Interaction method and electronic device

Abstract

This application provides an interaction method and an electronic device, relating to the field of artificial intelligence technology. The interaction method includes: during the output display of first content on an interface, obtaining a first additional input for a first segment; the first segment being a portion of content displayed on the interface, the first content being generated by a model based on a global input; generating second content based on the first additional input; presenting the second content as an additional output of the first segment at a position associated with the first segment; the global input triggering a new round of dialogue, and the first additional input not affecting the processing of the global input.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to an interaction method and electronic device. Background Technology

[0002] During the dialogue process of the model, when users read the answers generated by the model, they often have difficulty understanding certain specific content in the answers or need further clarification. However, the relevant technologies are mainly geared towards processing complete dialogue rounds and lack an interactive method for efficiently asking follow-up questions about specific parts of the answers. Summary of the Invention

[0003] In view of this, this application provides an interaction method and an electronic device.

[0004] According to a first aspect of this application, an interaction method is provided, comprising: obtaining a first additional input for a first segment during the output display of first content on an interface; the first segment being a portion of content displayed on the interface, the first content being generated by a model for a global input; generating second content based on the first additional input; presenting the second content as an additional output of the first segment at a position associated with the first segment; and triggering a new round of dialogue with the global input, wherein the first additional input does not affect the processing of the global input.

[0005] According to a second aspect of this application, an electronic device is provided, including a display screen for displaying an interface; a memory for storing computer instructions and a model; the computer instructions are loaded by at least one processor and are configured to: obtain a first additional input for a first segment during the output display of the interface based on first content; the first segment is a portion of the content displayed in the interface, and the first content is generated by the model in response to global input; generate second content based on the first additional input; present the second content as an additional output of the first segment at a position associated with the first segment; and trigger a new round of dialogue with the global input, wherein the first additional input does not affect the processing of the global input.

[0006] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent from the following description. Attached Figure Description

[0007] The above and other objects, features and advantages of this application will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0008] Figure 1 An application scenario diagram of an interaction method and electronic device provided in the embodiments of this application;

[0009] Figure 2 This is a flowchart illustrating one of the interaction methods provided in an embodiment of this application;

[0010] Figure 3 This is a schematic diagram illustrating an application scenario for content interaction processing provided in an embodiment of this application;

[0011] Figure 4 The illustration shows an application scenario diagram of the second content presentation method provided in the embodiments of this application;

[0012] Figure 5 This is a second flowchart illustrating an interaction method provided in an embodiment of this application;

[0013] Figure 6 This is a flowchart of an interaction method provided in an embodiment of this application;

[0014] Figure 7 This is a flowchart of an interaction method provided in an embodiment of this application;

[0015] Figure 8 This is a fifth flowchart illustrating an interaction method provided in an embodiment of this application;

[0016] Figure 9 This is a flowchart of an interaction method provided in an embodiment of this application;

[0017] Figure 10 This is a block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0018] The embodiments of this application will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of this application. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of this application for ease of explanation. However, it will be apparent that one or more embodiments may be implemented without these specific details. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concepts of this application.

[0019] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of this application. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0020] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0021] When using expressions such as "at least one of A, B, and C", they should generally be interpreted in accordance with the meaning that is commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include, but is not limited to, a system having A alone, having B alone, having C alone, having A and B, having A and C, having B and C, and / or having A, B, and C, etc.).

[0022] This application provides an interaction method and an electronic device.

[0023] The interaction method includes: obtaining a first additional input for the first segment during the output display of the first content on the interface; the first segment is a part of the content displayed on the interface, and the first content is generated by the model for the global input; generating second content based on the first additional input; presenting the second content as an additional output of the first segment at a position associated with the first segment; and triggering a new round of dialogue with the global input, wherein the first additional input does not affect the processing of the global input.

[0024] By adopting the technical solution of this application, a first supplementary input for the first segment is obtained during the first content output display process. The first segment is a portion of the content displayed on the interface rather than a complete answer, enabling targeted follow-up questions on local content within the answer. After generating second content based on the first supplementary input, the second content is presented as an additional output of the first segment at a position associated with the first segment. The follow-up question result is directly associated with the original content position that raised the question, allowing users to understand the correspondence between the follow-up question result and the original content without switching between content in different positions. Simultaneously, the first supplementary input does not affect the processing of global input, and the processing of local follow-up questions is independent of the main dialogue flow, maintaining the integrity of the original dialogue while achieving efficient follow-up questions on local content. Compared to related technologies that primarily focus on processing complete dialogue rounds, this application provides an interactive method for efficiently following up on local content within the answer, improving the efficiency and user experience of the dialogue application.

[0025] Figure 1 This is an application scenario diagram of an interaction method and electronic device provided in an embodiment of this application.

[0026] like Figure 1As shown, the application scenario according to the embodiments of this application may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 serves as a medium for providing a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired or wireless communication links or fiber optic cables. For example, a user can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 through the network 104 to receive or send information, etc.

[0027] The first terminal device 101, the second terminal device 102, and the third terminal device 103 can be electronic devices with display and interactive capabilities, such as smartphones, tablets, laptops, personal computers, smart wearable devices, and in-vehicle terminals. A target application can be installed and run on the terminal device. This target application can include, but is not limited to, applications that support model interaction, such as conversational applications, smart assistant applications, search applications, content creation assistance applications, online education applications, and office collaboration applications (for example only). Furthermore, this application embodiment does not limit the form of the target application, and includes, but is not limited to, applications installed on the terminal device, client programs, mini-programs, web applications, etc.

[0028] Server 105 can be a server providing various services, such as a backend server supporting users in running a target application using the first terminal device 101, the second terminal device 102, and the third terminal device 103 (this is just an example). The backend server can process received user input, call a model to generate corresponding content, and then feed the generated content back to the terminal devices. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services such as cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks, and basic cloud computing services such as big data and artificial intelligence. The server can be a backend server for the aforementioned target application, providing backend support such as model services for the target application.

[0029] It should be noted that the interaction method provided in the embodiments of this application can generally be executed by terminal devices 101-103. Accordingly, the interaction device provided in the embodiments of this application can generally be set in terminal devices 101-103. In some scenarios, the interaction method provided in the embodiments of this application can also be executed by server 105, and accordingly, the interaction device can also be set in server 105.

[0030] It should be understood that Figure 1The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0031] The following will be based on Figure 1 The following describes the interaction method of the embodiments of this application in detail, based on the described scenario.

[0032] Figure 2 This is a flowchart of one of the interaction methods provided in an embodiment of this application.

[0033] like Figure 2 As shown, the interaction method can specifically include the following operations.

[0034] In operation S210, during the output display of the first content on the interface, a first appended input for the first segment is obtained; the first segment is a part of the content displayed on the interface, and the first content is generated by the model based on the global input.

