Model output result display method and device and data processing system

Abstract

The invention discloses a model output result display method and device and a data processing system, and relates to the technical field of large models. The method comprises the steps that a layout adjusting instruction is responded, the layout of a display window of a front-end page is adjusted, the layout adjusting instruction is triggered by an updating event or triggered by user operation, and the updating event means that an output result of a large language model is updated; and displaying an output result of the large language model in the display window after layout adjustment. According to the method and the device, the technical problem that the model output result is difficult to meet the user demand because the user is difficult to obtain enough information to optimize the instruction data of the input model when the output result of the model is presented in a manner of expanding and retracting the content through a front-end page in the related technology is solved.

Description

Technical Field

[0001] The present application relates to the field of large model technology, and in particular to a method and device for displaying model output results, and a data processing system. Background Art

[0002] Interaction with the big model and the intelligent agent is a process of iterative optimization. If the user wants to optimize the next round of input command information, more information is needed, such as the steps of problem decomposition, the difference comparison of the output results before and after, etc. The application of big models in related technologies is mainly to present the thinking process of the big model in a straightforward manner through the expansion and collapse of cards in the form of dialogue.

[0003] It should be noted that the thinking process involved in large models and intelligent agents is complex and dynamic, covering multi-level data analysis and decision-making steps. The information displayed in the display method of related technologies is very limited. Users do not have sufficient information support to carry out the next round of instruction optimization, which affects the output results of the model.

[0004] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0005] The embodiments of the present application provide a method, device and data processing system for displaying model output results, so as to at least solve the technical problem in the related art that the output results of the model are presented in a manner of expanding and collapsing the content of the front-end page, and the user has difficulty in obtaining sufficient information to optimize the instruction data input into the model, resulting in the model output results being difficult to meet the user's needs.

[0006] According to one aspect of an embodiment of the present application, a method for displaying a model output result is provided, comprising: responding to a layout adjustment instruction, adjusting the layout of a display window of a front-end page, wherein the layout adjustment instruction is triggered by an update event or by a user operation, and the update event refers to an update of an output result of a large language model; and displaying the output result of the large language model in the display window after the layout is adjusted.

[0007] Optionally, in response to the layout adjustment instruction, adjusting the layout of the display window of the front-end page includes: in a case where the layout adjustment instruction is triggered by an update event, responding to the layout adjustment instruction, determining the total number of output results of the large language model, and adjusting at least one of the number, size and position of the display windows according to the total number of output results; in a case where the layout adjustment instruction is triggered by a user operation, responding to the layout adjustment instruction, adjusting at least one of the number, size and position of the display windows according to the type of user operation, wherein the type includes at least one of the following: dragging, scaling and deleting.

[0008] Optionally, if the output result of the large language model is updated, indicating that the large language model generates new output results, when the layout adjustment instruction is triggered by an update event, displaying the output result of the large language model in the display window after the layout is adjusted includes: obtaining the new output result of the large language model indicated by the update event; determining a first display window for displaying the new output result and a second display window for displaying historical output results among multiple display windows after the layout is adjusted; registering new components in the first display window according to the new output results, and rendering the new components; adjusting the size and position of the components in the second display window according to the size of the second display window, and re-rendering the adjusted components in the second display window according to the historical output results.

[0009] Optionally, registering a new component in the first display window according to the new output result includes: determining component definition information of the new component, and updating the component protocol of the front-end page according to the component definition information, wherein the component definition information includes at least a data source, configuration information, and linkage logic between components, the data source is the source of the new output result, and the configuration information is used to indicate how the component displays the new output result; determining layout information of the new component, and updating the layout protocol of the front-end page according to the layout information, wherein the layout information includes at least the size and position of the new component; determining communication information of the new component, and updating the communication protocol of the front-end page according to the communication information, wherein the communication information includes at least a communication method between components and a communication method between the component and the data source; registering the new component in the first display window according to the updated component protocol, updated layout protocol, and updated communication protocol of the front-end page.

[0010] Optionally, obtaining the newly added output results of the large language model indicated by the update event includes: receiving the output results generated by the server during the target phase of executing the text generation task, wherein the server is the server where the large language model is located; converting the output results into a format that supports rendering to obtain the newly added output results.

[0011] Optionally, when the layout adjustment instruction is triggered by a user operation, displaying the output results of the large language model in the display window after the layout is adjusted includes: adjusting the size and position of the components in the display window after the layout is adjusted according to the size and position of the display window after the layout is adjusted; obtaining the historical output results of the large language model associated with the component before the adjustment, and rendering the historical output results in the adjusted component.

[0012] Optionally, after displaying the output results of the large language model in the display window after the layout is adjusted, the method also includes: saving status data of different display windows after the layout is adjusted, wherein the status data at least includes the data, position and size displayed in the display window; after closing any display window, if an opening instruction of the display window is detected, restoring the display of the display window based on the status data of the display window; and when a reload instruction for any display window is detected, redisplaying the display window based on the status data of the display window.

[0013] According to another aspect of an embodiment of the present application, a display device for model output results is also provided, including: a first adjustment unit, used to respond to a layout adjustment instruction and adjust the layout of a display window of a front-end page, wherein the layout adjustment instruction is triggered by an update event or by a user operation, and the update event refers to an update of the output result of the large language model; a second adjustment unit, used to display the output result of the large language model in the display window after the layout is adjusted.

[0014] According to another aspect of an embodiment of the present application, a data processing system is also provided, including: a client, used to execute a method for displaying model output results, and display the output results of a large language model; a server, connected to the client, and used to send the output results of the large language model to the client.

[0015] According to another aspect of the embodiments of the present application, an electronic device is also provided, including: a memory storing an executable program; a processor connected to the memory via a bus and used to run the program, wherein the method of each embodiment of the present application is executed when the program is running.

[0016] According to another aspect of an embodiment of the present application, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the methods in each embodiment of the present application.

[0017] According to another aspect of an embodiment of the present application, a computer program product is also provided, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method in each embodiment of the present application is implemented.

