Data processing method and device and electronic equipment
By automatically triggering tasks and generating execution results using the target model in the large-scale model-assisted office monitoring operation, the problem of low efficiency caused by users frequently switching interfaces is solved, and the efficient execution of intelligent tasks is achieved.
Patent Information
- Application Number
- CN202511394839.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-26
- Publication Date
- 2025-12-23
AI Technical Summary
When using large-scale models to assist in office work, users need to frequently switch interfaces and input commands, resulting in low work efficiency and a poor user experience.
By monitoring operations in the target editing interface, tasks are automatically triggered and task information is obtained. The target model is used to generate execution results under the guidance of target prompts, thereby realizing the automated execution of intelligent tasks.
It improves the execution efficiency of intelligent tasks, reduces the frequency of interface switching, and enhances office efficiency and user experience.
Smart Images
Figure CN121189293A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of natural language processing technology, and in particular to a data processing method, apparatus and electronic device. Background Technology
[0002] Currently, when using large-scale models to assist in office work, it is necessary to input the required commands into the large-scale model's interactive interface to trigger the large-scale model to execute the office assistance tasks. Summary of the Invention
[0003] In view of the above, this application provides a data processing method, apparatus, and electronic device, as follows:
[0004] This application provides a data processing method, including:
[0005] In response to the triggering condition of the target task being hit by an operation on the target editing interface, the task information of the target task is obtained, the task information including the indication information of the target object and at least one initial prompt fragment;
[0006] Based on the indication information of the target object, obtain the content information corresponding to the target object on the target editing interface;
[0007] Based on at least one of the initial prompt fragments, target prompt information for the target task is determined, and the target prompt information is used to guide the target model to generate the execution result of the target task;
[0008] The content information of the target object and the target prompt information are input into the target model to obtain the execution result of the target task generated by the target model.
[0009] Preferably, it further includes:
[0010] In response to the task creation instruction of the target task, a task input box is displayed in the target editing interface;
[0011] Obtain the first task instruction input through the task input box;
[0012] Based on the first task instruction, the task information is determined;
[0013] Based on the first task instruction, determine the triggering conditions for the target task;
[0014] Based on the task information and the triggering conditions, the target task is created;
[0015] The target task is associated and bound to the target editing interface.
[0016] Preferably, creating the target task based on the task information and the triggering conditions includes:
[0017] Based on the task information and the triggering conditions, an initial task flow for the target task is created;
[0018] In response to receiving an adjustment instruction for the initial task flow, the initial task flow is adjusted according to the adjustment instruction to obtain the target task flow for the target task;
[0019] Based on the target task flow, the target task is created.
[0020] Preferably, determining the task information based on the first task instruction includes:
[0021] Parse the first task instruction to obtain the task type and the target object of the first task instruction;
[0022] Based on the task type of the first task instruction and the indication information of the target object, at least one first prompt fragment that matches the first task instruction is searched in the prompt fragment template library;
[0023] At least one of the first prompt fragments and corresponding matching information are displayed on the target editing interface, the matching information including matching degree and / or number of times it is used;
[0024] In response to a selection operation for at least one target first prompt fragment, each of the target first prompt fragments is determined as the initial prompt fragment.
[0025] Preferably, it further includes:
[0026] In response to an editing area creation command, the system obtains indication information of the target editing area for the target task, the indication information including location information.
[0027] Based on the indication information of the target editing area, the target editing area is created in the target editing interface; the task information also includes the indication information of the target editing area.
[0028] Preferably, after obtaining the execution result of the target task generated by the target model, the data processing method further includes:
[0029] The execution result of the target task is output to the target editing area.
[0030] Preferably, determining the target prompt information for the target task based on at least one of the initial prompt fragments includes:
[0031] At least one first prompt component is displayed on the target editing interface, and the first prompt component is associated with a second prompt fragment;
[0032] In response to an add operation for a target first prompt component in the at least one first prompt component, determine whether the add operation matches the target task;
[0033] If a match is found, the second prompt fragment associated with the target first prompt component is obtained;
[0034] Add the second prompt fragment associated with the first prompt component of the target to the target prompt information.