[0035] Operation S220 generates the second content based on the first appended input;

[0036] Operation S230 presents the second content as an additional output of the first segment at the position associated with the first segment; wherein, the global input is used to trigger a new round of dialogue, and the first additional input does not affect the processing of the global input.

[0037] In operation S210, the first content refers to the response generated by the model based on the global input submitted by the user. Global input refers to the question or command submitted by the user at the main input box entry point, which triggers the model to perform a new round of dialogue processing. Each input submitted from the main input box entry point triggers a complete call process of the model, generating the corresponding response content. If a new input is submitted again from the main input box entry point during the first content generation process, it will interrupt the current content generation process.

[0038] For example, if a user submits question 2 through the main input field while the model is generating an answer to question 1, the system should either prompt the user that new input cannot be sent at this time, or stop generating an answer to question 1 and switch to processing question 2. Based on this interaction method, it is difficult for users to ask follow-up questions about specific parts of the answer without interrupting the current response.

[0039] The model can generate corresponding answers based on the semantics of the global input, and these answers are displayed as the primary content in the interface. For example, the primary content can include, but is not limited to, dialogue answers, question answers, text, code, tables, images, audio data, and other forms of content generated by the model.

[0040] Based on the above-described process of generating the first piece of content, users may have questions or need further information about certain parts while reading it. In this case, users can select the first segment according to their needs.

[0041] The first segment refers to a portion of the first content, which can be understood as a local area selected or focused on by the user within the first content. This local area can be one or more sentences, paragraphs, phrases, or a specific expression within the first content. For example, the first segment could be a sentence in the first content whose meaning the user has questions about. Alternatively, the first segment could be a technical term or jargon in the first content, for which the user wants a detailed explanation.

[0042] For the first segment, the user can submit the first additional input. The first additional input refers to the input content submitted by the user for the first segment, which can be understood as a prompt word for the first segment, or used to express the user's further needs for the first segment.

[0043] For example, when the first segment is a description of the application of quantum entanglement in information transmission, the user can enter a request for a detailed explanation of the principles of quantum entanglement, or inquire about specific application scenarios. Similarly, when the first segment is a piece of code, the user can enter a request for the code's function, optimization suggestions, or a request to invoke a relevant program to simulate running the code.

[0044] In one implementation, an input control can be triggered after the user selects a first segment. The user then enters text content into this input control as the first appended input. The user can select a specific text range within the first segment by dragging with the mouse or by touching the screen. After detecting the selection, the system displays an input box or input button near the selected area. The user clicks the input box or input button to enter text, which becomes the first appended input.

[0045] In another implementation, a trigger area for appending input can be preset in the display area of ​​the first content. When the user interacts with the trigger area, the first appended input is obtained. An interactive icon can be set next to each paragraph or sentence of the first content. After the user clicks the icon, an input interface pops up. The content entered by the user in the input interface serves as the first appended input for the corresponding paragraph or sentence. This implementation eliminates the need for the user to perform a text selection operation first, reducing operational complexity.

[0046] In another implementation, the first additional input can be obtained via voice. After selecting a first segment, the user can speak their question or request regarding that segment through a voice input device. The system then converts the voice content into text format as the first additional input.

[0047] It should be noted that the acquisition of the first additional input occurs during the display of the first content. This process can be either the gradual generation and real-time display of the first content, or the user's browsing and viewing of the first content after it has been fully displayed. As long as the first content is displayed on the interface, the user can submit additional input for any segment of it. This allows the user to ask questions at any stage of reading the first content without waiting for the content to be fully generated or switching to a new dialogue turn.

[0048] In operation S220, the second content refers to the response content generated in response to the first additional input, which can be understood as a supplementary explanation, detailed explanation, or extended processing result based on the first fragment.

[0049] In one implementation, the first appended input can be directly processed by the model as a prompt word to generate the second content. Upon receiving the first appended input, the system can pass it to the model as an independent question or instruction. The model then generates a corresponding response based on the semantics of the first appended input, which becomes the second content.

[0050] In another implementation, the first fragment and the first appended input can be combined and input into the model to generate second content. The system can concatenate or combine the content of the first fragment with the content of the first appended input to form a new prompt. For example, the first fragment can be used as context information, and the first appended input can be used as a question for that context. The combined prompt can be in the form of "For the following content: [first fragment], answer the question: [first appended input]". The model generates the second content based on this combined prompt.

[0051] It's important to note that the generation of the second content is independent of the generation of the first content. Even if the first content is not yet fully generated, the generation of the second content will not interrupt the generation of the first content. The global input triggers a new round of dialogue addressing the overall question, while the first appended input only addresses a local segment of the first content. The submission and processing of the first appended input do not affect the processing of the global input. While the model is generating the second content in response to the first appended input, the process of generating the first content based on the global input can continue.

[0052] In practice, the generation of the first content and the generation of the second content can be performed as two tasks in parallel by the model. These two tasks can be handled by the same model or by two different models.

[0053] When using the same model, the model can handle two generation tasks simultaneously through multi-threading or multi-processing, maintaining two independent generation states respectively. When using two different models, the first model can be responsible for generating the first content, and the second model can be responsible for generating the second content. The two models run independently without interfering with each other, but the second model inherits the context of the first model, allowing multiple models to logically form a continuous thought process under the same brain, rather than being isolated islands. Inheritance methods can include sharing a context pool, packaging the memory of the previous model and passing it to the next model as input, or a main model + sub-model, etc. This embodiment does not limit these methods. This allows users to flexibly ask follow-up questions and explore specific content without affecting the original dialogue flow.

[0054] In operation S230, additional output refers to extra content generated for the first segment. It can be understood as a supplement or extension to the first segment, used to provide users with detailed information about the local segment without changing the main structure of the first content. The location associated with the first segment refers to the area in the interface that is spatially or logically related to the first segment. This area allows users to intuitively identify the correspondence between the second content and the first segment.

[0055] In one implementation, the second content can be presented directly below or to the side of the first segment. Once the second content is generated, the system can create a new display area adjacent to the text area containing the first segment and display the second content in that area. For example, a display area for the second content can be inserted below the first segment, and visual elements such as indentation, borders, or background color can be used to distinguish the second content from other parts of the first content.

[0056] It should be noted that the presentation of the second content as supplementary output differs from the presentation of the first content as the main answer. The first content is usually presented in the main display area of ​​the dialog interface, while the second content is presented as an adjunct to the first segment, allowing users to clearly distinguish between the overall answer to the global input and the supplementary explanation for a specific segment.

[0057] Please refer to Figure 3 , Figure 3 This is a schematic diagram illustrating an application scenario for content interaction processing provided in an embodiment of this application.