[0018] In an embodiment of the present application, the layout adjustment instruction of the front-end page is triggered by an update event of the output result of the large language model, or is triggered by a user operation. The front-end page responds to the layout adjustment instruction, adjusts the layout of the display window, and displays the output result of the large language model in the display window after the layout is adjusted, so that different display windows display different output results of the large language model, and the user can intuitively adjust the window layout to adapt to specific viewing and analysis needs, thereby solving the technical problem in the related art that the output result of the model is presented by expanding and collapsing the content of the front-end page, and the user finds it difficult to obtain sufficient information to optimize the instruction data input to the model, resulting in the model output result being difficult to meet user needs.

[0019] It is easy to notice that the above general description and the following detailed description are only for the purpose of exemplifying and explaining the present application, and do not constitute a limitation of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0021] Figure 1 is a schematic diagram of a data processing system according to an embodiment of the present application;

[0022] Figure 2 is a flow chart of a method for displaying model output results according to an embodiment of the present application;

[0023] Figure 3 is a schematic diagram of an optional front-end page according to an embodiment of the present application;

[0024] Figure 4 is a schematic diagram of an optional front-end page according to an embodiment of the present application;

[0025] Figure 5 is a schematic diagram of an optional front-end page according to an embodiment of the present application;

[0026] Figure 6 is a schematic diagram of an optional method for displaying model output results according to an embodiment of the present application;

[0027] Figure 7 is a schematic diagram of a display device for model output results according to an embodiment of the present application;

[0028] Figure 8 is a structural block diagram of a computing device according to an embodiment of the present application;

[0029] Fig. 9It is a structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0030] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0031] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0032] The technical solution provided in this application is mainly implemented using large-scale model technology. The large model here refers to a deep learning model with large-scale model parameters, which can usually contain hundreds of millions, tens of billions, hundreds of billions, trillions or even more than 10 trillion model parameters. The large model can also be called a foundation model / foundation model. The large model is pre-trained with large-scale unlabeled corpus to produce a pre-trained model with more than 100 million parameters. This model can adapt to a wide range of downstream tasks, and the model has good generalization ability, such as large-scale language model (LLM), multi-modal pre-training model, etc.

[0033] It should be noted that when the large model is actually applied, the pre-trained model can be fine-tuned through a small number of samples so that the large model can be applied to different tasks. For example, the large model can be widely used in natural language processing (NLP), computer vision, speech processing and other fields, specifically in computer vision tasks such as visual question answering (VQA), image caption (IC), image generation, etc. It can also be widely used in natural language processing tasks such as text-based sentiment classification, text summary generation, machine translation, etc. Therefore, the main application scenarios of the large model include but are not limited to digital assistants, intelligent robots, search, online education, office software, e-commerce, intelligent design, etc.

[0034] According to an embodiment of the present application, a method for displaying model output results is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0035] Considering the huge number of model parameters of the large model and the limited computing resources of the mobile terminal, the above method provided in the embodiment of the present application can be applied to Figure 1 The application scenarios shown are not limited to these. Figure 1 It is a schematic diagram of a data processing system according to an embodiment of the present application. The large model is deployed in a server 10. The server 10 can be connected to one or more client devices 20 through a local area network connection, a wide area network connection, an Internet connection, or other types of data networks. The client device 20 here can include but is not limited to: a smart phone, a tablet computer, a laptop computer, a PDA, a personal computer, a smart home device, a car device, etc. The client device 20 can interact with the user through a graphical user front-end page to implement the call of the large model, thereby implementing the method provided in the embodiment of the present application.

[0036] In an embodiment of the present application, a system composed of a client device and a server can perform the following steps: the client device executes: responding to a layout adjustment instruction, adjusting the layout of a display window of a front-end page, wherein the layout adjustment instruction is triggered by an update event or by a user operation, and the update event refers to an update of the output result of the large language model; and displaying the output result of the large language model in the display window after the layout is adjusted. The server executes: sending the output result of the large language model to the client.

[0037] It should be noted that, with the rapid development of high-performance computing units, in other application scenarios, the above method provided in the embodiment of the present application can also be applied to the model all-in-one machine. In an optional embodiment, a plurality of models are built into the model all-in-one machine, and the user can choose to adjust a model as needed to obtain the user's own model, so that the high-performance computing unit built into the model all-in-one machine can directly call the adjusted model to execute the above method provided in the embodiment of the present application. In another optional embodiment, a trained model is built into the large model all-in-one machine, so that the high-performance computing unit built into the model all-in-one machine can directly call the model to execute the above method provided in the embodiment of the present application.

[0038] Furthermore, when users need to train their own models, they can also upload their own data sets through the client, which are sent by the client to the server, so that the server can adjust the pre-trained model with the data set to obtain the user's own model and then deploy it to the production environment. In order to facilitate users' needs for model adjustment, the server can provide complete adjustment tools, development frameworks and processes, and can support multiple adjustment strategies, so that the adjusted model can better adapt to applications in different fields and achieve a high degree of customization.

[0039] Under the above operating environment, this application provides Figure 2 The display method of the model output results is shown. Figure 2 FIG. 1 is a flow chart of a method for displaying model output results according to an embodiment of the present application. Figure 2 As shown, the method may include the following steps:

[0040] Step S201, responding to a layout adjustment instruction, adjusting the layout of the display window of the front-end page, wherein the layout adjustment instruction is triggered by an update event or by a user operation, and the update event refers to an update of the output result of the large language model.

[0041] Among them, the layout adjustment instruction is a signal that triggers the change of the front-end page layout. It can be automatically triggered by an update event indicating that the output result of the large language model is updated, or it can be triggered by the user's operation, ensuring that the front-end page can respond immediately to changes in the output results of the large language model or changes in user operations, and adjust the layout of the display window of the front-end page.