[0035] Preferably, determining the target prompt information for the target task based on at least one of the initial prompt fragments further includes:
[0036] At least one second prompt component is displayed in the target editing interface, and each second prompt component is associated with one of the initial prompt fragments;
[0037] In response to a removal operation of a target second prompt component in the at least one second prompt component, an initial prompt fragment associated with the target second prompt component is obtained;
[0038] Remove the initial prompt fragment associated with the target second prompt component from the target prompt information.
[0039] A data processing apparatus, comprising:
[0040] The task triggering unit is used to obtain the task information of the target task in response to the triggering condition of the operation on the target editing interface hitting the target task. The task information includes the indication information of the target object and at least one initial prompt fragment.
[0041] The content extraction unit is used to obtain the content information corresponding to the target object on the target editing interface based on the indication information of the target object;
[0042] The prompt information acquisition unit is used to determine the target prompt information of the target task based on at least one of the initial prompt fragments, wherein the target prompt information is used to guide the target model to generate the execution result of the target task;
[0043] The result generation unit is used to input the content information of the target object and the target prompt information into the target model to obtain the execution result of the target task generated by the target model.
[0044] An electronic device includes: a memory, a processor, and a computer program stored in the memory, wherein an intelligent agent is deployed in the processor, and the processor invokes the computer program to perform:
[0045] In response to the triggering condition of the target task being hit by an operation on the target editing interface, the task information of the target task is obtained, the task information including the indication information of the target object and at least one initial prompt fragment;
[0046] Based on the indication information of the target object, obtain the content information corresponding to the target object on the target editing interface;
[0047] Based on at least one of the initial prompt fragments, target prompt information for the target task is determined, and the target prompt information is used to guide the target model to generate the execution result of the target task;
[0048] The content information of the target object and the target prompt information are input into the target model to obtain the execution result of the target task generated by the target model.
[0049] As can be seen from the above technical solutions, the data processing method apparatus and electronic device disclosed in this application, in response to the triggering condition of an operation on a target editing interface hitting a target task, acquires task information of the target task, including indication information of the target object and at least one initial prompt fragment. Based on the indication information of the target object, the content information corresponding to the target object on the target editing interface is obtained. Based on at least one initial prompt fragment, the target prompt information of the target task is determined, and the target prompt information is used to guide the target model to generate the execution result of the target task. The content information of the target object and the target prompt information are input into the target model to obtain the execution result of the target task generated by the target model. In summary, this solution automatically triggers the target task corresponding to the target editing interface by monitoring the operation of the target editing interface. By inputting the content information of the target object and the target prompt information into the target model, the target model, guided by the target prompt information, performs the corresponding target task processing on the target object on the target editing interface to obtain the execution result, thereby improving the efficiency of intelligent task execution for the target editing interface. Attached Figure Description
[0050] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0051] Figure 1 A flowchart illustrating a data processing method provided in an embodiment of this application;
[0052] Figure 2a A flowchart illustrating a task binding method provided in an embodiment of this application;
[0053] Figure 2b An example of a target editing interface diagram is provided;
[0054] Figure 3 A flowchart illustrating another data processing method provided in an embodiment of this application;
[0055] Figure 4a A flowchart illustrating another data processing method provided in an embodiment of this application;
[0056] Figure 4b This provides an example of another editing interface diagram;
[0057] Figure 5 A flowchart illustrating a method for creating an editing area provided in an embodiment of this application;
[0058] Figure 6 A flowchart illustrating another data processing method provided in an embodiment of this application;
[0059] Figure 7 A flowchart illustrating another data processing method provided in an embodiment of this application;
[0060] Figure 8 A flowchart illustrating another data processing method provided in an embodiment of this application;
[0061] Figure 9 This is a schematic diagram of the structure of a data processing device provided in an embodiment of this application;
[0062] Figure 10 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation
[0063] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0064] This application provides a data processing method, apparatus, and electronic device, aiming to improve the efficiency of intelligent task execution. The data processing method of this application embodiment will be described in detail below with reference to the accompanying drawings.