[0058] like Figure 3As shown, the display screen of the tablet computer 300 is divided into a first split screen 310 and a second split screen 320. The user enters the global input "Introduce the development history of quantum computing" in the input box 301 of the first split screen 310. The model responds to this global input and begins generating the first content, which is then gradually displayed in the display area 302. The first content includes an introduction to the development history of quantum computing, including content such as "quantum superposition is one of the core theories of quantum computing development."

[0059] While the first content is still being generated, the user has questions about the term "quantum superposition state" in the first content. They select the phrase as the first segment 303 via touch operation 304 and enter the first additional input 305, "Please explain the concept of quantum superposition state in detail," in the pop-up input control. Upon receiving the first additional input, the server generates the second content 306 based on the first additional input 305 and the first segment 303 without interrupting the model's generation of the first content. The second content 306, as an additional output of the first segment 303, is presented in the second split screen 320, providing a detailed explanation of the concept of quantum superposition state.

[0060] Furthermore, users can ask the model further questions in input box 307, such as "Please introduce quantum computing chips again," and the model will continue to output new content in the first split screen 310. Thus, users can continuously obtain complete answers to global questions while simultaneously viewing detailed supplementary explanations for specific local segments in the second split screen 320, achieving parallel processing of global dialogue and local exploration.

[0061] By adopting the technical solution of this embodiment, a first additional input for the first segment is obtained during the first content output display process. The first segment is a portion of the content displayed on the interface rather than a complete answer, enabling targeted follow-up questions on local content within the answer. After generating second content based on the first additional input, the second content is presented as an additional output of the first segment at a position associated with the first segment. The follow-up question result is directly associated with the original content position that raised the question, allowing users to understand the correspondence between the follow-up question result and the original content without switching between content in different positions. Simultaneously, the first additional input does not affect the processing of global input, and the processing of local follow-up questions is independent of the main dialogue flow, maintaining the integrity of the original dialogue while achieving efficient follow-up questions on local content. Compared to related technologies that primarily focus on processing complete dialogue rounds, this application provides an interactive method for efficiently following up on local content within an answer, improving the efficiency and user experience of the dialogue application.

[0062] Figure 4 The illustration shows an application scenario diagram of the second content presentation method provided in the embodiments of this application.

[0063] Based on the above embodiments, as an optional embodiment, the operation S230 of presenting the second content as an additional output of the first segment at a position associated with the first segment may include at least one of the following operations.

[0064] Operation S231: The second content is used as an additional output of the first segment and displayed in an embedded form as a card below the first segment.

[0065] Embedded display refers to inserting the second content directly into the display area of ​​the first content without altering the original text structure of the first content. Card format refers to using a container with independent boundaries and visual style to hold the second content.

[0066] Optionally, the system creates a display area below the location of the first segment, presenting the second content in a card-like format within this area. The card-like format may include visual elements such as borders, background colors, rounded corners, and shadows to distinguish the second content from other parts of the first content.

[0067] like Figure 4 As shown, the first content 401 includes a first segment 402. When the user submits a first append input for the first segment 402, the second content 404A of the first display mode is displayed as a card below the first segment 402. This card is embedded in the text stream of the first content 401, and when the user scrolls down the first content 401, they can directly see the second content 404A of the first display mode corresponding to the first segment 402.

[0068] Operation S232: The second content is used as an additional output of the first segment, and the first additional input, the first segment, and the second content are displayed in the side area of ​​the interface.

[0069] The side area refers to the side area of ​​the interface that is independent of the main content display area. This area can be the left or right side of the interface, or it can be an independent display area divided by a split-screen method.

[0070] Optionally, the system creates a side area on the side of the interface, where the first additional input, the first fragment, and the second content are displayed sequentially. The first fragment can be displayed as reference content, indicating to the user which part of the first content the current follow-up question refers to. The first additional input is displayed as a question or instruction, explaining the user's specific needs regarding the first fragment. The second content is displayed as an answer, presenting supplementary explanations generated in response to the first fragment and the first additional input.

[0071] like Figure 4As shown, the interface is divided into two areas, left and right. The left area displays the first content 401, while the right side area displays the first fragment 402, the first appended input 403, and the second content 404B displayed in a second mode. Users can continue to browse the first content 401 in the left area while viewing the follow-up questions about the first fragment 402 in the right area. The display of the two areas does not interfere with each other.

[0072] Operation S233: The second content is displayed as an additional output of the first segment in the form of a comment at the corresponding position of the first segment.

[0073] Annotation format refers to presenting second content in a way similar to document annotations or markings, so that the second content is attached to the position of the first segment in a floating, pop-up or annotation style.

[0074] Optionally, the system sets an annotation marker at the location of the first segment. When the user clicks the marker or hovers the mouse over the first segment, the second content is displayed in an annotation style. The annotation style can be a speech bubble floating above the first segment or a label box displayed to the side of the first segment. The display and hiding of the annotation style can respond to user interaction, automatically hiding it when the user does not need to view it, thereby reducing visual interference with the reading of the main content.

[0075] like Figure 4 As shown, annotation marks are displayed at the corresponding positions of the first segment 402, and the second content 404C, displayed in the third mode, is presented as an annotation above the first segment 402. When the user clicks or touches the first segment 402, the second content 404C in the third mode can be expanded to view, and the annotation automatically collapses when the user clicks outside the annotation area.

[0076] Operation S234: The second content is used as an additional output of the first segment, and the second content is presented in an expanded or collapsed state at the position associated with the first segment.

[0077] The expanded state refers to the state where the complete content of the second element is directly displayed on the interface. The collapsed state refers to the state where the second element is collapsed or hidden, and only the title, summary, or entry control is displayed.

[0078] Optionally, the system sets up a display area at the location associated with the first segment to hold the second content. This display area can switch between an expanded state and a collapsed state. When in the collapsed state, the display area only shows the title of the second content or a brief message of the first appended input, as well as a control for expanding it. When the user clicks the expand control, the display area switches to the expanded state, and the complete content of the second content is displayed in this area. The user can click the collapse control again to restore the display area to the collapsed state.

[0079] like Figure 4 As shown, a display control 405 is displayed below the first segment 402. This display control 405 includes a "Details" control and a "Close" control. When the second content is in a collapsed state, clicking the "Details" control expands the second content. When the second content is in an expanded state, clicking the "Close" control collapses or hides the second content.

[0080] Optionally, the system can automatically determine the initial display state based on the length or type of the second content. For example, when the length of the second content exceeds a preset threshold, the system automatically displays it in a collapsed state initially. When the second content is a brief explanation, the system automatically displays it in an expanded state initially.

[0081] Optionally, operations S231 to S234 can be used individually or in combination.

[0082] In one implementation, the user can customize the presentation method of the second content. The system provides the user with an entry point for selecting the presentation method, and the user selects one or more presentation methods according to their current reading needs. The system then presents the second content according to the user's selection.

