required content.

[0182] That is, when the content editing sub-area is presented in the first display area, prompt information may also be presented to prompt the user to generate content using the content generation model.

[0183] The prompt information may include at least one of a VPA icon corresponding to the content generation model, fifth content generated by the content generation model, and a content generation instruction, where the content generation instruction is used to instruct the content generation model to generate the required content.

[0184] In actual scenarios, when the prompt information includes the fifth content or content generation instructions, the prompt information may be determined based on at least one of the page context of the second display area and the currently associated content theme. In other words, when the input function is invoked, the user's intent to input content can be identified based on information such as the page context at the time of invocation and the currently associated content theme, and prompt information related to that intent can be presented. This approach improves the accuracy of the prompt information, thereby more effectively guiding users to utilize the content generation model.

[0185] Specifically, when the target application calls the input function, the target application can send a call request to the input function, and the call request can carry the page scene information where the input function is called. The page scene information includes at least one of the application name of the target application, the application type, the page name of the second display area, the page type, the operation triggered when the call is called, etc., so that the input function can identify the user's intention based on the page scene information, and thus obtain prompt information related thereto. For example, when the intention is to post a comment, chat, or status content, the corresponding internal prompt information can be actively generated according to different intentions, and the content that the user may need can be automatically displayed on the page of the input function (such as the page containing the keyboard), reminding the user that the content generation model can be used to assist in generating content.

[0186] The following illustrates several possible presentations of prompt information in conjunction with specific page scenarios.

[0187] In one possible implementation, when a user invokes an input function in a chat scenario, the input function (or content generation model) recognizes that the current page scenario is a chat scenario and automatically generates content including chat topics. The chat topics may be generated based on real-time hot topics on the Internet, or may be determined based on relevant information of the chat participants, such as historical social content. However, obtaining this information requires obtaining the user's permission or consent, and the collection, use, and processing of the relevant information must comply with the relevant laws, regulations, and standards of the relevant countries and regions.

[0188] See also Figure 11AAs shown in the figure, when the user clicks the input box to open the input function, the content that the user may need can be automatically displayed above the keyboard candidate word of the input function, reminding the user to use the help writing function of the content generation model. Here, the main purpose is to automatically provide the user with chat topics, such as Figure 11A The question “Have you seen the hot search topic “XXXXX” today?” (i.e., the fifth content mentioned above) guides users to use the content generation model.

[0189] It should be noted that when the content generation model is deployed in the server, the input function can send the operations and information received by the front end to the server in real time, so that the server executes the background processing logic, and the input function only needs to present the corresponding page based on the server's processing results.

[0190] When presenting the fifth content in the first display area of ​​the input function (such as the first display area currently only including the keyboard), it can be presented in the form of a content card in the aforementioned content generation sub-area. Alternatively, when presenting the fifth content, a VPA icon corresponding to the content generation model can also be presented, and the fifth content can be presented in the form of a bubble corresponding to the VPA icon, such as Figure 11A As shown, the VPA icon is designed as a cute dog icon, and the fifth content is presented next to the dog icon in the form of bubbles, which looks similar to the bubbles spit out by the dog when talking, and is presented more vividly.

[0191] In one possible implementation, when a user calls an input function in a chat scenario, the input function (or content generation model) recognizes that the current page scenario is a chat scenario, and if there is an associated content topic at the current moment, content related to the content topic can be automatically generated.

[0192] Among them, the content theme associated with the current moment means that the current moment is within a preset time range, and the preset time range is associated with a specific content theme. For example, the content theme can be a festival, anniversary, or a scheduled meeting, etc. The preset time range can be a time corresponding to the adjacent content theme, a time corresponding to the content theme, or a period of time after the time corresponding to the content theme. The specific time range can be set according to actual conditions, and the embodiments of the present application do not limit this.

[0193] See also Figure 11B As shown, if the current scene is identified as a chat scene and a holiday is approaching, holiday greetings can be automatically generated to facilitate users to send holiday greetings. Figure 11B As shown, taking the Mid-Autumn Festival as an example, when it is recognized that the Mid-Autumn Festival is approaching, the greetings related to the Mid-Autumn Festival can be automatically generated and presented in the first display area (such as the first display area including the content editing sub-area), as shown in FIG. Figure 11BIt can be presented in the form of a VPA icon spitting out holiday greetings.

[0194] In one possible implementation, when a user invokes an input function in a social content publishing scenario, the input function (or content generation model) recognizes that the current page scenario is a social content publishing scenario and automatically generates corresponding social content. The social content may be determined, for example, based on the application type or page type of the target application that invoked the input function.

[0195] Specifically, social content publishing scenarios include publishing articles, news, blogs, or personal status content on various social platforms. For example, when a user is identified as publishing on a social platform, the VPA icon can display a prompt "Help you write a hit title" to prompt the user to use the content generation model.

[0196] See also Figure 11C As shown, if the current scene is identified as a text note publishing scene, the publishing-related content can be automatically generated, such as the note title, as shown in Figure 11B The VPA icon shown spits out the content generation instruction "Click me to help you write a hit title", guiding users to use the content generation model.

[0197] Of course, in other scenarios, corresponding content or instructions can also be generated according to the characteristics of each scenario, and the embodiments of the present application do not limit this.

[0198] In this embodiment of the present application, after the prompt information is presented, any of the following processing methods may be performed:

[0199] (1) The user can operate the prompt information, and then in response to the trigger operation, the content generation sub-area can be presented in the first display area. The content generation sub-area is used to present the content generated by the content generation model, that is, after operating any of the VPA icon, prompt content (such as the fifth content mentioned above) or prompt instruction (such as the above-mentioned "Click me to help you write a popular title") presented in the first display area, the user can enter the page of the content generation model, or in other words, the first display area containing the content generation sub-area can be presented on the first interface, that is, the first interface in step S31 or step S51 can be presented. For example, clicking the VPA icon, prompt content or prompt instruction hides the keyboard area and switches to displaying the content generation sub-area.

[0200] In one possible embodiment, when presenting the content generation sub-area, the content presented in the content generation sub-area may be determined based on the prompt information presented before the triggering operation. For example, when the prompt content includes prompt content or prompt instructions, more content related to the prompt content or instruction may be presented in the first display area.

[0201] In another possible implementation, the content presented in the content generation sub-area may be determined based on the object of the user's operation. For example, if the user operates on a VPA icon, the content generation sub-area may not present recommended content, i.e., may present blank content. Alternatively, the content generation sub-area may present pre-configured initial content, such as content that instructs the user on how to use the content generation model. Alternatively, if the user operates on a prompt or instruction, the content generation sub-area may present more content related to the prompt or instruction.