[0065] Reference Figure 1 , Figure 1 This is a flowchart illustrating a data processing method provided in an embodiment of this application, as shown below. Figure 1As shown, this method includes steps 101 to 104, as follows:
[0066] Step 101: In response to the operation on the target editing interface hitting the trigger condition of the target task, obtain the task information of the target task.
[0067] In this embodiment, the task information includes indication information for the target object and at least one initial prompt fragment. The indication information is used to locate the position of the target object within the target editing interface. For example, the indication information for the target object is "third paragraph," indicating that the target object's position within the target editing interface is the third paragraph. The initial prompt fragment includes at least a prompt word indicating the task type. For example, initial prompt fragment A1 is "Summarize the core viewpoint of the third paragraph in one sentence," meaning the task type of the target task is a summary.
[0068] In this embodiment, the target editing interface is an interface in an editing state. For example, in a PPT office scenario, the target editing interface is the PPT editing page interface. Operations on the target editing interface include saving, editing (deleting and adding), and hovering the mouse over a specific position for more than a target time. The triggering condition for the target task includes at least an operation that triggers the target task. For example, the triggering condition for the target task includes receiving either a save operation or an Enter key operation on the target editing interface. Therefore, when the user clicks the save button on the target editing interface, triggering the save operation, it is determined that the save operation has met the triggering condition of the target task.
[0069] In this embodiment, the target editing interface can be associated with at least one task. The operation on the target editing interface is monitored. When the operation hits the trigger condition of at least one target task, the task information of each target task is obtained.
[0070] Step 102: Based on the indication information of the target object, obtain the content information corresponding to the target object on the target editing interface.
[0071] In this embodiment, based on the location of the target object indicated by the target indication information on the target editing interface, the original data corresponding to that location is extracted from the target editing interface. For example, if the indication information of the target object is "third paragraph", then the set of all paragraph elements on the content editing page of the target editing interface is obtained, the third paragraph is located by paragraph index, and data is extracted from the third paragraph to obtain the content information corresponding to the target object.
[0072] Step 103: Based on at least one initial cue fragment, determine the target cue information for the target task.
[0073] In this embodiment, the target prompt information is used to guide the target model to generate the execution result of the target task. The target prompt information is obtained by updating based on the initial prompt fragment, and the update includes adding or deleting.
[0074] For example, the task information for the target task includes two initial prompts: A1 "Summarize the core viewpoint of the third paragraph" and A2 "Keep the word count within 20 words". The updated target prompts include A1 "Summarize the core viewpoint of the third paragraph", A2 "Keep the word count within 20 words" and A3 "Conversational expression".
[0075] Step 104: Input the content information of the target object and the target prompt information into the target model to obtain the execution result of the target task generated by the target model.
[0076] In this embodiment, the target model is a pre-trained or fine-tuned intelligent model, which can be one of the following: a Large Language Model (LLM), a multimodal large model, a model capable of generating text, a model with more than 100 million adjustable parameters, a model that can output corresponding results based on input natural language instructions, a model that can convert input natural language instructions into feature sequences and infer output results based on feature sequences, or a model that can convert input information into feature sequences and infer subsequent features of the feature sequences. Optionally, for different target tasks, task-specific models can be configured according to the modality of the task type and content information. Specifically, the target model performs feature parsing and logical reasoning on the content information based on the prompt fragments in the target prompt information to generate structured or unstructured execution results.
[0077] For example, if the target task's prompts include A1, A2, and A3, and the content information is a text W, then W, A1, A2, and A3 are input into the target model. Guided by A1, A2, and A3, the target model extracts information from W, performs semantic compression, and performs style transfer to generate a natural language summary that satisfies multiple constraints (word limit and colloquial style) representing the core content of W.