[0083] In another implementation, the presentation method of the second content can be automatically selected based on the display area of ​​the interface. When the display area is greater than a preset threshold, the system uses a side area display or a card-style expanded display. When the display area is less than the preset threshold, the system uses an annotation or card-style collapsed display.

[0084] In another implementation, multiple presentation methods can be used simultaneously to display the second content. For example, the system can display the second content in card form below the first segment while simultaneously displaying the complete record of the follow-up question in a side area. In this implementation, the system can present content with different levels of detail based on the display area corresponding to different presentation methods. Specifically, the display area of ​​the first display area is larger than that of the second display area; the system displays a summary of the second content in the second display area and detailed information about the second content in the first display area.

[0085] By adopting the technical solution of this embodiment, multiple presentation methods for the second content are provided to meet the display needs in different scenarios. The card-style embedded display directly links the second content to the first segment, facilitating simultaneous viewing of supplementary explanations during reading. The side-area display method presents the first and second content separately, avoiding interference with the main structure of the first content. The annotation-style display provides a more flexible viewing method, allowing users to expand or hide the second content at any time as needed. The expand and collapse switching function allows users to control the display of the second content according to actual needs, expanding it for detailed viewing and collapsing it to focus on the main content, improving the flexibility and usability of the interface.

[0086] The above embodiments have provided a detailed description of the presentation method of the second content. The generation process of the second content will be described below.

[0087] Figure 5 The second flowchart is shown as an embodiment of an interaction method provided in this application.

[0088] Based on the above embodiments, as an optional embodiment, the above operation S220 may specifically include the following operations.

[0089] Operation S510 determines the corresponding first anchor point based on the first segment;

[0090] Operation S520: Obtain context information from the first content based on the first anchor point; the context information includes the first segment and content segments located within a predetermined range before and after the corresponding position of the first anchor point.

[0091] Operation S530 generates second content based on the first appended input and context information.

[0092] In operation S510, the first anchor point refers to the positioning information used to identify the position and content characteristics of the first segment in the first content. It can be understood as the association basis for binding the first segment with the subsequently generated second content.

[0093] Optionally, after the user selects the first segment, the system obtains the location information and text information of the first segment within the first content. The system generates a first anchor point based on the obtained location and text information. The location information is used to identify the specific location of the first segment within the first content. The text information is used to identify the content features of the first segment.

[0094] In one implementation, the position information may include the starting character position and the ending character position of the first segment. The system may assign character numbers to the first content, assigning a sequential position number to each character in the first content. When a user selects the first segment, the system can determine the starting character position of the first segment within the first content based on the selection range, and determine the ending character position of the first segment within the first content as the ending character position.

[0095] In another implementation, the text information may include the original text content of the first segment. The system can extract the text content corresponding to the first segment selected by the user and record that text content as part of the first anchor point.

[0096] In another implementation, the text information may further include text feature information generated based on the first segment. The system can extract features from the original text content of the first segment to generate information that identifies the features of the segment's content. The text feature information may be a checksum generated based on the text content and length of the first segment, or it may be a feature identifier generated based on the keywords of the first segment.

[0097] Optionally, the first anchor point may also include reference text before and after the first segment. Reference text refers to the text content located before and after the first segment, used to help confirm the position of the first segment within the first content. The system can extract a predetermined number of characters of text before the first segment as the preceding reference text, extract a predetermined number of characters of text after the first segment as the following reference text, and use the preceding and following reference texts as components of the first anchor point.

[0098] It should be noted that the first anchor point is used to establish the relationship between the first segment and the second content, so that the second content can accurately correspond to the position of the first segment. Through the first anchor point, the system can locate the first segment in the first content, thereby presenting the second content in the position associated with the first segment.

[0099] In operation S520, contextual information refers to the relevant text content provided to the model when generating the second content, which enables the model to understand the context and background of the first segment.

[0100] Optionally, the system determines the position of the first segment within the first content based on the position information in the first anchor point. Using this position as a reference, the system extracts content segments within a predetermined range both forward and backward. The predetermined range can be a preset number of characters, sentences, or paragraphs. The system combines the first segment with the extracted preceding and following content segments to form contextual information.

[0101] In one implementation, one or more sentences preceding and following the first segment can be extracted as part of the context information. The system can segment the first content into sentences and identify the sentence boundaries of the first segment. The system extracts several sentences preceding and following the first segment, and combines these sentences with the first segment to form the context information.

[0102] In another implementation, the scope of context information extraction can be determined based on the length of the first segment. When the first segment is short, the system can extract a wide range of preceding and following content segments to supplement the context information. When the first segment is long, the system extracts a small range of preceding and following content segments or only the first segment itself. For example, when the first segment is a single word, the system extracts the complete sentence containing that word, along with the sentences before and after it, as context information. When the first segment is a complete paragraph, the system can extract only that paragraph and its title as context information.

[0103] In another implementation, the scope of context information extraction can be determined based on the type of the first appended input. When the first appended input requests an explanation of the meaning of a term, the system can extract the first segment and the sentence it contains as context information. When the first appended input requests an explanation of the reasoning basis or a verification conclusion, the system extracts the first segment and a large portion of the preceding content as context information.

[0104] It should be noted that the contextual information includes at least the first segment itself, as well as the adjacent content before and after it. By providing contextual information, the model can understand the specific context of the first segment within the first content when generating the second content, thereby generating more accurate and relevant supplementary explanations.

[0105] In operation S530, the system processes the first appended input and contextual information together as a prompt word input model.

[0106] In one implementation, the scope of the first segment can be clearly defined in the prompt. The system can mark the first segment in contextual information, such as by adding markers before or after the first segment or using a specific format to identify it. This allows the model to clearly distinguish the first segment from its preceding and following content, thereby providing a more accurate answer based on the first segment.

[0107] In another implementation, instructions regarding the requirements for generating the second content can be added to the prompt. The system can add formatting requirements, detail requirements, or content type requirements for the second content to the prompt.

[0108] Optionally, the system can control the length of the context information. When the length of the context information exceeds a preset threshold, the system compresses or truncates the context information, retaining only the part most relevant to the first segment. For example, it can retain only the complete sentence containing the first segment, as well as adjacent sentences that have a referential or logical relationship with the first segment.

[0109] By adopting the technical solution of this embodiment, and generating a first anchor point for the first segment, the system can accurately identify the position and content features of the first segment within the first content, providing positioning for the subsequent presentation of the second content. Based on the first anchor point, contextual information including the first segment and its preceding and following adjacent content is obtained, enabling the model to understand the specific context of the first segment when generating the second content, avoiding semantic bias caused by interpreting the first segment in isolation. Providing the first supplementary input along with the contextual information to the model allows the model to generate accurate supplementary explanations based on the user's specific needs and the actual context of the first segment, improving the accuracy of the second content.