[0202] See also Figure 12 , which is a schematic diagram of a processing method provided in an embodiment of the present application. Figure 12 Specifically, the prompt content is a holiday greeting as an example. After operating on "This night, the Mid-Autumn moon shines brightly on 100,000 homes", the following information may be displayed: Figure 12 The content generation subarea shown in the middle right image. This content presentation page displays multiple pieces of content related to "This night, the Mid-Autumn moon shines brightly, a hundred thousand houses are cleared," also serving as Mid-Autumn Festival greetings. This content can be generated by the content generation model simultaneously with the creation of "This night, the Mid-Autumn moon shines brightly, a hundred thousand houses are cleared," or it can be newly generated by the content generation model using "This night, the Mid-Autumn moon shines brightly, a hundred thousand houses are cleared" as a prompt.

[0203] (2) When the prompt information includes the fifth content (i.e., the prompt content) and the user performs a trigger operation on the fifth content, the fifth content can be directly sent to the social object in response to the trigger operation, and the fifth content can be presented in the content presentation sub-area included in the second display area, and the content presentation sub-area is used to present the social content sent to the social object. This method is similar to the continuous input mode, that is, the fifth content can be directly sent to the upper screen for display.

[0204] (3) When the prompt information includes the fifth content (i.e., the prompt content) and the user performs a trigger operation on the fifth content, the fifth content can be presented in the content input sub-area included in the second display area in response to the trigger operation. The content input sub-area is used to present the content currently edited by the input function, such as the input box area. The user can further optimize and edit the content in the input box and send it to the social object. At the same time, the area originally presenting the fifth content can be updated to present other content, such as other content with similar intent to the fifth content.

[0205] (4) When the prompt information includes a content generation instruction, if the user performs a trigger operation in response to the content generation instruction, a first display area including a content input sub-area may be presented in response to the trigger operation, and the content presented in the first display area is generated by the content generation model according to the content generation instruction, and the content generation instruction may be any of the instructions mentioned above. In this manner, the first interface in step S31 or step S51 may also be presented, that is, at least one first content may be presented in the first display area.

[0206] See also Figure 13 As shown, it is an example of combining a large model with an input function in the related art. Figure 13 The input function of the example still uses the big model dialogue method. An input box for big model communication is embedded in the input function area. The input box displays "Tell me about your communication problems" to remind users to use this function. After users enter questions in the input box, the big model provides content by answering. Figure 13 As can be seen from the chat scene interface shown, in addition to the chat input box on the interface, this method adds a process of dialogue with the large model on the basis of the original scene interaction. A new input box for communicating with the large model appears outside the user's scene. The main path for user input content will be changed, which may easily cause users to have questions about the two input boxes, easily cause confusion, and increase user communication and input costs.

[0207] Based on this, in the embodiments of this application, see Figure 12 As shown, the input box in the second display area serves as a communication path with the content generation model. That is, the user can use this input box to input prompt content for the content generation model. As the user enters content, the content generated by the content generation model based on the content will appear in the first display area simultaneously. This method can avoid user confusion about the input path and avoid increasing the user's learning cost.

[0208] In the embodiment of the present application, after the first interface is presented, if it is determined that the user is not currently in the continuous input mode or that the conditions for opening the continuous input mode are not met, the user can proceed according to the execution logic of the normal input mode. In addition, the content generation capability of the content generation model is interspersed during the input process in the embodiment of the present application, helping the user to more efficiently input content that more accurately expresses the user's intention and provides higher quality content during the uninterrupted input process. Figure 14 As shown, it is another flow chart of the content input method provided by the embodiment of the present application, and the flow includes the following steps: Figure 14 Steps S141 to S144 are shown.

[0209] S141: Presenting a first interface, the first interface including a first display area corresponding to an input function and a second display area corresponding to a target application, the first display area presenting at least one first content generated by a content generation model.

[0210] The process of step S141 is similar to that of step S31 or S51, so please refer to the introduction in the aforementioned part and will not be repeated here.

[0211] S142: In response to a triggering operation on a fourth target content in the at least one first content, present the fourth target content in the content input sub-area.

[0212] The second display area in the first interface may also include a content input sub-area, see Figure 4A The area where the input box is located is the content input sub-area, which is used to present the content currently being edited.

[0213] Specifically, if it is determined that the user is not currently in continuous input mode or that the conditions for enabling continuous input mode are not met, the execution logic of normal input mode may be used. That is, when the user triggers the fourth target content, the fourth target content will be entered into the input box, and the user will be allowed to edit it or click "Send" to send it to the corresponding social network partner. The fourth target content will then be presented in the content presentation sub-area. The content presentation sub-area is an area for presenting social content sent to a social network partner, such as a record presentation area, a status content presentation sub-area, or a comment presentation area.

[0214] See also Figure 15A As shown, when the user clicks on the content card corresponding to the fourth target content, namely "When we look into each other's eyes, will your heart also say welcome to me?", the content will enter the input box, so that the user can further optimize the content in the input box to obtain the content that the user needs more. After the user clicks on the content, the area in the first display area that displays the content generated by the content generation model can remain the same as the original interface, that is, the content is not updated; or the content in the area can also be updated accordingly, see Figure 15A The specific illustration is based on this example.

[0215] S143: Present at least one candidate adjustment instruction in the first display area, wherein each candidate adjustment instruction is used to represent an adjustment requirement for the content output by the content generation model, and at least one candidate adjustment instruction is determined according to the fourth target content.

[0216] In an embodiment of the present application, in order to more intelligently assist the user in generating more accurate content, after the user triggers an operation on the fourth target content, at least one candidate adjustment instruction may be determined based on the fourth target content selected by the user, and the at least one candidate adjustment instruction may be presented in the first display area. The adjustment instruction may represent the user's adjustment requirements for the content output by the content generation model, that is, how further adjustments need to be made to the currently selected fourth target content or the content generated by the content generation model.

[0217] See also Figure 15A As shown, after the user clicks on the fourth target content, multiple candidate adjustment instructions will be presented in the first display area, namely "be humorous", "express in classical Chinese", "add emoji", etc. For example, "be humorous" can be an instruction to express the fourth target content selected by the user in a more humorous tone. Among them, multiple candidate adjustment instructions can be displayed in the instruction row sub-area, and the instruction row sub-area is the area where the instructions are presented. The embodiment of the present application does not impose specific restrictions on the number and position of the instruction row sub-area in the first display area. It can be configured according to the situation during specific implementation, or the user can configure its number and position by himself.

[0218] In one possible implementation, before presenting at least one candidate adjustment instruction, at least one original instruction is presented in the first display area. The original instruction includes at least one of a type instruction and an adjustment instruction, and the type instruction is used to indicate the type of content generated by the content generation model, such as generated text, image, or translated content.