[0078] As can be seen from the above technical solutions, the data processing method provided in this application, in response to the triggering condition of an operation on a target editing interface hitting a target task, obtains task information of the target task. The task information includes indication information of the target object and at least one initial prompt fragment. Based on the indication information of the target object, the content information corresponding to the target object on the target editing interface is obtained. Based on at least one initial prompt fragment, the target prompt information of the target task is determined. The target prompt information is used to guide the target model to generate the execution result of the target task. The content information of the target object and the target prompt information are input into the target model to obtain the execution result of the target task generated by the target model. In summary, this solution automatically triggers the target task corresponding to the target editing interface by monitoring the operation of the target editing interface. By inputting the content information of the target object and the target prompt information into the target model, the target model, guided by the target prompt information, performs the corresponding target task processing on the target object on the target editing interface to obtain the execution result, thereby improving the intelligent task execution efficiency of the target editing interface.
[0079] Furthermore, this application also includes the step of associating and binding the target editing interface with at least one target task, see [link to relevant documentation]. Figure 2a , Figure 2a This is a flowchart illustrating a task binding method provided in an embodiment of this application, as shown below. Figure 2a As shown, this method specifically includes steps 201 to 206, which are described in detail below.
[0080] Step 201: In response to the task creation command of the target task, display the task input box in the target editing interface.
[0081] In this embodiment, the task creation instruction is used to instruct the creation of a target task. For example, a task creation button is displayed in the target editing interface, and in response to clicking the task creation button, a task input box is displayed in the target editing interface. Optionally, the task input box is a text interaction area that pops up in the target editing interface.
[0082] Figure 2b An example of a target editing interface diagram is provided, such as... Figure 2b As shown, the editing interface of presentation 1 is equipped with a task creation button. After the user clicks the task creation button, a task input box pops up on the editing interface, where the user can enter task instructions.
[0083] Step 202: Obtain the first task instruction entered through the task input box.
[0084] like Figure 2bAs shown, after the user enters the text "summarize the core viewpoint of the third paragraph" in the task input box and triggers the submission operation, the agent obtains the first task instruction carrying the text through the interface.
[0085] Step 203: Determine the task information based on the first task instruction.
[0086] In this embodiment, firstly, the first task instruction is parsed, and operation keywords indicating the task type and object keywords indicating the target object are extracted from the first task instruction. Taking the input text "summarize the core viewpoint of the third paragraph" as an example, the operation keyword is extracted as "summarize" and the object keyword is "third paragraph". Then, the indication information of the target object is obtained based on the object keyword. Based on the operation keyword and the object keyword, the initial prompt fragment is determined. For example, the indication information of the target object includes an index for indicating the third paragraph and the initial prompt fragment "summarize the third paragraph".
[0087] Step 204: Based on the first task instruction, determine the triggering conditions of the target task.
[0088] In this embodiment, if the first task instruction includes conditional keywords for indicating the triggering operation, the operation keywords are directly extracted, and triggering conditions are generated based on the conditional keywords. For example, the first task instruction is "update the core summary of the entire page every time it is saved", where the conditional keywords include "every time" and "save". The triggering conditions obtained based on the conditional keywords include the save operation, that is, every save operation for the target editing interface hits the triggering conditions of the target task.
[0089] If the first task instruction does not include conditional keywords, then the corresponding triggering conditions can be matched based on the task type of the target task indicated by the first task instruction. For example, if the task type is "summary," the triggering conditions include the Enter key operation; if the task type is "refinement," the triggering condition is entering a period. It can be understood that different triggering operations can be automatically matched for different target tasks based on their task types.
[0090] Step 205: Create the target task based on the task information and triggering conditions.
[0091] In this embodiment, the target object, initial prompt information, and triggering conditions are combined into a complete task description. Based on the task description, a workflow for executing the target task is generated. Furthermore, the target object, initial prompt information, triggering conditions, and workflow are associated and bound.