[0110] Based on the above embodiments, as an optional embodiment, the above interaction method may further include the following steps.

[0111] At least one of the following is identified as context information: a summary of the first content and similar content fragments retrieved based on the text feature information of the first fragment.

[0112] In one implementation, the system may use a summary of the first content as part of the contextual information. The summary is used to summarize the main points and core content of the first content.

[0113] The system can extract a summary from the first piece of content, obtaining a summary that reflects the overall theme and main information of the first piece of content. The system combines the summary with the first piece of content to form contextual information. By providing the summary, the model can understand the overall background and thematic direction of the first piece of content when generating the second piece of content.

[0114] In another implementation, the system can retrieve similar content fragments based on the textual feature information of the first fragment, and use the retrieved similar content fragments as part of the context information. Similar content fragments refer to other text fragments that are semantically or content-related to the first fragment.

[0115] The system can extract textual feature information from the first segment of the first content, and use this textual feature information as a search condition to search other parts of the first content or an external knowledge base. The system obtains content segments with similar semantics or related topics to the first segment, and combines these similar content segments with the first segment to form contextual information.

[0116] In another implementation, the system can use a summary of the first content and similar content fragments together as contextual information. The system first extracts the summary of the first content as overall background information, and then retrieves content fragments similar to the first fragment as supplementary detailed information. The system combines the summary, the first fragment, and similar content fragments to form complete contextual information.

[0117] Optionally, when the context information contains content from multiple sources, the system can prioritize the content. The system can place the first segment as the core content at the center of the context information, place the summary as overall background information at the beginning, and place similar content segments as supplementary information at the end.

[0118] By employing the technical solution of this embodiment, and by providing a summary of the first content as contextual information, the model can understand the overall background of the first segment, avoiding the situation of focusing only on local content while ignoring the overall theme. By providing similar content segments retrieved based on the first segment, the model can refer to more relevant information, improving the comprehensiveness and accuracy of the second content.

[0119] Based on the above embodiments, as an optional embodiment, the operation of presenting the first segment at the position associated with the first segment in the above operation S230 may specifically include at least one of the following operations.

[0120] Operation S235: Determine the display position of the first segment in the first content based on the first anchor point, and present the second content in the adjacent area corresponding to the display position;

[0121] In operation S236, when the first content changes, the first segment is repositioned based on the first anchor point to maintain the association display relationship between the second content and the first segment;

[0122] In operation S237, when the content of the first segment changes, the second content is regenerated based on the first anchor point, and the content display at the associated position is updated.

[0123] In operation S235, the system determines the display position of the first segment within the first content based on the position information recorded in the first anchor point. The system then presents the second content in the adjacent area corresponding to that display position. The adjacent area can be determined by referring to the presentation methods of operations S231 to S234 described above, and will not be elaborated further here.

[0124] In operation S236, the change of the first content refers to the continuous addition of new content during the streaming generation process of the first content, or the change of the text position of the first content due to reasons such as re-layout or font adjustment.

[0125] When the first content changes, the character position of the first segment within the first content may change accordingly. The system repositions the first segment based on the first anchor point to ensure that the second content is still displayed in the position associated with the first segment. After the system completes the repositioning, it updates the position information recorded in the first anchor point, so that the display position of the second content follows the new position of the first segment.

[0126] In operation S237, the change in the content of the first segment means that while the first content is still being generated, the text content at the corresponding position of the first segment has been changed or supplemented.

[0127] When the system detects a change in the content of the first segment, it re-acquires the changed segment content based on the first anchor point. This changed content, along with the first appended input, is used as the prompt word input model to generate updated second content. The system then displays the updated second content at the position associated with the first segment, replacing the original second content.

[0128] In one implementation, the system can periodically detect the content of the first segment based on the text feature information recorded in the first anchor point. When the system detects that the content of the first segment is inconsistent with the original text of the segment recorded in the first anchor point, it determines that the content of the first segment has changed and triggers the regeneration of the second content.

[0129] In another implementation, the system can monitor content changes at corresponding positions of the first segment during the generation of the first content. When a change in the content at that position is detected, the system triggers the regeneration of the second content in real time, ensuring that the second content remains consistent with the latest content of the first segment.

[0130] It should be noted that there is no fixed execution order between operations S231 to S234 and operations S235 to S237. The system can execute any of the operations S235 to S237 simultaneously with or after executing any of the operations S231 to S234. For example, the system can reposition the first segment while presenting the second content in card format. Alternatively, after displaying the second content in a side area, the system can regenerate the second content when the content of the first segment changes.

[0131] By adopting the technical solution of this embodiment, the first segment is repositioned based on the first anchor point, ensuring that the second content remains accurately displayed in the position associated with the first segment even when the first content changes, thus maintaining the display relationship between the second content and the first segment. When the content of the first segment changes, the second content is regenerated, ensuring that the second content remains consistent with the latest content of the first segment, thereby improving the accuracy of the second content.

[0132] Figure 6 The third flowchart illustrates an interaction method provided in an embodiment of this application.

[0133] Based on the above embodiments, as an optional embodiment, when the first content changes, the above operation S236 may further include the following operations.

[0134] Operation S610 performs initial positioning of the first segment based on the position information in the first anchor point;

[0135] Operation S620 verifies the initial positioning result based on the text feature information in the first anchor point;

[0136] Operation S630: In case of discrepancy, match based on text similarity in the first content;

[0137] In operation S640, if a content segment corresponding to the first segment exists, the new position is determined based on the corresponding content segment, and the position information in the first anchor point is updated.

[0138] In operation S610, the location information may include the start character position and the end character position of the first segment. Based on the start character position and the end character position, the system extracts the corresponding range of text content from the first content as the initial positioning result.

[0139] For example, when the starting character position recorded in the first anchor point is 100 and the ending character position is 150, the system extracts the text content between the 100th and 150th characters in the first content as the initial positioning result.

[0140] In operation S620, the system verifies the initial positioning result based on the text feature information recorded in the first anchor point. The text feature information is used to identify the content features of the first segment.

[0141] In one implementation, the text feature information may include the original text content of the first segment. The system compares the initial positioning result with the original text content recorded in the first anchor point. If the two are completely identical, the system confirms that the position of the first segment has not changed.

[0142] In another implementation, the text feature information may include a checksum generated based on the text content and length of the first segment. The system generates a corresponding checksum for the initial positioning result and compares the generated checksum with the checksum recorded in the first anchor point. If the two checksums match, the system confirms that the position of the first segment has not changed.

[0143] It should be noted that inconsistent verification results may be due to the first segment's position within the first content being offset, causing the initial positioning result to point to incorrect text content. It could also be because the text content at the corresponding position of the first segment has been changed or deleted.