[0219] Then, see Figure 15B As shown, the first display area originally presents multiple types of instructions. After the first target content is triggered, the multiple types of instructions can be updated to at least one candidate adjustment instruction. Alternatively, see Figure 15C As shown, multiple types of instructions and at least one candidate adjustment instruction may also be presented in the first display area, that is, these two instructions may be presented at the same time. These two instructions may be presented in different instruction row sub-areas or in the same instruction row sub-area, and there is no specific limitation on this. Figure 15C The example above uses the candidate adjustment instructions being presented below the type instruction, but in practice, there are no restrictions on their placement. Alternatively, when multiple candidate adjustment instructions are initially presented in the first display area, after a triggering operation is performed on the first target content, the multiple candidate adjustment instructions can be updated to at least one candidate adjustment instruction.

[0220] S144: In response to a triggering operation on a target adjustment instruction in at least one candidate adjustment instruction, updating and presenting at least one fourth content in the first display area, where the at least one fourth content is generated by the content generation model according to the adjustment requirement corresponding to the target adjustment instruction.

[0221] That is, when the user selects a target adjustment instruction to perform an operation, the content generation model is triggered to generate new content based on the adjustment requirements corresponding to the target adjustment instruction and present it in the first display area. In one embodiment, the target adjustment instruction can also be presented in the content generation box (the area for presenting generated content) in the first display area, and the content generated based on the target adjustment instruction is presented immediately thereafter.

[0222] See also Figure 15D As shown, when the user clicks the "Add emoji" command, the "Add emoji" command will be reflected in the content generation frame area and the content card generated according to the "Add emoji" command will be presented immediately afterwards. In addition, see Figure 15D As shown, the command line will rematch and present the optimization commands that can be used for the newly generated content.

[0223] Considering that instruction matching is an important link in connecting content production and user input experience of the input function to realize the content generation model, if the instruction matching is inaccurate or impractical, the user will not be able to obtain the desired content, thereby affecting the user's experience of intelligent input. Based on this, a variety of instruction matching methods are provided in the embodiments of the present application. When determining at least one candidate adjustment instruction, it can be determined according to one or more combinations of the following methods, which are introduced below.

[0224] (1) Determine at least one candidate adjustment instruction according to a page scenario when the target application calls an input function.

[0225] In other words, different page scenarios can correspond to different adjustment instructions, thereby assisting the content generation model to accurately generate content.

[0226] In practical applications, an exhaustive transformation approach can be used to first list commonly used input scenarios, such as chat, comment, forum, social sharing, status sharing, and movie review scenarios. Based on these scenarios, the user's intent can be determined and associated with specific commands. This means that based on commonly used input scenarios, a command library corresponding to user intent is created, and the overall input content generation model is used to generate content. Subsequently, a large model can be trained to expand user scenarios, matching the command library to user intent, enabling the input method to more accurately cover user intelligent input.

[0227] For example, in a movie review scenario, when a user wants to post a movie review, the matching instructions may be instructions related to movie reviews. Or, in a chat scenario, when a user wants to send a leave request message, the matching instructions may be instructions related to leave requests.

[0228] In actual scenarios, the page scenario can also combine the content input by the user to determine the user's intention. For example, if the user enters "I like you" in a chat scenario, it indicates that the user may need to confess to the social object, and the instructions related to the intention can be matched, such as generating content using different forms of expression.

[0229] (2) Determine at least one candidate adjustment instruction based on the content intention corresponding to the fourth target content.

[0230] That is, content recognition may be performed on the fourth target content selected by the user to obtain the corresponding content intent, thereby determining at least one candidate adjustment instruction according to the content intent.

[0231] For example, if the content selected by the user indicates that the user wants to confess his / her love, the matched instruction may be a content optimization instruction related to confession.

[0232] In actual scenarios, the fourth target content can also be combined with the page context to determine the content intent. For example, if a user selects content expressing holiday greetings in a chat scenario, this indicates that the user may want to express holiday greetings to their social partner. Therefore, instructions related to this intent can be matched, such as generating content with a different holiday greeting tone.

[0233] (3) Determine at least one candidate adjustment instruction based on the changes in the content input sub-area.

[0234] That is, based on the real-time status of the input box, every time the content of the input box changes, such as selecting content, adding or removing words, etc., the user's intention can be re-identified and the instructions displayed in the instruction area can be matched and refreshed.

[0235] For example, after inputting the fourth target content into the content input sub-area, the user can edit the fourth target content, and at least one candidate adjustment instruction can be determined based on the change of the fourth target content in the content input sub-area. That is to say, each time the fourth target content is triggered to edit, the adjustment instruction can be re-matched according to the modified fourth target content and the instruction displayed in the first display area can be updated.

[0236] (4) Determine at least one candidate adjustment instruction based on the content status in the content input sub-area.

[0237] The content status can be used to indicate the input stage corresponding to the content in the content input sub-area. For example, the input stage can be represented by the content length, that is, the content status can include the content length. Therefore, a target interval range for the content length in the content input sub-area can be determined from at least one preset interval range, and then at least one candidate adjustment instruction can be determined based on the candidate adjustment instruction set corresponding to the target interval range.

[0238] In this way, the embodiment of the present application can match corresponding differentiated instructions for users to choose at each input stage of the user, such as when the input box is empty, a word phrase, a sentence paragraph or a note article, to help users understand how to communicate with the big model to obtain content production. Among them, the empty input box represents that nothing has been entered in the current input box, the word phrase represents that the length of the content in the input box is in the first interval range, such as the first interval range is 0 to 15 characters or 2 to 15 characters, the sentence paragraph represents that the length of the content in the input box is in the second interval range, such as the second interval range is 15 to 50 characters or 16 to 99 characters, and the note article represents that the length of the content in the input box is in the third interval range, such as the third interval range is 100 to 500 characters. The specific interval range can be configured according to the actual scenario, and the embodiment of the present application does not limit this.

[0239] For example, see Figure 16 The figure shows a flow chart of intelligent input assistance for content generation model provided in an embodiment of the present application.

[0240] Among them, Figure 16 In the process, the content generation model can automatically identify the page scene the user is currently on and add intelligent assistance steps to the user's input progress, helping users to produce, polish, expand, and personalize the input content. Among them, the intelligent assistance function can cover various scenarios such as social networking, commenting, publishing, and documents. It recognizes the user's intention through the scene and can associate it with specific instructions. Based on the real-time status of the user's input box, every time the content of the input box changes, the user's intention is re-identified and the instructions in the instruction area are matched and refreshed.