[0042] Exemplarily, triggering a layout adjustment instruction by an update event of the output result of a large language model means that when a large language model performs a text generation task, it may involve multiple stages. At the end of each stage, the large language model will generate output results. In each stage, output results of different dimensions can also be generated. When the output result is updated, an update event will be generated. The event carries information about the newly added output result, such as the content of the newly added output result, the associated stage, etc., to trigger the layout adjustment instruction. For example, in a new round of interaction, the user inputs the constructed instruction information into the large language model. The large language model generates output results based on the instruction information currently input by the user, and generates an update event, which is captured and converted into a layout adjustment instruction through the event system of the front-end page. For another example, after the front-end page displays the instruction reply information generated by the large language model based on the instruction information, the large language model also generates thinking process information. At this time, an update event is generated, which is captured and converted into a layout adjustment instruction through the event system of the front-end page.

[0043] Exemplarily, triggering a layout adjustment instruction by a user operation means that the user can adjust the window layout directly by operating the front-end page to adapt to the viewing requirements of the model's output results. For example, the user may need to enlarge a display window to view detailed information, or reduce a display window to increase the visibility of other display windows. The user's operation (for example, dragging the edge of the display window, clicking an adjustment button) is captured by the event system of the front-end page and converted into a layout adjustment instruction.

[0044] After the front-end page executes the layout adjustment instruction, the display window on the front-end page will be adjusted according to the information in the layout adjustment instruction. This may include increasing or decreasing the number of display windows, changing the size of display windows, moving the position of display windows, etc., to ensure that each output result can be displayed in a suitable and visible display window. The layout adjustment process can be handled by the layout system of the front-end page, which will recalculate the position and size of each display window according to the layout protocol and layout adjustment instructions to adapt to the new layout requirements.

[0045] Figure 3 This is a schematic diagram of an optional front-end page according to an embodiment of the present application. Figure 1 ,like Figure 3As shown, the front-end page can be used to display the analysis results of the large language model on the document. The front-end page includes a PDF viewer window, an operation viewer window and an operation viewer copy window, wherein the PDF viewer window can display the content of the document, and the user can read the document through the display window to facilitate understanding of the output results of the large language model. The operation viewer window and the operation viewer copy window can display the output results of different analysis dimensions of the document, or the output results of different analysis stages. On the one hand, if the large language model generates new output results of new dimensions or new stages, the front-end page can adjust the number of operation viewer copy windows to display the new output results. On the other hand, the user can adjust the position and size of the existing display window of the front-end page, and can also close the display window. The front-end page can respond to the user's operation and adjust the existing display window.

[0046] Figure 4 This is a schematic diagram of an optional front-end page according to an embodiment of the present application. Figure 2 ,like Figure 4 As shown, the front-end page can be used to display the processing results of the large language model on the dialogue task. The front-end page includes a dialogue window, a task window and a log window. The dialogue window displays the user's instructions under different interactive tasks and the reply information of the large language model. The task window is used to display the input and output, content description and processing steps of each interactive task. The log window is used to display the log information of each interactive task. On the one hand, when the large language model generates new output results of a new interactive task, the front-end page can adjust the number of sub-windows in the task window to display the new output results. On the other hand, the user can adjust the position and size of the existing display window of the front-end page, and can also close the display window. The front-end page can respond to the user's operation and adjust the existing display window.

[0047] Figure 5 This is a schematic diagram of an optional front-end page according to an embodiment of the present application. Figure 3 ,like Figure 5As shown, the front-end page can be used to display the parsing results of the large language model on the chart. The front-end page includes a chart display window and a parsing window. The chart display window can display images, tables and related text descriptions. Users can read and analyze images and tables through the display window to facilitate understanding of the output results of the large language model. The parsing window may include multiple display pages that can be switched to display the output results of different analysis dimensions of the chart, for example, displaying the overview information, table content, formula content and picture content of the chart. On the one hand, when the large language model generates new output results of new dimensions, the front-end page can adjust the number of display pages that can be switched in the parsing window to display the new output results. On the other hand, the user can adjust the position and size of the existing display window of the front-end page, and can also close the display window. The front-end page can respond to the user's operation and adjust the existing display window.

[0048] Step S202: display the output result of the large language model in the display window after the layout is adjusted.

[0049] The display window includes components, and different components are designed to display different types of data in the output results, such as text, images, tables, charts, etc. For example, the text summary component can be used to display the text results generated by the model, and the chart component can be used to display the visualization results of data analysis. By matching components and data types, it is ensured that the model results can be presented to users in an appropriate way.

[0050] The changes in the layout of the display window include changes in the size and position of the display window and changes in the number of display windows. After the size and position of the display window are changed, the components in the display window are adjusted accordingly, such as size changes, position changes, etc., to ensure that the layout and size of the components are consistent with the adjusted window. When the number of display windows changes, new components are determined for the new windows to display the new output results of the large language model.

[0051] Furthermore, the components in the display window after the layout is adjusted are re-rendered according to the output results of the large language model. The rendering process is implemented by calling the rendering function of the component to create or update display elements such as charts, images, and texts to display the output results of the large language model in the component, providing users with an interactive dynamic front-end page to view and compare different results of the model.

[0052] In an embodiment of the present application, the layout adjustment instruction of the front-end page is triggered by an update event of the output result of the large language model, or is triggered by a user operation. The front-end page responds to the layout adjustment instruction, adjusts the layout of the display window, and displays the output result of the large language model in the display window after the layout is adjusted, so that different display windows display different output results of the large language model, and the user can intuitively adjust the window layout to adapt to specific viewing and analysis needs, thereby solving the technical problem in the related art that the output result of the model is presented by expanding and collapsing the content of the front-end page, and the user finds it difficult to obtain sufficient information to optimize the instruction data input to the model, resulting in the model output result being difficult to meet user needs.

[0053] In order to ensure the flexibility and user-friendliness of the front-end page when displaying different output results of the large language model, optionally, in the display method of the model output results of the embodiment of the present application, responding to the layout adjustment instruction, adjusting the layout of the display window of the front-end page includes: when the layout adjustment instruction is triggered by an update event, responding to the layout adjustment instruction, determining the total number of output results of the large language model, and adjusting at least one of the number, size and position of the display windows according to the total number of output results; when the layout adjustment instruction is triggered by a user operation, responding to the layout adjustment instruction, adjusting at least one of the number, size and position of the display windows according to the type of user operation, wherein the type includes at least one of the following: dragging, zooming and deleting.