[0144] In operation S630, the system performs a text similarity match in the first content to determine whether the first fragment still exists in the first content.

[0145] The system can extract multiple candidate texts from the first content and calculate the similarity between each candidate text and the text feature information recorded in the first anchor point. The system identifies candidate texts with similarity scores higher than a preset threshold as content fragments corresponding to the first segment.

[0146] In one implementation, the system can sequentially extract text segments of the same or similar length as the first segment from the first content as candidate texts. The system calculates the degree of word overlap between each candidate text and the original text of the first segment, and uses the degree of word overlap as the basis for judging similarity.

[0147] In another implementation, the system can perform matching by combining the preceding and following reference texts recorded in the first anchor point. The preceding reference text refers to the portion of text content preceding the first segment, and the following reference text refers to the portion of text content following the first segment. The system searches for positions in the first content that match the preceding and following reference texts, and determines the text between the matching positions as the content segment corresponding to the first segment.

[0148] Optionally, the system may prioritize matching within the vicinity of the original position of the first segment. If no candidate text with a similarity higher than a preset threshold is found within the vicinity, the system expands the matching scope to the entire first content. The vicinity can be a range of a predetermined number of characters before and after the original position, or a range of a predetermined number of sentences before and after the original position.

[0149] In operation S640, the system can obtain the start and end character positions of the corresponding content fragment within the first content, and use this position information as the new position information of the first fragment. The system updates the start and end character positions recorded in the first anchor point, causing the first anchor point to point to the new position of the first fragment.

[0150] The system can adjust the presentation position of the second content based on the updated location information, ensuring that the second content continues to appear in the position associated with the first segment. For example, when the second content is presented as a card below the first segment, the system will adjust the card position of the second content to below the new position of the first segment.

[0151] If no content segment corresponding to the first segment exists in the first content, it indicates that the content of the first segment has been substantially changed or deleted. The system can either notify the user that the content of the first segment has changed in the presentation area of ​​the second content, or hide the presentation of the second content.

[0152] By adopting the technical solution of this embodiment, the system can quickly obtain the expected position of the first segment within the first content through initial positioning based on location information. Through verification based on text feature information, the system can determine whether the content at the expected position is still the first segment. When the verification is inconsistent, by performing global matching based on text similarity within the first content, the system can accurately find the new position of the first segment even if its position has shifted, thus improving the correlation between the second content and the first segment.

[0153] Figure 7 The fourth flowchart illustrates an interaction method provided in an embodiment of this application.

[0154] Based on the above embodiments, as an optional embodiment, when the content of the first segment is changed, the above operation S237 may further include the following operations.

[0155] Operation S710 detects changes in the content of the first segment based on the text feature information in the first anchor point;

[0156] In operation S720, if the content of the first segment changes, the context information is reconstructed based on the changed first segment and the first appended input.

[0157] Operation S730 generates updated second content based on the first appended input and the updated context information;

[0158] Operation S740 replaces the text feature information in the first anchor point with the updated information of the second content, or updates the content display at the associated position of the first segment.

[0159] In operation S710, the system can periodically acquire the current text content of the first segment within the first content and compare the acquired current text content with the text feature information recorded in the first anchor point. If the two match, the system confirms that the content of the first segment has not changed. If the two do not match, the system confirms that the content of the first segment has changed.

[0160] In another implementation, the system can trigger a detection when the first content is updated. When new content is output during the generation of the first content, the system determines whether the newly output content is within the position range corresponding to the first segment. If the newly output content is within the position range of the first segment, the system performs a content comparison on the first segment to confirm whether the content of the first segment has changed.

[0161] Optionally, the system can set conditions for determining content changes. When the change in the text content of the first segment exceeds a preset proportion, the system determines that the content has been changed. When the text content of the first segment only changes punctuation marks or spaces, the system can determine that the content has not been substantially changed.

[0162] In operation S720, the system obtains the current text content of the modified first segment within the first content. Based on this current text content, the system extracts content segments within a predetermined range, both forward and backward. The system combines the modified first segment with the extracted preceding and following content segments to form updated context information.

[0163] For example, if the first segment was originally "Quantum superposition is a core concept of quantum computing" and is changed to "Quantum superposition and quantum entanglement are two core concepts of quantum computing," the system re-extracts adjacent sentences based on the changed text content to construct context information containing the changed first segment. The construction of context information can be performed in the manner described in operation S520 above.

[0164] In operation S730, the system combines the first appended input with the updated context information to form a new prompt word, which is then input into the model for processing. The model understands the modified first segment content based on the updated context information and generates the updated second content by combining it with the specific requirements of the first appended input.

[0165] For example, when the first segment is changed from "Quantum superposition is the core concept of quantum computing" to "Quantum superposition and quantum entanglement are two core concepts of quantum computing", if the first appended input is "Please explain this concept in detail", the second content generated by the system based on the changed first segment will include explanations of both the concepts of quantum superposition and quantum entanglement, instead of only explaining quantum superposition.

[0166] In operation S740, the system acquires the text content of the modified first segment and generates new text feature information based on this text content. The system replaces the original text feature information in the first anchor point with the new text feature information, enabling the first anchor point to accurately identify the modified first segment.

[0167] The system displays the updated second content at the location associated with the first segment, replacing the original second content. If the second content is displayed as a card, the system updates the content within the card. If the second content is displayed in the side panel, the system updates the answer content in the side panel.

[0168] Optionally, the system can provide a change notification to the user when updating the second content. The system can add an indicator to the display area of ​​the second content to notify the user that the answer has been updated according to the changes in the first segment. The user can understand the update status of the second content based on the notification.

[0169] By employing the technical solution of this embodiment, the system can promptly detect changes in the content of the first segment by detecting such changes. Based on the changed first segment, the system reconstructs the context information and generates updated second content, ensuring that the second content remains consistent with the latest content of the first segment and avoiding any mismatch between the two.

[0170] Figure 8 The fifth flowchart illustrates an interaction method provided in an embodiment of this application.

[0171] Based on the above embodiments, as an optional embodiment, the above interaction method may further include the following operations.

[0172] Operate S810 to obtain a second appended input for the second segment of the second content;

[0173] Operate S820 to generate the third content based on the second appended input;

[0174] Operation S830 outputs the third content as an additional output of the second segment, forming a nested display structure with the second segment in a related hierarchy.

[0175] In operation S810, the second fragment refers to the text fragment selected by the user in the second content. The second append input refers to the append request or question made by the user regarding the second fragment.

[0176] When the system presents the second content, the user can select any text segment from it. After the user selects a text segment, the system designates that segment as the second segment. The user can submit additional input for the second segment through text input, voice input, or preset options.