[0092] Step 206: Associate and bind the target task with the target editing interface.
[0093] In this embodiment, the identifier of the target task and the identifier of the target editing interface are associated and stored together, so that, based on the triggering conditions of the target task associated with the target editing interface, the operation on the target editing interface is monitored, and it is determined whether the operation hits the triggering conditions of any target task. When it does, the execution logic of the target task is started.
[0094] As can be seen from the above technical solution, this solution improves the convenience and efficiency of operation by setting a task creation button on the target editing interface, which pops up an input box after clicking. It also associates and binds the target task with the target editing interface and monitors the operation on the target editing interface. When the operation hits the trigger condition, the target task is executed in a timely manner, improving the task execution efficiency. Furthermore, this solution supports both extracting conditional keywords from task instructions to generate trigger conditions and automatically matching trigger operations according to the task type, enabling different tasks to have flexible and diverse triggering methods and enhancing the adaptability of task execution.
[0095] Based on the above embodiments, see Figure 3 , Figure 3 This is a flowchart illustrating another data processing method provided in an embodiment of this application. Figure 3 Step 205 is shown, which is the specific implementation method for creating the target task based on task information and triggering conditions, such as... Figure 3 As shown, this method specifically includes steps 301 to 303, which are described in detail below.
[0096] Step 301: Based on the task information and triggering conditions, create the initial task flow for the target task.
[0097] In this embodiment, the execution steps of the target task are generated based on the task information and triggering conditions to obtain an initial task flow, wherein the task flow includes various execution steps arranged in the execution order.
[0098] Step 302: In response to receiving the adjustment instruction for the initial task flow, adjust the initial task flow according to the adjustment instruction to obtain the target task flow for the target task.
[0099] In this embodiment, the configuration file for configuring the initial task flow is displayed in the target editing interface or the editing interface of the process editor, and adjustment instructions for the initial task flow are received. The adjustment instructions can be triggered by drag and drop operations, including adjustment instructions for any execution step, such as swapping step 2 "call webpage 1" and step 3 "call webpage 2" in the initial task flow.
[0100] In this embodiment, the target task flow of the target task is an adjusted flow that includes at least one execution step arranged in the execution order.
[0101] Step 303: Create the target task based on the target task flow.
[0102] In this embodiment, the target object, initial prompt information, triggering conditions, and target workflow are associated and bound to execute each target task according to the target task flow of each target task.
[0103] Based on the above embodiments, see Figure 4a , Figure 4a This is a flowchart illustrating another data processing method provided in an embodiment of this application. Figure 4a Step 203 is shown, which is the specific implementation method for determining the task information based on the first task instruction, such as... Figure 4a As shown, this method specifically includes steps 401 to 404, which are described in detail below.
[0104] Step 401: Parse the first task instruction to obtain the task type and target object of the first task instruction.
[0105] In this embodiment, firstly, the first task instruction is parsed, and operation keywords indicating the task type are extracted from the first task instruction to determine the task type. Then, object keywords indicating the target object are extracted to determine the target object. Taking the input text "summarize the core viewpoint of the third paragraph" as an example, the extracted operation keyword is "summary," and the object keyword is "third paragraph." Therefore, the task type is determined to be a summary task, and the indicator information for the target object is the index identifier corresponding to the third paragraph text.
[0106] Step 402: Based on the task type and target object indication information of the first task instruction, search for at least one first prompt fragment that matches the first task instruction in the prompt fragment template library.
[0107] In this embodiment, the prompt fragment template library includes multiple historical prompt fragments and / or template prompt fragments, and any prompt fragment contains operation keywords and object keywords.
[0108] In this embodiment, for any historical prompt fragment or template prompt fragment, the matching degree with the first task instruction is determined based on the similarity between the operation keywords and object keywords in the prompt fragment and the task type and target object of the first task instruction. Historical prompt fragments and template prompt fragments with a matching degree greater than the matching degree threshold are selected to obtain at least one first prompt fragment that matches the first task instruction.