[0054] Exemplarily, the window layout can be dynamically adjusted through layout adjustment instructions triggered by update events. Specifically, the event system of the front-end page continuously monitors the update events of the output results from the large language model or intelligent agent service (a system or program driven by artificial intelligence technology, with a large language model as its core). When the update event of the output result is detected, the layout system of the front-end page enters the response state, determines the total number of output results, and automatically adjusts the number of display windows on the front-end page according to the total number of output results, ensuring that each result has an independent display space. At the same time, the layout system also dynamically adjusts the size and position of each display window according to the type and size of the output result, making the window layout beautiful and practical.

[0055] Exemplarily, the window layout can also be dynamically adjusted through layout adjustment instructions triggered by user operations (including dragging the display window edge, scaling the display window size, and deleting the display window, etc.). When the user operation is detected by the event system of the front-end page, the layout system of the front-end page flexibly adjusts the window layout according to the type of user operation. For example, when the user drags the edge of the display window to adjust the size, the layout system calculates the new display window size and rearranges the positions of all display windows to maintain the coordination of the overall layout. When the user scales the display window, the layout system processes the scaling instruction to ensure that the content in the display window can adapt to the new size. When the user deletes a display window, the layout system removes the corresponding display window and automatically adjusts the positions of the remaining display windows to fill the gap.

[0056] In this embodiment, on the one hand, when the model output results are updated, the front-end page can respond immediately and dynamically adjust the number and size of display windows to ensure that each result is reasonably displayed, so that users can fully understand the thinking process and decision-making basis of the model to reasonably construct the next round of instruction information. On the other hand, the front-end page has the ability to respond to user operations, and users can intuitively adjust the window layout to adapt to specific viewing and analysis needs, making the window layout more personalized and improving the efficiency of human-computer collaboration and the quality of decision-making.

[0057] In order to make the comparison of the model thinking process and results intuitive and efficient, optionally, in the method for displaying the model output results of the embodiment of the present application, if the output result of the large language model is updated to indicate that the large language model generates a new output result, when the layout adjustment instruction is triggered by an update event, displaying the output result of the large language model in the display window after the layout is adjusted includes: obtaining the new output result of the large language model indicated by the update event; determining a first display window for displaying the new output result and a second display window for displaying the historical output result among the multiple display windows after the layout is adjusted; registering a new component in the first display window according to the new output result, and rendering the new component; adjusting the size and position of the component in the second display window according to the size of the second display window, and re-rendering the adjusted component in the second display window according to the historical output result.

[0058] Exemplarily, the event system of the front-end page continuously listens for update events of the output results from the large language model. When the large language model generates new output results, the event system captures the update event and obtains the new output results from the communication layer based on the indication information carried by the update event.

[0059] After obtaining the newly added output results, the layout system of the front-end page determines which display windows are used to display the newly added output results and which display windows continue to display the historical output results. Specifically, the layout system creates a first display window according to the defined layout protocol, determines that the newly added output results will be displayed in the newly created first display window, and the historical output results continue to be retained in the second display window.

[0060] When displaying the newly added output results, you can select a component that can effectively display the type of data according to the type and content of the newly added output results, and register it in the first display window. Then, the data distribution system of the front-end page sends the newly added output results to the component, and the component renders according to the received output results, and the rendering results display the newly added output results of the large language model.

[0061] When displaying historical output results, considering the increase in the number of display windows, the layout system recalculates the size and position of the second display window to ensure that it does not block other display windows while maintaining a good front-end page layout. For components in the second display window, the data distribution system sends historical output results to the components based on the data source, and the components render based on the received output results to ensure that the components can display historical output results more reasonably and that the size and position of the components are adapted to the adjusted second display window.

[0062] In this embodiment, when the update event of the output result of the large language model triggers a layout adjustment instruction, by creating a new first display window, dynamically registering and rendering the new components in the first display window, it is possible to quickly respond to the iterative update of the large language model and provide the user with the latest model results. At the same time, by re-layouting and re-rendering the historical components in the second display window, the reasonable display of the historical output results is ensured. That is, not only can the new output results be efficiently introduced and displayed, but also the existing front-end page layout can be intelligently adjusted to ensure the continuous accessibility of the historical output results. When processing the dynamic and iterative output results of the large language model, the user is allowed to see the different output results of the large language model on the same front-end page at the same time for comparison and analysis, making the complex model thinking process and result comparison intuitive and efficient.

[0063] Before rendering a newly added component in the first display window, it is necessary to register the newly added component first. Optionally, in the display method of the model output result of the embodiment of the present application, registering the newly added component in the first display window according to the newly added output result includes: determining component definition information of the newly added component, and updating the component protocol of the front-end page according to the component definition information, wherein the component definition information at least includes a data source, configuration information, and linkage logic between components, the data source is the source of the newly added output result, and the configuration information is used to indicate the way in which the component displays the newly added output result; determining layout information of the newly added component, and updating the layout protocol of the front-end page according to the layout information, wherein the layout information at least includes the size and position of the newly added component; determining communication information of the newly added component, and updating the communication protocol of the front-end page according to the communication information, wherein the communication information at least includes the communication method between components and the communication method between the component and the data source; registering the newly added component in the first display window according to the updated component protocol, updated layout protocol, and updated communication protocol of the front-end page.

[0064] Exemplarily, when a new output result is received, the new output result is first analyzed to determine the new component type and component definition information that need to display the new output result. The component definition information includes data source, configuration information, and linkage logic between components, where the data source indicates the source of the data that the new component will use to display, the configuration information describes how the component processes and displays data, and the linkage logic defines how the component responds to events and data updates of other components. The process of updating the component protocol includes modifying the existing component list in the component protocol, adding new component description information, including its data source, configuration information, and linkage logic with existing components, ensuring that the new component can be correctly identified and configured in the layout system, preparing for subsequent registration and rendering.