[0241] Under these scenarios, differentiated instruction options can be provided to users at different stages of user input to help them understand how to communicate with the big model to produce content. The following examples illustrate the content or instructions that can be provided at different stages of input:

[0242] (1) When the input box is empty, two types of content can be provided. The first type is to provide chat topics, chat greetings, comments or hot releases based on scene recognition, such as holiday greetings, weather changes or theme selection. The second type is to provide prompt instructions based on fuzzy matching scenes, such as help you write or you can try asking me, etc., to guide users to use content generation models, or provide a few days, weather or hot topics based on dimensions, etc., to help users quickly input content.

[0243] (2) When the user inputs a phrase (i.e., the content length is 2 to 15 characters), content generation instructions can be provided, such as helping to write comments, leave notes, or blessings, or using high emotional intelligence expressions, euphemistic expressions, or affectionate expressions, to improve the user's content input efficiency.

[0244] (3) When the user enters a sentence paragraph (i.e., the content length is 15 to 50 words), polishing instructions can be provided, such as making it more humorous, expressing it in English, expressing it in classical Chinese, adding emojis, or lengthening and expanding it, to assist the user in optimizing the content and improving the richness of the content.

[0245] (4) When the user enters a note article (i.e., the content length is 100 to 500 words), professional instructions can be provided, such as organization, personalization, summary generation, or paragraph formatting, to assist the user in improving the content and enhancing its accuracy.

[0246] The above solution is introduced below with an example, and the following example is illustrated by taking holiday greetings in a chat scenario as an example.

[0247] See also Figure 17 The figure shows an example diagram of intelligent input assistance provided by an embodiment of the present application.

[0248] S171: When the user clicks the input box to call the input function, the content generation model recognizes that the user is in a chat scene and the Mid-Autumn Festival is approaching, and generates text content of Mid-Autumn Festival-related greetings and blessings, which is displayed at the corresponding position of the VPA icon.

[0249] S172: The user clicks on the text content generated in step S171, the keyboard disappears, and three related content cards are displayed on the page. Also, a line below the input box displays instructions for the types of content that can be generated, such as AI writing assistance, AI illustration, or translation.

[0250] S173: The user clicks on a piece of content, the content enters the input box, and the optimization instructions that can be used for the content are displayed in the instruction line, for example Figure 17 Use more humor, classical Chinese expressions, or add emojis, etc.

[0251] S174: The user clicks the add emoji command, and the command is displayed on the page, and the content card generated according to the command is displayed immediately afterwards. At the same time, the command line will match and display the optimization commands that can be used for the content, such as Figure 17 Add poems, English expressions, or more emojis as shown.

[0252] S175: The user clicks on the English expression, and the instruction will be displayed on the page, and the content card generated according to the instruction will be displayed immediately afterwards. At the same time, the instruction line will match and display the optimization instructions that can be used for the content, such as Figure 17 Add poetry, some humor, or emojis as shown in the example.

[0253] S176: The user can slide up and down in the content generation sub-area of ​​the first display area to review the previously generated content, such as Figure 17 As shown in , when the user slides down, the content generated by the last emoji command will be presented on the page, thereby helping the user make decisions and select the content they need.

[0254] When the user needs to send a piece of content, he can click on the result content card, the content card will be selected, the content will enter the input box, click send to complete the screen, or the user can drag to enter the continuous input mode to directly put the content on the screen.

[0255] See also Figure 18 As shown, it is another example of combining a large model with an input function in the related art. Figure 18 In the example, the user inputs "Have the years stolen your youth recently?", and the content generated by the model "You look brand new recently, more mature and charming!" will be presented in the content generation box area. In addition, a "change sentence" icon is provided in the content generation box area, and the user can click the icon to update the content in the content generation box area. If the user needs to change a type of content, the keyboard editing can be called up in the original scene input box. However, after the update, the content in the content generation box area will be refreshed, that is, replaced by the newly produced content. The user cannot flexibly compare the content produced by the large model, and the decision-making efficiency is low. For example, it may be necessary to update several times in succession to determine the content you need, and the efficiency of content input is not high.

[0256] Compared with the methods in related arts, in the embodiments of the present application, previous content can be retraced by sliding up and down, avoiding the complete refreshing of previous content. When the user needs previously generated content, he may need to refresh and obtain it repeatedly, reducing the complexity of the user's operation and thus improving the efficiency of content input.

[0257] In the embodiment of the present application, if the instructions presented in the instruction line do not match the user's request, the user can also customize the instructions. Figure 19 As shown, when the user clicks on a custom command, the custom box for the command can be expanded and the keyboard can be called up. The user can enter the command he needs in the custom box, for example Figure 19 Expand a few lines of poetry as shown, and click OK in the lower right corner of the keyboard to generate a custom command.

[0258] The custom box can be an input box, which can reduce the number of input boxes on the page and avoid user confusion. Alternatively, after clicking the custom command, you can jump to a new page, that is, the custom page, to customize the command.

[0259] like Figure 19 As shown, after clicking confirm, the instruction can be sent to the content generation model to generate new content. Alternatively, the instruction can be presented on the page after clicking confirm, and when the user clicks the instruction, the content generation model is triggered to generate new content.

[0260] In one possible implementation, the custom command input box can also serve as a means of communicating with the content generation model, and corresponding content can be generated based on the instructions entered into the custom command input box. For example, if a user enters a topic for today's post in the custom command input box, multimodal results such as text, images, or videos suitable for the post topic of the social platform currently used by the user can be generated for the user to select.

[0261] In one possible implementation, the system can intelligently identify the scenario and present corresponding prompts in the custom command input box to assist users in using the content generation model. For example, if it recognizes that a user is posting a movie review, the system can directly expand the custom command input box and prompt the user to enter the movie title. After the user enters the movie title and clicks "Confirm", a detailed movie review will be generated for the user, with support for optimized opinions.

[0262] See also Figure 20 As shown, it is a schematic diagram of the interaction logic between the user, input function and content generation model provided in an embodiment of the present application.

[0263] When a user enters a page scene and calls the input function, the input function can obtain the scene information returned by the target application, and can judge the user's intention based on the scene information, or, based on the words and phrases input by the user, it can combine the scene information and the words and phrases to judge the user's intention, and then match the instructions related to the user's intention and display the matching instructions. Here we mainly take the adjustment instruction as an example.

[0264] If the user clicks on one of the adjustment instructions, it will trigger the adjustment instruction to be sent to the content generation model. After the content generation model generates content according to the adjustment instruction, it will display the generated content on the page of the input function.

[0265] If the user clicks on one of the content, it will be entered into the input box of the target application's current page. If the user modifies the content of the input box, the input keyboard will be called for the user to enter the content. After the modification, the user can click the send control to send the content to the screen. At the same time, based on the content status of the input box, the new instructions will be matched in real time and displayed on the page.