[0109] Step 403: Display at least one first prompt fragment and its corresponding matching information in the target editing interface.
[0110] In this embodiment, the matching information includes matching degree and / or usage count. The matching degree indicates the relevance between the first prompt fragment and the first task instruction, and the usage count indicates the number of times the first prompt fragment was selected by the user corresponding to the target editing interface.
[0111] In this embodiment, a candidate panel is displayed at the corresponding position of the task input box of the target task, and a first prompt fragment and matching information of the first prompt fragment are displayed on the candidate panel.
[0112] Figure 4b This example provides a schematic diagram of another editing interface, such as... Figure 4b As shown, a candidate panel is displayed below the task input box of the target task. The candidate panel displays three first prompt fragments that match the first task instruction "summarize the core viewpoint of the third paragraph". They are: B1 "summarize the main idea of the third paragraph", B2 "briefly describe the core content of the third paragraph" and B3 "summarize the key points of the third paragraph". After each first prompt fragment, a first label indicating the matching degree and a second label indicating the number of times it is used are displayed. For example, B1 has a matching degree of 88% and is used 5 times.
[0113] Step 404: In response to the selection operation of at least one target first prompt fragment, determine each target first prompt fragment as the initial prompt fragment.
[0114] In this embodiment, the selection operation can be a click operation on the target first prompt segment or a selection operation on the checkbox of the target first prompt segment. Figure 4b As shown, the system receives a click operation on the target first prompt segment B2 "Briefly describe the core content of the third paragraph", determines B2 "Briefly describe the core content of the third paragraph" as the target first prompt segment, and obtains the initial prompt segment.
[0115] It should be noted that a drop-down button can be displayed in the task input box. When a click operation is received on the drop-down button, steps 401 to 404 will be executed. If no click operation is received on the drop-down button, the first task instruction will be used as the initial prompt segment.
[0116] As can be seen from the above technical solution, this solution generates an initial task flow containing sequentially arranged execution steps based on task information and triggering conditions. It supports triggering adjustment commands through drag-and-drop operations, flexibly adjusting the order of each execution step in the task flow to meet diverse task requirements. By associating and binding the target object, initial prompt information, triggering conditions, and target workflow, it ensures that the target task can be executed strictly according to the adjusted target task flow, guaranteeing accurate and error-free task execution.
[0117] Based on the above embodiments, the task information also includes indication information of the target editing area, which is the text input area on the target editing interface. Figure 5 A method for creating an editing area is provided in the embodiments of this application, such as Figure 5 As shown, this method specifically includes steps 501 to 502, which are described in detail below.
[0118] Step 501: In response to the editing area creation command, obtain the indication information of the target editing area of the target task.
[0119] In this embodiment, the indication information of the target task includes location information, such as coordinate information, which is used to indicate the relative position of the target editing area on the target editing interface. The editing area creation instruction and the task creation instruction are triggered synchronously. For example, the editing area creation instruction is triggered in response to the selection operation of the task creation button.
[0120] In this embodiment, there are several methods for obtaining the indication information of the target editing area of the target task. For example, the indication information of the target editing area corresponding to the target area can be determined by the parameter configuration operation of the area configuration interface corresponding to the target task. Another example is to obtain the target task and the corresponding indication information of the target editing area that match the target editing interface from the configuration information of historical target editing interfaces, based on the type of the target editing interface. Yet another example is to receive a drag-and-drop selection operation in the target editing area and parse the indication information of the target editing area corresponding to the drag-and-drop selection operation. Specifically, in response to the click operation of the task creation button, both the editing area creation command and the task creation command are triggered simultaneously. The mouse becomes a draggable selection state, and the user's selection operation of a rectangular / polygonal area at any position on the target editing interface is received. The coordinate information corresponding to the selection operation is recorded to obtain the indication information of the target editing area.