[0065] After determining the definition information of the newly added component, the layout system of the front-end page determines the layout information of the newly added component, including size and position, based on the newly added output results. The update of layout information includes adjusting the display window allocation scheme in the layout protocol, allocating appropriate space to the newly added component, and taking into account the coordination and aesthetics of the overall layout. For example, if the newly added output result contains a large amount of data, a larger display window needs to be allocated; if the newly added output result is in contrast or correlation with the historical output result, the layout system will ensure that the two components are visually adjacent to facilitate user comparative analysis. The update of the layout protocol ensures that the newly added component can be displayed in the first display window with an appropriate size and position, while maintaining the consistency of the overall front-end page layout.

[0066] In order to achieve linkage and data sharing between components, it is also necessary to determine the communication information between the new component and the historical component and data source, and update it in the communication protocol. The communication information includes the communication method between components and the communication method between components and data sources. The update of the communication protocol ensures that the new component can seamlessly interact with other components and data sources, respond to events and data updates, and achieve dynamic communication and information synchronization between multi-window containers.

[0067] After the component protocol, layout protocol and communication protocol are updated, the registration process is executed according to the updated component protocol, layout protocol and communication protocol to register the newly added component in the first display window. The registration process includes component initialization, data binding and event listener setting. After the newly added component is registered, the component can be rendered according to its configuration information and data source to display the newly added output results.

[0068] This embodiment achieves the purpose of accurately displaying the newly added output results through components in the first display window by dynamically updating the component protocol, layout protocol and communication protocol. It not only improves the display flexibility and response speed of the front-end page, but also optimizes the collaboration and data sharing between components, ensuring that users can instantly see the thinking process and the latest output results of the large language model, and can also easily compare and analyze historical output results.

[0069] In order to enable the user to know the thinking process and result changes of the model in real time during the operation of the model, optionally, in the display method of the model output results of the embodiment of the present application, obtaining the new output results of the large language model indicated by the update event includes: receiving the output results generated by the server at the target stage of executing the text generation task, wherein the server is the server where the large language model is located; converting the output results into a format that supports rendering to obtain new output results.

[0070] It should be noted that the output data of the large language model can be transmitted from the server where the model is located to the front-end page of the client through the data stream network (Stream Network). The transmission process can be real-time or near real-time, and can be achieved through real-time communication mechanisms such as SSE (Server-Sent Events) or WS (WebSocket, network communication protocol), ensuring that the output results of the large language model can be quickly delivered to the front-end page, providing a data basis for dynamically updating and displaying the output results of the large language model.

[0071] After the front end receives the output results sent by the server, the next step is to parse these output results. The parsing process involves converting the raw data generated by the model, such as text, images, charts, etc., into a structured or standardized data format so that the components can understand and display this data. Parsing the output results can include word segmentation analysis of text, size adaptation of images, sorting of table data, etc., to ensure that the data can adapt to the display requirements of each component. After the parsing is completed, the parsed results are converted into a format that supports rendering, which can be a data format such as JSON or XML that is easy for components to read and process. The conversion process ensures that the component can accurately and efficiently display the output results of the model without additional processing or parsing of the original data, which improves the efficiency and response speed of the display process.

[0072] This embodiment realizes the instant update and visual display of the output results of the large language model through real-time data transmission, data analysis and format conversion, ensuring that users can see the latest output results of the large language model. It can also parse and format the output results according to the display requirements of the components so that the output results can be presented, providing users with an intuitive and efficient model result comparison and analysis environment.

[0073] When the number of display windows decreases, or when the number of display windows remains unchanged, the components also need to be re-rendered. Optionally, in the method for displaying the model output results of the embodiment of the present application, when the layout adjustment instruction is triggered by a user operation, displaying the output results of the large language model in the display window after the layout is adjusted includes: adjusting the size and position of the component in the display window after the layout is adjusted according to the size and position of the display window after the layout is adjusted; obtaining the historical output results of the large language model associated with the component before the adjustment, and rendering the historical output results in the adjusted component.

[0074] Exemplarily, the event of reducing the number of display windows can be triggered by a user operation, for example, closing a display window. When the event system of the front-end page detects the event of closing the display window, the layout system of the front-end page recalculates the size and position of the remaining display windows according to the number of display windows. First, determine whether it is necessary to adjust the division ratio or direction between the display windows after the display windows are reduced to optimize the display space of the remaining display windows. When a new layout scheme is determined, the layout system notifies all components to update their position and size information to ensure that even if the number of display windows is reduced, the components can still clearly and fully display the output results of the model in their respective display windows.

[0075] For example, when the number of display windows does not change, user operations (such as dragging to adjust the size of the display window) may also cause the size and position of the display window to change. The layout system will also update the layout protocol and adjust the size and position of the display window to adapt to new content display requirements or user preferences. And according to the adjusted size and position information of the display window, the size and position of the component are adaptively adjusted. For example, the chart component may need to be redrawn to adapt to the new window size, and the text summary component may need to be reformatted to ensure the readability of the text.

[0076] After the size and position of the component are adjusted, the component is re-rendered according to the output results associated with the component. Specifically, data is obtained from the data source and then rendered according to the component's configuration information and data processing logic. The re-rendering process includes updating the visual elements such as charts, lists, images, etc. in the component, and ensuring that the layout and size inside the component match the display window size after the layout is adjusted.

[0077] In the case where the number of display windows is reduced or the size of the display windows is adjusted, this embodiment not only adapts to the dynamic changes in the number of model results, but also optimizes the display of components and improves the readability of information and the efficiency of comparative analysis by dynamically adjusting the position and size of components and re-rendering components.

[0078] In order to ensure the persistence of data display and the continuity of user experience, optionally, in the method for displaying the model output results of an embodiment of the present application, after displaying the output results of the large language model in the display window after adjusting the layout, the method also includes: saving the status data of different display windows after adjusting the layout, wherein the status data includes at least the data, position and size displayed in the display window; after closing any display window, if an opening instruction of the display window is detected, restoring the display of the display window based on the status data of the display window; and when a reload instruction for any display window is detected, redisplaying the display window based on the status data of the display window.