[0266] After the content is displayed, if it is detected that the user long-presses and drags the content upward, it will trigger the entry into continuous input mode, and the content dragged by the user will be directly sent to the screen for display. After the content is displayed on the screen, it will trigger the sending of instructions to the content update model to generate new content. The content generation model will continuously generate new content based on the current intention and display the newly generated content. The user can also drag and swipe these contents upward to quickly send the content to the screen. In addition, the adjustment instructions displayed in the instruction row sub-area will switch to style instructions in the continuous input mode. After the user clicks to select a style instruction, the style instruction will be sent to the content generation model. The content generation model will continuously generate new content according to the corresponding content style and display the newly generated content. The user can also drag and swipe these contents upward to quickly send the content to the screen.

[0267] In an embodiment of the present application, it is also possible to collect user usage data on custom instructions during the process of users using the content generation model to generate content, and then use these usage data as training data to perform small model training on the content generation model in the vertical field of content generation, and fine-tune the content generation model to improve the accuracy of instruction matching and content generation.

[0268] In summary, in the embodiment of the present application, the user's intention to input content can be identified based on the scene information returned when the input function is called up, and content can be actively generated and displayed to the user. After the user selects content and inputs it into the input box, appropriate instructions that match the content of the user's input box are displayed to help the user optimize the existing content of the input box so that the content is more in line with the user's input requirements. In addition, the embodiment of the present application also provides a special input mode. When it is determined that the user is in a special input scenario based on the user's specific behavior, instructions in a special input mode are provided. When the user clicks on the instruction or the user uses a gesture to press and hold to drag a generated content to send it to the screen, an instruction to automatically generate new content is triggered immediately, thereby supporting users to achieve a quick burst mode in which they can continuously obtain large-scale model generated content without having to issue instructions each time, and send content to the screen through simple gestures.

[0269] In the solution of the embodiment of the present application, the content generation model and the input function form are more naturally and closely integrated. The user does not need to change the input scene to conduct the content generation model dialogue, which saves the user's time to think and ask questions, lowers the usage threshold, realizes intelligent assisted input, and makes input more convenient and efficient. The quick burst function has an insight into and splits the specific demands of the user for input content, and innovatively extends the product functions of the large model embedded input method. It deeply explores the differences in users' demands for screen efficiency before content accuracy in specific scenarios. Through the innovative burst mode, it supports users to input continuously and quickly on the screen, improves scene coverage, and better solves the problem that the user's input efficiency cannot keep up with the speed of thinking. It gives full play to the advantages of the large model and brings users a user experience of free speech and flowing ideas.

[0270] It should be noted that the content input sub-area (such as the input box) in the embodiment of the present application may belong to the target application or to the input function, and there is no limitation on this.

[0271] See Figure 21 Based on the same inventive concept, the embodiment of the present application further provides a content input device 210, which includes:

[0272] The interface presenting unit 2101 is configured to present a first interface, the first interface including a first display area corresponding to an input function and a second display area corresponding to a target application, the first display area presenting at least one first content generated by a content generation model;

[0273] a content updating unit 2102 configured to, in response to a triggering operation on a first target content in the at least one first content, present the first target content in the second display area, and update and present at least one second content in the first display area, where the at least one second content is generated by the content generation model according to the content intent corresponding to the first target content;

[0274] The content updating unit 2102 is further configured to present the second target content in the second display area in response to a triggering operation on the second target content in the at least one second content.

[0275] In a possible implementation, the second display area includes a content presentation sub-area, and the content presentation sub-area is used to present social content sent to the social object;

[0276] The content updating unit 2102 is specifically configured to present the first target content or the second target content in the content presentation sub-area.

[0277] In a possible implementation, the apparatus further includes an instruction presenting unit 2103, configured to:

[0278] presenting at least one candidate style instruction associated with the first target content in the first display area, wherein each candidate style instruction is used to indicate a content style of the content generated by the content generation model;

[0279] In response to a triggering operation on a target style instruction in at least one candidate style instruction, at least one third content is updated and presented in the first display area. The at least one third content is generated by the content generation model according to the content style corresponding to the target style instruction.

[0280] In a possible implementation, the apparatus further includes an instruction matching unit 2104, configured to:

[0281] Performing intent recognition on the first target content to obtain a first intent recognition result;

[0282] Performing intent recognition on at least one input content to obtain a second intent recognition result; wherein each input content is content obtained based on a received input operation, and the at least one first content is generated by a content generation model based on the at least one input content;

[0283] determining, based on at least one of the first intent recognition result and the second intent recognition result, a content intent corresponding to the first target content;

[0284] At least one candidate style instruction matching the content intent corresponding to the first target content is obtained.

[0285] In a possible implementation, the interface presenting unit 2101 is specifically configured to:

[0286] Presenting at least one candidate adjustment instruction in the first display area, wherein each candidate adjustment instruction is used to represent an adjustment requirement for content output by the content generation model;

[0287] The instruction presenting unit 2103 is specifically configured to update and present at least one candidate adjustment instruction as at least one candidate style instruction; or present at least one candidate adjustment instruction and at least one candidate style instruction in the first display area.

[0288] In a possible implementation, the apparatus further includes a determining unit 2105, configured to:

[0289] Determining that a condition for enabling a continuous input mode is met; wherein, in the continuous input mode, each time content is output to a target application, an update of content presented in the first display area is triggered;

[0290] The instruction presenting unit 2103 is specifically configured to present at least one candidate style instruction in the first display area when it is determined that the continuous input mode is in progress.

[0291] In one possible implementation, the activation condition includes at least one of the following conditions:

[0292] Receiving a preset operation triggered by content presented in the first display area;

[0293] Receive the start command corresponding to the continuous input mode;

[0294] Determining, based on at least one input content, a content intention to enable a continuous input mode;

[0295] Determine, based on at least one social content sent historically, whether the content has the intention to enable the continuous input mode.

[0296] In a possible implementation, the second display area further includes a content input sub-area, and the content input sub-area is used to present the content currently being edited by the input function. Then, the determining unit 2105 is further used to:

[0297] When it is determined that the end condition of the continuous input mode is met, the continuous input mode is closed;

[0298] In response to a triggering operation on a third target content in the at least one second content, presenting the third target content in the content input sub-area;

[0299] In response to a sending operation performed on the third target content presented in the content input sub-region, the third target content is presented in the content presentation sub-region.