[0121] Furthermore, based on the coordinate information, the corresponding editing area is displayed on the target editing interface, and a task instruction input box is displayed within the editing area.
[0122] Step 502: Based on the indication information of the target editing area, create the target editing area in the target editing interface.
[0123] In this embodiment, the indicator information of the target editing area is added to the task information of the target task, and the target editing area is displayed on the target editing interface to realize the creation of the target editing area.
[0124] Based on the above embodiments, this application embodiment further includes: outputting the execution result of the target task to the target editing area. That is, after obtaining the execution result of the target task, the execution result is output to the corresponding position of the target editing interface according to the instruction information of the target editing area of the target task.
[0125] As can be seen from the above technical solution, this solution quickly searches for matching prompt fragments in the prompt fragment template library based on the task type and target object indication information. It then uses a similarity algorithm to filter out prompt fragments with high matching degrees. The first matching prompt fragment, along with matching information such as matching degree and usage frequency, is displayed intuitively in the target editing interface as a candidate panel. This displays the degree of association between each prompt fragment and the task, as well as its usage frequency, providing a basis for selection. Users can flexibly determine the target's first prompt fragment as the initial prompt fragment through clicks or checkboxes, improving the matching degree between the initial prompt fragment and the current target task, and enhancing the flexibility of initial prompt fragment configuration.
[0126] Based on the above embodiments, see Figure 6 , Figure 6 This is a flowchart illustrating another data processing method provided in an embodiment of this application. Figure 6 The following illustrates step 103, which involves determining the target prompt information for the target task based on at least one initial prompt fragment: Figure 6 As shown, this method specifically includes steps 601 to 603, which are described in detail below.
[0127] Step 601: Display at least one first prompt component in the target editing interface, and associate the first prompt component with a second prompt fragment.
[0128] In this embodiment, at least one first prompt component is displayed in the first area of the target editing interface, and the first prompt segments are different.
[0129] In one optional embodiment, based on the task information of the target task, at least one second prompt segment that is different from the first prompt segment is determined, and a first prompt component associated with the second prompt segment is displayed in the candidate panel corresponding to the target task.
[0130] Step 602: In response to an add operation for a target first prompt component in at least one first prompt component, determine whether the add operation matches the target task.
[0131] In this embodiment, the addition operation for the target first prompt component can be a selection operation or a drag operation for the target first prompt component. The addition operation matching the target task means that the target first prompt component is the first prompt component corresponding to the target task. For example, the end position of the drag operation is within the target editing area or task input box corresponding to the target task. For another example, for the first target task, when there are multiple candidate panels of target tasks in the target editing interface, it is determined whether the target first prompt component corresponding to the click operation is located within the candidate panel of the first target task. If so, a match is determined.
[0132] Step 603: If a match is found, obtain the second prompt fragment associated with the target first prompt component.
[0133] In this embodiment, the association and binding of the prompt component and the prompt fragment are determined by the interaction between the front-end and back-end information to determine the prompt fragment associated with the prompt component.
[0134] Step 604: Add the second prompt fragment associated with the first prompt component of the target to the target prompt information.
[0135] In this embodiment, after the second prompt segment associated with the target prompt component, namely the target second prompt segment, is added to the target prompt information, the target prompt information includes the initial prompt segment and the target second prompt segment.
[0136] Based on the above embodiments, see Figure 7 , Figure 7 This is a flowchart illustrating another data processing method provided in an embodiment of this application. Figure 7 This illustrates another specific implementation method for step 103, namely, determining the target cue information of the target task based on at least one initial cue fragment, such as... Figure 7 As shown, this method specifically includes steps 701 to 706, which are described in detail below.
[0137] Step 701: Display at least one first prompt component and at least one second prompt component in the target editing interface.
[0138] In this embodiment, a first prompt component is associated with a second prompt segment, and a second prompt component is associated with an initial prompt segment. That is, prompt segments from the task information are displayed simultaneously in the target editing interface.