[0079] It should be noted that when the user operates in a multi-window interactive environment, such as adjusting the display window size, moving the display window position, interacting with the components in the display window, etc., the state data of these display windows can be saved. The state data at least includes the display data (i.e., the model output results displayed in the current display window), the display window position, and the display window size information. The storage component is responsible for saving these data. The storage mechanism of the storage component can be a storage mechanism based on the secondary cache, which ensures that even when the user temporarily closes the display window or refreshes the display window, all important state information can be persisted, providing a basis for subsequent display recovery operations.

[0080] For example, after a user closes a display window, if an opening instruction of the display window is detected again later, the rendering container executes to read the state data from the storage component, reloads and renders the component according to the component protocol and data source, and restores the original display state of the display window (including the position and size of the display window, and the display content of the component in the display window). By restoring the state data, the user can quickly return to the previous operating environment without having to manually adjust the display window layout or reload the model results, which greatly improves the user's work efficiency and experience continuity.

[0081] For example, if the user issues a reload instruction for any display window, the rendering container first reads the state data of the display window from the storage component, then obtains the model output result from the data source, re-renders the components in the display window based on the model output result and the previously saved position and size information, and quickly redisplays the display window to ensure that the user sees the latest and consistently laid out display content. The reload instruction may be triggered by a direct user operation, for example, to refresh the content of the display window to obtain the latest display in order to ensure the real-time and consistency of data.

[0082] This embodiment saves and restores the status data of the display window, thereby ensuring that even after a brief display window closing or reloading operation, the user can quickly return to the previous operating environment without repeating the previous complex layout adjustment and data loading steps, thereby greatly improving work efficiency and operational fluency. When processing dynamic large model output results, it can help users maintain a coherent understanding of the model thinking process and reduce the risk of information interruption due to frequent operations.

[0083] In the above operating environment, this application also provides Figure 6 The display method of the model output results is shown. Figure 6 is a schematic diagram of an optional method for displaying model output results according to an embodiment of the present application, such as Figure 6 As shown, the method includes:

[0084] The user inputs the instruction data of the text generation task through the front-end page. The corresponding service of the front-end page converts the input into a format that the large language model can understand, and then sends it to the LLM (Large Language Model) or Agent service. The large language model or agent service starts to execute the text generation task, and the execution process can involve multiple stages of thinking and result generation.

[0085] The output of the large language model or agent service is encapsulated and transmitted to the front-end page in the form of data streams through the Stream Network as defined in the communication protocol. The transmission process can be real-time or near real-time, and can be achieved through real-time communication mechanisms such as SSE (Server-Sent Events) or WS (WebSocket, network communication protocol). The output of the large language model can be in the form of dialogue mode, report mode, summary mode, etc.

[0086] On the front-end page, the layout protocol in the registered skeleton is used to define how to dynamically create a multi-window container. The component protocol is used to define the component data source, configuration information, and how components are linked (actions, events). The communication protocol defines which data needs to be shared globally and how to request them.

[0087] After receiving the model results, the front-end page performs content parsing. The parsing process includes converting the original data generated by the model, such as text, images, charts, etc., into a structured or standardized data format so that the components can understand and display the data. After the parsing is completed, the parsing results are converted into a format that supports rendering, which can be a data format such as JSON or XML that is easy for components to read and process. Then, the data distribution system of the rendering container layer distributes the data to each component according to the data source defined in the component protocol. The layout system of the rendering container layer ensures that the component is initialized in the correct window, and dynamically adjusts the window size to adapt to different component display requirements. The component presents the thinking process and stage results of the model based on the received data. The runtime of the rendering container layer provides an environment for the operation of the component.

[0088] In multiple display windows, users can annotate and pick model results. The event system captures these operations and links with components in other windows according to the communication protocol in the registration skeleton, such as updating related data, triggering new requests, or adjusting display content, ensuring that user operations are immediately reflected in all related windows, providing a unified and coherent interactive experience.

[0089] Users can switch freely between multiple windows, compare the results of different stages, and conduct in-depth analysis. Each window can show different perspectives or details. By comparing multiple views, users can better understand the decision-making process and result changes of the model, and generate new instruction data, which is passed to the large language model or intelligent agent service again, triggering a new round of thinking and result generation. Through this iterative cycle, the large language model or intelligent agent service can gradually optimize its results to be closer to user needs.

[0090] In order to ensure the stability of the user's front-end page, the storage component will save the current layout state and component data. In this way, even if the user returns to view the results at a different time point, they can start from the last layout state without resetting the window or component, which improves the user experience and efficiency.

[0091] This implementation responds to user-constructed instruction information through a large language model or intelligent agent service to generate model results, and then uses multiple windows and components to dynamically display these model results, while supporting user interaction and feedback on model results. It not only demonstrates the complex thinking process of the large model, but also provides users with a basis for in-depth understanding and guidance of the model, and ultimately improves the accuracy and availability of model results through iterative optimization.

[0092] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0093] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0094] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, disk, CD), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0095] According to an embodiment of the present application, a data display mask for implementing the above-mentioned method for displaying the model output result is also provided. Figure 7 is a schematic diagram of a display device for the output result of the model according to an embodiment of the present application, such as Figure 7 As shown, the device comprises:

[0096] The first adjustment unit 701 is used to respond to a layout adjustment instruction and adjust the layout of the display window of the front-end page, wherein the layout adjustment instruction is triggered by an update event or a user operation, and the update event refers to an update of the output result of the large language model;

[0097] The second adjustment unit 702 is used to display the output result of the large language model in the display window after the layout is adjusted.

[0098] Optionally, in the display device of the model output results of the embodiment of the present application, the first adjustment unit 701 includes: a first adjustment module, which is used to respond to the layout adjustment instruction when the layout adjustment instruction is triggered by an update event, determine the total number of output results of the large language model, and adjust at least one of the number, size and position of the display windows according to the total number of output results; and a second adjustment module, which is used to respond to the layout adjustment instruction when the layout adjustment instruction is triggered by a user operation, and adjust at least one of the number, size and position of the display windows according to the type of user operation, wherein the type includes at least one of the following: dragging, scaling and deleting.