[0177] In operation S820, the system can generate third content in a similar manner to generating the second content. The system determines the corresponding second anchor point based on the second segment and obtains the contextual information of the second segment within the second content. The system combines the second appended input with the contextual information of the second segment to form a prompt word, which is then input into the model to generate the third content.

[0178] It should be noted that the process of generating the third content can refer to the method of generating the second content in operations S510 to S530 above, and will not be repeated here.

[0179] In operation S830, the system outputs the third content as an additional part of the second segment, forming a nested display structure with the second segment in a hierarchical relationship. A nested display structure refers to a display method organized according to the hierarchical relationship of appended inputs, used to reflect the hierarchical relationship between the first, second, and third content.

[0180] In one implementation, the system can represent nesting hierarchy through indentation. The system presents the second content at a position associated with the first segment, and the third content at a position associated with the second segment. The system increases the indentation of the third content by one level relative to the second content, making the third content visually subordinate to the second content, and the second content visually subordinate to the first content.

[0181] In another implementation, the system can represent nested levels using connectors or a tree structure. The system uses the first segment as the root node, the second content as a first-level child node, and the third content as a second-level child node. Connectors link the content at each level, forming a nested tree structure.

[0182] In another implementation, the system can represent the nesting hierarchy through card nesting. The system displays the second content in the first card and the third content in the second card. The system also nests the second card inside or near the first card, using the card nesting relationship to represent the hierarchical relationship of the content.

[0183] Optionally, the system can set different visual styles for content at different levels. The system can set different background colors, border styles, or font sizes for the second and third content, so that users can intuitively identify the hierarchical relationship of the content at each level.

[0184] Optionally, the system can support folding and expanding operations. Users can fold the second content to hide the display of the third content, or expand the second content to display the third content. Through folding and expanding operations, users can control the display status of each level of content as needed.

[0185] It should be noted that users can submit additional input to the third content segment, and the system can generate a fourth content segment, further forming a deeper nested display structure. The system supports multi-level nested displays, enabling users to continuously explore and learn about content of interest.

[0186] By adopting the technical solution of this embodiment, and by supporting the addition of second content to generate third content, the system can support users to further inquire about the supplementary explanations, meeting users' needs for in-depth learning at each level. Through the nested display structure, the system can clearly show the relationship between content at each level, enabling users to intuitively understand the hierarchical relationship of the content and improving the readability of information organization.

[0187] Figure 9 The sixth flowchart illustrates an interactive method provided in an embodiment of this application.

[0188] Based on the above embodiments, as an optional embodiment, the above operation 820 may further include the following operations.

[0189] Operate S910 to determine the corresponding second anchor point based on the second segment;

[0190] Operation S920 retrieves context information from the second content based on the second anchor point;

[0191] Operate S930 to determine the corresponding first anchor point based on the first segment;

[0192] Operation S940: Obtain context information from the first content based on the first anchor point;

[0193] Operator S950 generates third content based on appended input and context information.

[0194] In operations S910 and S920, the system determines a corresponding second anchor point based on the second segment, and obtains contextual information from the second content based on the second anchor point. The second anchor point is used to identify the position and content characteristics of the second segment within the second content. The contextual information in the second content includes the second segment and content segments located within a predetermined range before and after the second segment.

[0195] The method for generating the second anchor point and the method for obtaining the context information in the second content can refer to the processing methods for the first anchor point and the context information in the first content in the above operations S510 and S520, and will not be repeated here.

[0196] In operations S930 and S940, the system determines a corresponding first anchor point based on the first segment, and obtains contextual information from the first content based on the first anchor point. The first anchor point is used to identify the position and content characteristics of the first segment within the first content. The contextual information in the first content includes the first segment and content segments located within a predetermined range before and after the first segment.

[0197] It should be noted that the first anchor point is already determined when generating the second content. When generating the third content, the system can reuse the existing first anchor point and retrieve the context information from the first content again.

[0198] In operation S950, the system generates third content based on appended input and context information. The appended input may include a second appended input, or both a first appended input and a second appended input. The context information may include context information from the first content and context information from the second content.

[0199] In one implementation, the system can generate third content based solely on the second appended input and contextual information from the second content. The system combines the second appended input with the contextual information from the second content to form prompts, and the input model generates the third content. This approach is suitable when the second appended input only prompts further questions about the second segment itself.

[0200] In another implementation, the system can generate third content based on a second appended input, contextual information from the first content, and contextual information from the second content. The system combines the contextual information from the first and second content to form complete contextual information. The system then combines the second appended input with the complete contextual information to form a prompt word, which is input into the model to generate the third content. This approach allows the model to understand the context of both the first and second segments simultaneously when generating the third content, improving the accuracy of the third content.

[0201] In another implementation, the system can generate third content based on a combination of the first and second appended inputs and complete contextual information. The system associates the first and second appended inputs to form a comprehensive appended input containing two levels of follow-up intent. The system combines this comprehensive appended input with the contextual information of the first and second content to form a prompt word, and the input model generates the third content. This approach allows the third content to simultaneously respond to both the first and second levels of follow-up intent.

[0202] Optionally, the system can identify the contextual information of the first and second content when combining contextual information. The system can indicate the source hierarchy of the contextual information of the first and second content in the prompt words, enabling the model to distinguish between different levels of contextual content.

[0203] By adopting the technical solution of this embodiment, and simultaneously acquiring the contextual information of the first and second content, the system can provide the model with a more complete contextual background. By combining multiple levels of additional input and contextual information to generate the third content, the system can generate more accurate and comprehensive supplementary explanations, improving the quality of answers in multi-level follow-up questioning scenarios.

[0204] This application also discloses an electronic device, including:

[0205] A display screen is used to display the interface; a memory is used to store computer instructions and models; the computer instructions are loaded by at least one processor and are used to: obtain a first additional input for a first segment during the output display of the first content on the interface; the first segment is a portion of the content displayed on the interface, and the first content is generated by the model in response to the global input; generate second content based on the first additional input; present the second content as an additional output of the first segment at a position associated with the first segment; the global input triggers a new round of dialogue, and the first additional input does not affect the processing of the global input.

[0206] It should be noted that the specific operation process and technical effects implemented when the above computer instructions are executed by the processor 1030 can be referred to the relevant descriptions in the above method embodiments, and will not be repeated here.

[0207] Figure 10 This is a block diagram of an electronic device provided in an embodiment of this application. Figure 10 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0208] like Figure 10 As shown, an electronic device 1000 according to an embodiment of this application includes a display screen (not shown) and a processor 1001, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a memory 1008 into a random access memory (RAM) 1003. The processor 1001 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 1001 may also include onboard memory for caching purposes. The processor 1001 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of this application.