[0300] In one possible implementation, the termination condition includes at least one of the following conditions:

[0301] receiving a triggering operation of presenting a content editing sub-area with an input function in the first display area, the content editing sub-area being used for editing content;

[0302] Receive a close command corresponding to the continuous input mode;

[0303] Determining, based on at least one input content, a content intention to close the continuous input mode;

[0304] Determine, based on at least one social content sent historically, whether the content has a content intention to close the continuous input mode.

[0305] In a possible implementation, the second display area includes a content input sub-area, and the content input sub-area is used to present the content currently being edited by the input function. Then, the instruction presenting unit 2103 is further used to:

[0306] In response to a triggering operation on a fourth target content in the at least one first content, presenting the fourth target content in the content input sub-area; and presenting at least one candidate adjustment instruction in the first display area, wherein each candidate adjustment instruction is used to represent an adjustment requirement for content output by the content generation model, and the at least one candidate adjustment instruction is determined based on the fourth target content;

[0307] In response to a triggering operation on a target adjustment instruction in at least one candidate adjustment instruction, at least one fourth content is updated and presented in the first display area, where the at least one fourth content is generated by the content generation model according to the adjustment requirement corresponding to the target adjustment instruction.

[0308] In a possible implementation, the instruction presenting unit 2103 is specifically configured to:

[0309] presenting at least one original instruction in the first display area, the original instruction including at least one of a type instruction and an adjustment instruction, the type instruction being used to indicate a content type of content generated by the content generation model;

[0310] presenting at least one original instruction update as at least one candidate adjustment instruction; or,

[0311] At least one original instruction and at least one candidate adjustment instruction are presented in the first display area.

[0312] In a possible implementation, the instruction matching unit 2104 is specifically configured to:

[0313] At least one candidate adjustment instruction is determined by at least one of the following methods:

[0314] determining at least one candidate adjustment instruction according to the content intent corresponding to the fourth target content;

[0315] Determining at least one candidate adjustment instruction according to the page scene corresponding to the second display area;

[0316] determining at least one candidate adjustment instruction based on the change in the content input sub-area;

[0317] At least one candidate adjustment instruction is determined according to the content state in the content input sub-area.

[0318] In a possible implementation, the content status includes the content length; the instruction matching unit 2104 is specifically configured to:

[0319] Determine a target interval range for the content length in the content input sub-area from at least one preset interval range;

[0320] At least one candidate adjustment instruction is determined according to the candidate adjustment instruction set corresponding to the target interval range.

[0321] In a possible implementation, the apparatus further includes a scene recognition unit 2106 configured to:

[0322] In response to an input operation triggered on a target application, presenting a first display area, wherein the first display area includes a content editing sub-area, and the content editing sub-area is used to edit content; and

[0323] Prompt information is presented in other areas of the first display area except for the content editing sub-area. The prompt information is used to prompt the use of the content generation model to generate content. The prompt information includes at least one of the fifth content generated by the content generation model and the content generation instruction. The prompt information is determined based on the page scene of the second display area and at least one of the content themes associated at the current moment. The content generation instruction is used to indicate the use of the content generation model to generate the required content.

[0324] In a possible implementation, the scene recognition unit 2106 is further configured to:

[0325] In response to a triggering operation on the prompt information, presenting a content generation sub-area in the first display area, where the content generation sub-area is used to present content generated by the content generation model;

[0326] When the prompt information includes the fifth content, in response to a triggering operation on the fifth content, presenting the fifth content in a content presentation sub-area included in the second display area, the content presentation sub-area being used to present social content sent to the social object;

[0327] When the prompt information includes the fifth content, in response to a triggering operation on the fifth content, the fifth content is presented in a content input sub-area included in the second display area, where the content input sub-area is used to present the content currently being edited by the input function;

[0328] When the prompt information includes a content generation instruction, in response to a triggering operation on the content generation instruction, a content generation sub-area is presented in the first display area, and the content generation sub-area includes content generated by the content generation model according to the content generation instruction.

[0329] Through the above device, the content displayed in the first display area can be automatically updated after one content is output, so that the required content can be continuously selected in the first display area. Without the need for other input operations or update operations, only a simple trigger operation is required to achieve the effect of continuously outputting content to the target application, greatly improving the efficiency of content input.

[0330] The device can be used to execute the methods shown in the various embodiments of the present application. Therefore, for the functions that can be implemented by the various functional modules of the device, please refer to the description of the aforementioned embodiments and no further details will be given.

[0331] See Figure 22 Based on the same technical concept, the present application also provides a computer device. In one embodiment, the computer device can be Figure 1 The server shown or Figure 2 The device corresponding to the cloud shown in FIG. Figure 22 As shown, it includes a memory 2201, a communication module 2203 and one or more processors 2202.

[0332] Memory 2201 is used to store computer programs executed by processor 2202. Memory 2201 may primarily include a program storage area and a data storage area. The program storage area may store an operating system and programs required for running instant messaging functions, while the data storage area may store various instant messaging messages and operating instruction sets.

[0333] Memory 2201 may be a volatile memory, such as random-access memory (RAM); a non-volatile memory, such as read-only memory, flash memory, a hard disk drive (HDD), or a solid-state drive (SSD); or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto. Memory 2201 may be a combination of the aforementioned memories.

[0334] The processor 2202 may include one or more central processing units (CPUs) or digital processing units, etc. The processor 2202 is configured to implement the above-mentioned content input method when calling the computer program stored in the memory 2201 .

[0335] The communication module 2203 is used to communicate with terminal devices and other servers.

[0336] The specific connection medium between the memory 2201, the communication module 2203 and the processor 2202 is not limited in the embodiment of the present application. Figure 22 The memory 2201 and the processor 2202 are connected via a bus 2204. Figure 22 The connections between the other components are shown in bold lines, which are only for illustration and are not intended to be limiting. The bus 2204 can be divided into an address bus, a data bus, a control bus, etc. For ease of description, Figure 22 The diagram shows a single thick line, but this does not indicate that there is only one bus or one type of bus.

[0337] The memory 2201 stores a computer storage medium, which stores computer executable instructions. The computer executable instructions are used to implement the content input method of the embodiments of the present application. The processor 2202 is used to execute the content input method of the above embodiments.

[0338] In another embodiment, the computer device may also be a terminal device, such as Figure 1 In this embodiment, the structure of the computer device can be as follows: Figure 23 As shown, it includes: a communication component 2310, a memory 2320, a display unit 2330, a camera 2340, a sensor 2350, an audio circuit 2360, a Bluetooth module 2370, a processor 2380 and other components.

[0339] The communication component 2310 is used to communicate with the server. In some embodiments, it may include a wireless fidelity (WiFi) module. The WiFi module is a short-range wireless transmission technology. The computer device can help users send and receive information through the WiFi module.