[0099] Optionally, in the display device of the model output result of the embodiment of the present application, if the output result of the large language model is updated to indicate that the large language model generates a new output result, the second adjustment unit 702 includes: an acquisition module, used to acquire the new output result of the large language model indicated by the update event; a determination module, used to determine, among the multiple display windows after the layout is adjusted, a first display window for displaying the new output result, and a second display window for displaying the historical output result; a registration module, used to register a new component in the first display window according to the new output result, and render the new component; a third adjustment module, used to adjust the size and position of the component in the second display window according to the size of the second display window, and re-render the adjusted component in the second display window according to the historical output result.

[0100] Optionally, in the display device of the model output results of the embodiment of the present application, the registration module includes: a first determination submodule, used to determine the component definition information of the newly added component, and update the component protocol of the front-end page according to the component definition information, wherein the component definition information includes at least a data source, configuration information and linkage logic between components, the data source is the source of the newly added output results, and the configuration information is used to indicate the way in which the component displays the newly added output results; a second determination submodule, used to determine the layout information of the newly added component, and update the layout protocol of the front-end page according to the layout information, wherein the layout information includes at least the size and position of the newly added component; a third determination submodule, used to determine the communication information of the newly added component, and update the communication protocol of the front-end page according to the communication information, wherein the communication information includes at least the communication method between components and the communication method between the component and the data source; a registration submodule, used to register the newly added component in the first display window according to the updated component protocol, updated layout protocol and updated communication protocol of the front-end page.

[0101] Optionally, in the display device of the model output results of the embodiment of the present application, the acquisition module includes: a receiving sub-module, used to receive the output results generated by the server at the target stage of executing the text generation task, wherein the server is the server where the large language model is located; a conversion sub-module, used to convert the output results into a format that supports rendering to obtain newly added output results.

[0102] Optionally, in the display device of the model output results of the embodiment of the present application, when the layout adjustment instruction is triggered by a user operation, the second adjustment unit 702 includes: a fourth adjustment module, used to adjust the size and position of the components in the display window after the layout is adjusted according to the size and position of the display window after the layout is adjusted; a rendering module, used to obtain the historical output results of the large language model associated with the component before the adjustment, and render the historical output results to the adjusted component.

[0103] Optionally, in the display device of the model output result of the embodiment of the present application, the device also includes: a maintaining and saving unit, which is used to save the state data of different display windows after the layout is adjusted after the output result of the large language model is displayed in the display window after the layout is adjusted, wherein the state data at least includes the data, position and size displayed in the display window; wherein the state data at least includes the data, position and size displayed in the display window; a first recovery unit, which is used to restore the display of the display window based on the state data of the display window if an opening instruction of the display window is detected after closing any display window; and a second recovery unit, which is used to redisplay the display window based on the state data of the display window when a reload instruction for any display window is detected.

[0104] It should be noted that the above-mentioned determination unit corresponds to the steps in the above-mentioned embodiment, and the examples and application scenarios implemented by the unit and the corresponding steps are the same, but are not limited to the contents disclosed in the above-mentioned embodiment. It should be noted that the above-mentioned modules or units can be hardware components or software components stored in a memory and processed by one or more processors, and the above-mentioned modules can also be run as part of the device in the server 10 provided in the above-mentioned embodiment.

[0105] It should be noted that the preferred implementation scheme involved in the above embodiments of the present application is the same as the scheme provided in the above embodiments, as well as the application scenario and implementation process, but is not limited to the scheme provided in the above embodiments.

[0106] An embodiment of the present application may provide a computing device. Figure 8 is a structural block diagram of a computing device according to an embodiment of the present application. Figure 8 As shown, the computing device 100 may include: one or more ( Figure 8 (only one is shown) processor 102, memory 104, storage controller, and peripheral interfaces.

[0107] The above-mentioned computing device can be understood as an integrated intelligent terminal, including but not limited to a server, a desktop computer, a PC (Personal Computer), a model all-in-one machine, etc., and the computing device can be pre-installed with the model described in the above-mentioned embodiments of the present application.

[0108] Specifically, the computing device can pre-set multiple types of models, including but not limited to models in the fields of natural language processing, visual processing, speech processing, code processing, multimodal task processing, etc., so as to provide a variety of model choices. In different product forms, the computing device can support one or more model usage methods, including but not limited to model training, model calling, model fine-tuning, model deployment, model reasoning and application, etc. In some product forms, the computing device also supports model management, including but not limited to multi-type model management (supporting the management of multiple types of models such as discriminant and generative models), model version control (supporting the control of different model versions), model evaluation (based on model evaluation tools, evaluating the performance and effect of the model), etc. In other product forms, the computing device can also create applications based on models, provide API calling capabilities, and can call models to the created applications through API interfaces, while providing application management tools to achieve management and monitoring of applications.

[0109] Furthermore, the computing device can also include data management (supporting the creation and management of model tuning data sets), training center (providing rich training resources to help users learn and master AI technology), and basic management and control capabilities (providing enterprise-level basic management and control capabilities to ensure the security and efficient operation of the system). Through the above functions, a comprehensive, integrated AI development, training, deployment and application device is provided.

[0110] Among them, the memory can be used to store software programs and modules, such as program instructions / modules corresponding to the methods and devices in the embodiments of the present application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, implementing the methods in the above embodiments. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the terminal A via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0111] The processor may call the executable program stored in the memory through the transmission device to execute the method described in any one of the above embodiments.

[0112] An embodiment of the present application may provide an electronic device. Fig. 9 is a structural block diagram of an electronic device according to an embodiment of the present application. Fig. 9 As shown, the electronic device may include: an input / output device 112 ; a memory 114 and a processor 116 , wherein the processor 116 is connected to the input / output device 112 and the memory 114 via a bus 118 .