[0209] RAM 1003 stores various programs and data required for the operation of electronic device 1000. Processor 1001, ROM 1002, and RAM 1003 are interconnected via bus 1004. Processor 1001 executes various operations of the method flow according to embodiments of this application by executing programs in ROM 1002 and / or RAM 1003. It should be noted that the programs may also be stored in one or more memories other than ROM 1002 and RAM 1003. Processor 1001 may also execute various operations of the method flow according to embodiments of this application by executing programs stored in said one or more memories.

[0210] According to embodiments of this application, the electronic device 1000 may further include an input / output (I / O) interface 1005, which is also connected to a bus 1004. The system 1000 may also include one or more of the following components connected to the input / output (I / O) interface 1005: an input device 1006 including a keyboard, mouse, etc.; an output device 1007 including a cathode ray tube (CRT), liquid crystal display (LCD), display screen, etc., and a speaker, etc.; a memory 1008 including a hard disk, etc.; and a communication section 1009 including a network interface card such as a LAN card, modem, etc. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the input / output (I / O) interface 1005 as needed. A removable medium 1011, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 1010 as needed so that computer programs read from it can be installed into the memory 1008 as needed.

[0211] According to embodiments of this application, the method flow according to embodiments of this application can be implemented as a computer software program. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable storage medium, the computer program containing program code for performing the methods shown in the flowchart. In such embodiments, the computer program can be downloaded and installed from a network via communication section 1009, and / or installed from removable medium 1011. When the computer program is executed by processor 1001, it performs the functions defined in the system of embodiments of this application. According to embodiments of this application, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0212] This application also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of this application.

[0213] According to embodiments of this application, the computer-readable storage medium can be a non-volatile computer-readable storage medium. Examples include, but are not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0214] For example, according to embodiments of this application, a computer-readable storage medium may include the ROM 1002 and / or RAM 1003 described above and / or one or more memories other than ROM 1002 and RAM 1003.

[0215] Embodiments of this application also include a computer program product comprising a computer program containing program code for performing the methods provided in the embodiments of this application. When the computer program product is run on an electronic device, the program code is used to enable the electronic device to implement the methods provided in the embodiments of this application.

[0216] When the computer program is executed by the processor 1001, it performs the functions defined in the system / apparatus of this application embodiment. According to the embodiments of this application, the systems, apparatuses, modules, units, etc. described above can be implemented by computer program modules.

[0217] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and may be downloaded and installed via the communication section 1009, and / or installed from a removable medium 1011. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0218] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions. Those skilled in the art will understand that the features recited in the various embodiments and / or claims of this application can be combined and / or combined in various ways, even if such combinations or combinations are not expressly stated in this application. In particular, the various embodiments and / or features described in the claims of this application may be combined and / or combined in various ways without departing from the spirit and teachings of this application. All such combinations and / or combinations fall within the scope of this application.

[0219] The embodiments of this application have been described above. However, these embodiments are merely illustrative and not intended to limit the scope of this application. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. The scope of this application is defined by the appended claims and their equivalents. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of this application, and all such substitutions and modifications should fall within the scope of this application.

[0220] The embodiments of this application have been described above. However, these embodiments are merely illustrative and not intended to limit the scope of this application. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. The scope of this application is defined by the appended claims and their equivalents. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of this application, and all such substitutions and modifications should fall within the scope of this application.

Claims

1. An interaction method, comprising: During the output and display of the first content on the interface, the first additional input for the first segment is obtained; The first segment is a portion of the content displayed on the interface, and this content is generated by the model based on the global input. Generate the second content based on the first appended input; The second content is presented as an additional output of the first segment at the location associated with the first segment; The global input triggers a new round of dialogue, and the first appended input does not affect the processing of the global input.

2. The method of claim 1, wherein presenting the second content as additional output to the first segment at a location associated with the first segment includes at least one of the following: The first segment is displayed in an embedded card format below it. The first additional input, the first fragment, and the second content are displayed in the side area of ​​the interface; Displayed as annotations at the corresponding positions in the first segment; The second content is displayed in either an expanded or collapsed state at the location associated with the first segment.

3. The method according to claim 1, wherein generating the second content based on the first appended input comprises: Determine the corresponding first anchor point based on the first segment; Context information in the first content is obtained based on the first anchor point; The context information includes the first segment and content segments located within a predetermined range before and after the position corresponding to the first anchor point; The second content is generated based on the first appended input and the context information.

4. The method of claim 3, wherein presenting at the location associated with the first segment comprises at least one of the following: Based on the first anchor point, the display position of the first segment in the first content is determined, and the second content is presented in the adjacent area corresponding to the display position. When the first content changes, the first segment is repositioned based on the first anchor point to maintain the association between the second content and the first segment in display. When the content of the first segment changes, the second content is regenerated based on the first anchor point, and the content display at the associated location is updated.

5. The method according to claim 4, wherein repositioning the first segment based on the first anchor point comprises: The first segment is initially located based on the position information in the first anchor point; The initial positioning result is verified based on the text feature information in the first anchor point; In the event of a discrepancy, a matching based on text similarity is performed within the first content; If a content segment corresponding to the first segment exists, a new position is determined based on the corresponding content segment, and the position information in the first anchor point is updated.

6. The method according to claim 4, wherein regenerating the second content based on the first anchor point and updating the content display at the associated location includes: The content changes of the first segment are detected based on the text feature information in the first anchor point; If the content of the first segment changes, the context information is reconstructed based on the changed first segment and the first appended input. The updated second content is generated based on the first appended input and the updated context information; Replace the text feature information in the first anchor point with the updated information of the second content, or update the content display at the associated position of the first segment.

7. The method according to claim 3, further comprising: The context information is determined by at least one of the following: a summary of the first content and similar content fragments retrieved based on the text feature information of the first fragment.

8. The method according to any one of claims 1-7, further comprising: Obtain a second appended input for the second segment of the second content; Generate the third content based on the second appended input; The third content is used as an additional output of the second segment, forming a nested display structure with the second segment in a related hierarchy.

9. The method according to claim 8, wherein generating the third content based on the second appended input comprises: Determine the corresponding second anchor point based on the second segment; Contextual information in the second content is obtained based on the second anchor point; Determine the corresponding first anchor point based on the first segment; Context information in the first content is obtained based on the first anchor point; The third content is generated based on the additional input and context information.

10. An electronic device, comprising: A display screen, used to show the interface; Memory, used to store computer instructions and models; The computer instructions are loaded by at least one processor and are used for: During the output display process on the interface based on the first content, a first additional input for the first segment is obtained; the first segment is a part of the content displayed on the interface, and the first content is generated by the model for the global input. Generate the second content based on the first appended input; The second content is presented as an additional output of the first segment at the location associated with the first segment; The global input triggers a new round of dialogue, and the first appended input does not affect the processing of the global input.

Images