[0340] The memory 2320 can be used to store software programs and data. The processor 2380 executes various functions and data processing of the terminal device by running the software programs or data stored in the memory 2320. The memory 2320 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. The memory 2320 stores an operating system that enables the terminal device to run. In the present application, the memory 2320 can store the operating system and various application programs, and may also store code for executing the content input method of the embodiment of the present application.

[0341] The display unit 2330 can also be used to display information input by the user or information provided to the user, as well as a graphical user interface (GUI) of various menus of the terminal device. Specifically, the display unit 2330 may include a display screen 2332 disposed on the front of the terminal device. The display screen 2332 may be configured in the form of a liquid crystal display, a light-emitting diode, etc. The display unit 2330 can be used to display various interfaces or pages in the embodiments of the present application.

[0342] The display unit 2330 can also be used to receive input digital or character information and generate signal input related to the user settings and function control of the terminal device. Specifically, the display unit 2330 may include a touch screen 2331 arranged on the front of the terminal device, which can collect user touch operations on or near it, such as clicking a button, dragging a scroll box, etc.

[0343] The touch screen 2331 can be covered on the display screen 2332, or the touch screen 2331 and the display screen 2332 can be integrated to realize the input and output functions of the terminal device. The integrated display screen can be simply called a touch display screen. The display unit 2330 in this application can display the various interfaces or pages involved above.

[0344] The camera 2340 can be used to capture still images, and users can post comments on the images captured by the camera 2340 through the application. The camera 2340 can be one or more. The object generates an optical image through the lens and projects it onto the photosensitive element. The photosensitive element can be a charge coupled device (CCD) or a complementary metal-oxide-semiconductor (CMOS) phototransistor. The photosensitive element converts the light signal into an electrical signal, and then transmits the electrical signal to the processor 2380 to convert it into a digital image signal.

[0345] The terminal device may further include at least one sensor 2350, such as an acceleration sensor 2351, a distance sensor 2352, a fingerprint sensor 2353, and a temperature sensor 2354. The terminal device may also be configured with other sensors such as a gyroscope, a barometer, a hygrometer, a thermometer, an infrared sensor, a light sensor, and a motion sensor.

[0346] The audio circuit 2360, speaker 2361, and microphone 2362 provide an audio interface between the user and the terminal device. The audio circuit 2360 can convert the received audio data into an electrical signal and transmit it to the speaker 2361, which converts it into a sound signal for output. The terminal device may also be equipped with a volume button for adjusting the volume of the sound signal. Meanwhile, the microphone 2362 converts the collected sound signal into an electrical signal, which is then received by the audio circuit 2360 and converted into audio data. The audio data is then output to the communication component 2310 for transmission to, for example, another terminal device, or the audio data is output to the memory 2320 for further processing.

[0347] The Bluetooth module 2370 is used to exchange information with other Bluetooth devices having a Bluetooth module through the Bluetooth protocol. For example, the terminal device can establish a Bluetooth connection with a wearable computer device (such as a smart watch) that also has a Bluetooth module through the Bluetooth module 2370 to exchange data.

[0348] The processor 2380 is the control center of the terminal device. It uses various interfaces and lines to connect various parts of the entire terminal. By running or executing software programs stored in the memory 2320 and calling data stored in the memory 2320, it performs various functions of the terminal device and processes data. In some embodiments, the processor 2380 may include one or more processing units; the processor 2380 may also integrate an application processor and a baseband processor, wherein the application processor mainly processes the operating system, user interface, and application programs, and the baseband processor mainly processes wireless communications. It is understandable that the above-mentioned baseband processor may not be integrated into the processor 2380. In this application, the processor 2380 can run the operating system, application programs, user interface display and touch response, as well as the content input method of the embodiment of the present application. In addition, the processor 2380 is coupled to the display unit 2330.

[0349] Based on the same inventive concept, an embodiment of the present application further provides a storage medium storing a computer program. When the computer program runs on a computer, the computer executes the steps of the content input method according to various exemplary embodiments of the present application described above in this specification.

[0350] In some possible implementations, various aspects of the content input method provided by the present application may also be implemented in the form of a computer program product, which includes a computer program. When the program product is run on a computer device, the computer program is used to enable the computer device to execute the steps of the content input method according to various exemplary embodiments of the present application described above in this specification. For example, the computer device can execute the steps of each embodiment.

[0351] The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or component, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0352] The program product of the embodiment of the present application may be a portable compact disc read-only memory (CD-ROM) and include a computer program, and can be run on a computer device. However, the program product of the present application is not limited thereto. In the present application, a readable storage medium may be any tangible medium containing or storing a program, and the computer program included therein may be used by or in conjunction with a command execution system, apparatus, or device.

[0353] A readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries a readable computer program. Such a propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium that can transmit, propagate, or transfer a program for use by or in conjunction with a command execution system, apparatus, or device.

[0354] The computer program embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0355] Computer programs for performing the operations of the present application may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages.

[0356] It should be noted that although several units or subunits of the device are mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, depending on the embodiment of the application, the features and functions of two or more units described above can be embodied in a single unit. Conversely, the features and functions of a single unit described above can be further divided and embodied by multiple units.

[0357] Furthermore, although the operations of the method of the present application are described in a particular order in the accompanying drawings, this does not require or imply that the operations must be performed in this particular order, or that all illustrated operations must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps.

[0358] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0359] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0360] Obviously, those skilled in the art may make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application is intended to include these modifications and variations.

Claims

1. A content input method, characterized in that: The method comprises: Presenting a first interface, the first interface including a first display area corresponding to an input function and a second display area corresponding to a target application, the first display area presenting at least one first content generated by a content generation model; In response to a triggering operation on a first target content among the at least one first content, presenting the first target content in the second display area, and updating and presenting at least one second content in the first display area, where the at least one second content is generated by the content generation model according to the content intent corresponding to the first target content; In response to a triggering operation on a second target content among the at least one second content, the second target content is presented in the second display area.

2. The method according to claim 1, wherein The second display area includes a content presentation sub-area, and the content presentation sub-area is used to present social content sent to the social object; Then presenting the first target content or the second target content in the second display area includes: The first target content or the second target content is presented in the content presentation sub-area.

3. The method according to claim 1, wherein The triggering operation in response to the first target content in the at least one first content further includes: presenting at least one candidate style instruction associated with the first target content in the first display area, wherein each candidate style instruction is used to indicate a content style of content generated by the content generation model; In response to a triggering operation on a target style instruction in the at least one candidate style instruction, at least one third content is updated and presented in the first display area, where the at least one third content is generated by the content generation model according to the content style corresponding to the target style instruction.