[0113] Among them, the memory can be used to store software programs and modules, such as program instructions / modules corresponding to the methods and devices in the embodiments of the present application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, implementing the methods in the above embodiments. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the terminal A via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0114] The processor may call the executable program stored in the memory through the transmission device to execute the method described in any one of the above embodiments.

[0115] It will be understood by those skilled in the art that the structure shown in the figure is for illustration only, and the computing device may also be a terminal device such as a smart phone, a tablet computer, a PDA, a mobile Internet device (MID), a PAD, etc. The figure does not limit the structure of the above computing device. For example, the computing device 100 may also include more or fewer components (such as a network interface, a display device, etc.) than those shown in the figure, or have a configuration different from that shown in the figure.

[0116] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0117] The embodiment of the present application further provides a computer-readable storage medium. Optionally, in this embodiment, the computer-readable storage medium can be used to store the program code executed by the method provided in the above embodiment.

[0118] Optionally, in this embodiment, the above storage medium may be located in a computing device.

[0119] Optionally, in this embodiment, the computer-readable storage medium is configured to store an executable program, and when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the method described in any one of the above embodiments.

[0120] The embodiment of the present application further provides a computer program product. Optionally, in this embodiment, the computer program product may include a computer program, and the computer program implements the method provided in the embodiment when executed by a processor.

[0121] The embodiments of the present application also provide a computer program product. Optionally, the computer program product may include a non-volatile computer-readable storage medium, which may be used to store a computer program, and when the computer program is executed by a processor, the method provided in the embodiments is implemented.

[0122] The embodiment of the present application further provides a computer program. Optionally, in this embodiment, when the computer program is executed by a processor, the method provided in the above embodiment is implemented.

[0123] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0124] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0125] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0126] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0127] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk, etc., which can store program code.

[0128] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for displaying model output results, characterized in that: include: In response to a layout adjustment instruction, adjusting the layout of the display window of the front-end page, wherein the layout adjustment instruction is triggered by an update event or by a user operation, and the update event refers to an update of the output result of the large language model; The output result of the large language model is displayed in the display window after the layout is adjusted.

2. The method according to claim 1, characterized in that Responding to the layout adjustment instruction, adjusting the layout of the display window of the front-end page includes: In a case where the layout adjustment instruction is triggered by the update event, responding to the layout adjustment instruction, determining a total number of output results of the large language model, and adjusting at least one of the number, size, and position of the display window according to the total number of output results; In the case where the layout adjustment instruction is triggered by the user operation, in response to the layout adjustment instruction, at least one of the number, size and position of the display windows is adjusted according to the type of the user operation, wherein the type includes at least one of the following: dragging, scaling and deleting.

3. The method according to claim 1 or 2, characterized in that: If the output result of the large language model is updated, indicating that the large language model generates a new output result, and when the layout adjustment instruction is triggered by the update event, displaying the output result of the large language model in the display window after the layout is adjusted includes: Obtaining the newly added output result of the large language model indicated by the update event; Determining, from among the multiple display windows after the layout is adjusted, a first display window for displaying the newly added output result and a second display window for displaying the historical output result; registering a new component in the first display window according to the new output result, and rendering the new component; The size and position of the components in the second display window are adjusted according to the size of the second display window, and the adjusted components in the second display window are re-rendered according to the historical output results.

4. The method according to claim 3, characterized in that Registering a new component in the first display window according to the new output result includes: Determine the component definition information of the newly added component, and update the component protocol of the front-end page according to the component definition information, wherein the component definition information at least includes a data source, configuration information, and linkage logic between components, the data source is the source of the newly added output result, and the configuration information is used to indicate the way in which the component displays the newly added output result; Determine the layout information of the newly added component, and update the layout protocol of the front-end page according to the layout information, wherein the layout information at least includes the size and position of the newly added component; Determine the communication information of the newly added component, and update the communication protocol of the front-end page according to the communication information, wherein the communication information at least includes the communication mode between components and the communication mode between components and data sources; The newly added component is registered in the first display window according to the updated component protocol, the updated layout protocol and the updated communication protocol of the front-end page.

5. The method according to claim 3, characterized in that: Obtaining the newly added output result of the large language model indicated by the update event includes: Receiving an output result generated by a server in a target phase of executing a text generation task, wherein the server is a server where the large language model is located; The output result is converted into a format that supports rendering to obtain the newly added output result.

6. The method according to claim 1 or 2, characterized in that: In the case where the layout adjustment instruction is triggered by the user operation, displaying the output result of the large language model in the display window after the layout is adjusted includes: According to the size and position of the display window after the layout adjustment, adjusting the size and position of the components in the display window after the layout adjustment; The historical output results of the large language model associated with the component before adjustment are obtained, and the historical output results are rendered in the component after adjustment.

7. The method according to claim 1 or 2, characterized in that: After displaying the output result of the large language model in the display window after the layout is adjusted, the method further includes: Saving state data of different display windows after the layout adjustment, wherein the state data at least includes data, position and size displayed in the display window; After closing any display window, if an opening instruction of the display window is detected, restoring the display of the display window based on the state data of the display window; When a reload instruction for any display window is detected, the display window is redisplayed based on the state data of the display window.

8. A display device for model output results, characterized in that: include: A first adjustment unit, configured to adjust the layout of the display window of the front-end page in response to a layout adjustment instruction, wherein the layout adjustment instruction is triggered by an update event or a user operation, and the update event refers to an update of an output result of the large language model; The second adjustment unit is used to display the output result of the large language model in the display window after the layout is adjusted.

9. A data processing system, characterized in that: include: A client, configured to execute the method according to any one of claims 1 to 7, and display the output result of the large language model; The server is connected to the client and is used to send the output result of the large language model to the client.

10. An electronic device, characterized in that: include: A memory storing an executable program; A processor, connected to the memory via a bus, and configured to run the program, wherein the program executes the method described in any one of claims 1 to 7 when running.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored executable program, wherein when the executable program is executed, the device where the computer-readable storage medium is located is controlled to execute the method according to any one of claims 1 to 7.

12. A computer program product, characterized in that The invention comprises a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images