4. The method according to claim 3, wherein Before presenting at least one candidate style instruction associated with the first target content in the first display area, the method further includes: Performing intent recognition on the first target content to obtain a first intent recognition result; Performing intent recognition on at least one input content to obtain a second intent recognition result; wherein each input content is content obtained according to a received input operation, and the at least one first content is generated by the content generation model according to the at least one input content; determining, based on at least one of the first intent recognition result and the second intent recognition result, a content intent corresponding to the first target content; The at least one candidate style instruction that matches the content intent corresponding to the first target content is obtained.

5. The method according to claim 3, wherein The presenting of the first interface includes: presenting at least one candidate adjustment instruction in the first display area, wherein each candidate adjustment instruction is used to represent an adjustment requirement for the content output by the content generation model; Then presenting at least one candidate style instruction associated with the first target content in the first display area includes: updating the at least one candidate adjustment instruction to present the at least one candidate style instruction; or The at least one candidate adjustment instruction and the at least one candidate style instruction are presented in the first display area.

6. The method according to claim 3, wherein Before presenting at least one candidate style instruction associated with the first target content in the first display area, the method further includes: Determining that a condition for enabling a continuous input mode is met; wherein, in the continuous input mode, each time content is output to the target application, an update of content presented in the first display area is triggered; Then, presenting at least one candidate style instruction associated with the first target content in the first display area includes: When it is determined that the continuous input mode is in place, the at least one candidate style instruction is presented in the first display area.

7. The method according to claim 6, wherein The opening condition includes at least one of the following conditions: receiving a preset operation triggered by content presented in the first display area; receiving a start instruction corresponding to the continuous input mode; Determining, based on at least one input content, a content intention to enable a continuous input mode; Determine, based on at least one social content sent historically, whether the content has the intention to enable the continuous input mode.

8. The method according to claim 6, wherein The second display area further includes a content input sub-area, and the content input sub-area is used to present the content currently edited by the input function. The method further includes: When determining that the end condition of the continuous input mode is met, closing the continuous input mode; In response to a triggering operation on a third target content in the at least one second content, presenting the third target content in the content input sub-area; In response to a sending operation performed on the third target content presented in the content input sub-area, the third target content is presented in the second display area.

9. The method according to claim 8, wherein The termination condition includes at least one of the following conditions: receiving a triggering operation for presenting a content editing sub-area of ​​the input function in the first display area, the content editing sub-area being used for editing content; Receiving a closing instruction corresponding to the continuous input mode; Determining, based on at least one input content, a content intention to close the continuous input mode; Determine, based on at least one social content sent historically, whether the content has a content intention to close the continuous input mode.

10. The method according to any one of claims 1 to 9, wherein: The second display area includes a content input sub-area, and the content input sub-area is used to present the content currently edited by the input function. The method further includes: In response to a triggering operation on a fourth target content in the at least one first content, presenting the fourth target content in the content input sub-area; and presenting at least one candidate adjustment instruction in the first display area, wherein each candidate adjustment instruction is used to represent an adjustment requirement for the content output by the content generation model, and the at least one candidate adjustment instruction is determined based on the fourth target content; In response to a triggering operation on a target adjustment instruction in the at least one candidate adjustment instruction, at least one fourth content is updated and presented in the first display area, where the at least one fourth content is generated by the content generation model according to the adjustment requirement corresponding to the target adjustment instruction.

11. The method according to claim 10, wherein Before presenting at least one candidate adjustment instruction in the first display area, the method further includes: presenting at least one original instruction in the first display area, the original instruction including at least one of a type instruction and an adjustment instruction, the type instruction being used to indicate a type of content generated by the content generation model; Then, presenting at least one candidate adjustment instruction in the first display area includes: updating the at least one original instruction to present the at least one candidate adjustment instruction; or The at least one original instruction and the at least one candidate adjustment instruction are presented in the first display area.

12. The method according to claim 10, wherein Determining at least one candidate adjustment instruction includes at least one of the following methods: determining, according to the content intent corresponding to the fourth target content, the at least one candidate adjustment instruction; determining, according to a page scene corresponding to the second display area, the at least one candidate adjustment instruction; determining the at least one candidate adjustment instruction according to the change in the content input sub-area; The at least one candidate adjustment instruction is determined according to the content state in the content input sub-area.

13. The method according to claim 12, wherein: The content status includes the content length; and determining the at least one candidate adjustment instruction according to the content status in the content input sub-area includes: Determining a target interval range for the content length in the content input sub-area from at least one preset interval range; The at least one candidate adjustment instruction is determined according to a candidate adjustment instruction set corresponding to the target interval range.

14. The method according to any one of claims 1 to 9, wherein: The method further comprises: In response to an input operation triggered on the target application, presenting the first display area, wherein the first display area includes a content editing sub-area, and the content editing sub-area is used to edit content; and Prompt information is presented in other areas of the first display area except the content editing sub-area, and the prompt information is used to prompt the use of the content generation model to generate content. The prompt information includes at least one of the fifth content generated by the content generation model and content generation instructions. The prompt information is determined based on the page scene of the second display area and at least one of the content themes associated at the current moment. The content generation instructions are used to indicate the use of the content generation model to generate the required content.

15. The method according to claim 14, wherein After presenting prompt information in other areas of the first display area except the content editing sub-area, the method further includes any of the following steps: In response to a triggering operation on the prompt information, presenting a content generation sub-area in the first display area, wherein the content generation sub-area is used to present content generated by the content generation model; When the prompt information includes the fifth content, in response to a triggering operation on the fifth content, presenting the fifth content in a content presentation sub-area included in the second display area, the content presentation sub-area being used to present social content sent to the social object; When the prompt information includes the fifth content, in response to a triggering operation on the fifth content, presenting the fifth content in a content input sub-area included in the second display area, the content input sub-area being used to present the content currently being edited by the input function; When the prompt information includes the content generation instruction, in response to a triggering operation on the content generation instruction, a content generation sub-area is presented in the first display area, and the content generation sub-area includes content generated by the content generation model according to the content generation instruction.

16. A content input device, characterized in that: The device comprises: an interface presenting unit, configured to present a first interface, the first interface comprising a first display area corresponding to an input function and a second display area corresponding to a target application, the first display area presenting at least one first content generated by a content generation model; a content updating unit, configured to, in response to a triggering operation on a first target content in the at least one first content, present the first target content in the second display area, and update and present at least one second content in the first display area, where the at least one second content is generated by the content generation model according to the content intent corresponding to the first target content; The content updating unit is further configured to present the second target content in the second display area in response to a triggering operation on a second target content in the at least one second content.

17. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 15 are implemented.

18. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 15 are implemented.

19. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 15 are implemented.

Images