[0139] Step 702: In response to an add operation for a target first prompt component in at least one first prompt component, determine whether the add operation matches the target task.
[0140] Step 703: If a match is found, obtain the second prompt fragment associated with the target first prompt component.
[0141] Step 704: Add the second prompt fragment associated with the first prompt component of the target to the target prompt information.
[0142] Step 705: In response to a removal operation of a target second prompt component in at least one second prompt component, obtain the initial prompt fragment associated with the target second prompt component.
[0143] In this embodiment, the removal operation includes selecting or dragging the removal button. For example, a removal button for the second prompt component is displayed on the target editing interface, and the selection of the removal button for the target second prompt component is received to obtain the initial prompt fragment associated with the target second prompt component. Alternatively, a drag operation on the target second prompt component is received; if the end position of the drag operation exceeds the candidate panel, the initial prompt fragment associated with the target second prompt component is obtained.
[0144] Step 706: Remove the initial prompt fragment associated with the target second prompt component from the target prompt information.
[0145] It should be noted that steps 701 to 706 can be performed during the process from when the target task is triggered to when the execution result is obtained. That is, the target prompt information input to the target model can be updated based on the selection operation of the first target prompt component and / or the removal operation of the second target prompt component. Therefore, updating the target prompt information interactively improves the adaptability of the target prompt information to the target task and enhances the accuracy of the target task's execution result.
[0146] In summary, the data processing method provided in this application can be specifically applied to intelligent office applications with large-model-assisted office functions. During intelligent office work, users need to switch to the large-model interactive interface to input requirements and send commands to complete the task. However, requirements in office scenarios are repetitive and frequent, thus necessitating high-frequency interface switching and command sending operations. For example, each page may require a summary of key points in one sentence, and each key point may require detailed explanations. Repeatedly switching to the large-model interactive interface and inputting requirements results in low work efficiency and a poor user experience.
[0147] Based on this, the data processing method provided in this application is applied to an intelligent agent configured in office software. The intelligent agent pre-configures multiple tasks for the editing interface and automatically triggers these tasks. This eliminates the need for frequent interface switching, enabling automatic triggering and execution of intelligent tasks, thus improving task execution efficiency and overall office efficiency. For example, in an office scenario based on a presentation document (PPT), see [link to relevant documentation]. Figure 8 , Figure 8 A data processing flowchart is provided for an embodiment of this application, such as Figure 8 As shown, data processing includes five stages, as follows:
[0148] 1. Define the editing area.
[0149] Specifically, in the editing interface of the office software, clicking the intelligent auxiliary floating button triggers an instruction to create an editing area. The mouse then becomes a draggable selection area, allowing the user to drag the mouse to select a rectangular or polygonal area anywhere on the screen, supporting the creation of multiple areas in parallel. The intelligent agent records the area information of each area, including its coordinate range and display level, and generates a visual editing area mapping table to provide spatial anchoring for subsequent task binding and content output.
[0150] 2. Define intelligent tasks.
[0151] Specifically, within the editing area, a task input box pops up to receive the user's natural language input request and obtain the first task instruction, such as "summarize the core viewpoints of the current PPT page" or "supplement case data to support the argument". The task input box supports multiple languages and quick keyword input.
[0152] The intelligent agent parses the first task instruction, extracts operation keywords indicating the task type and object keywords indicating the target object, and generates structured task metadata.
[0153] 3. Generate the candidate suggestion component.
[0154] Specifically, the intelligent agent accesses the user's historical task library and an industry-standard prompt template library. Combining this with task metadata, it uses semantic matching and scenario adaptation algorithms to filter out highly relevant prompt fragments to the first task instruction. A candidate panel pops up next to the editing area, displaying alternative prompt components corresponding to highly relevant prompt fragment options with different styles / emphases, and indicating usage frequency and matching degree tags. This allows the user to flexibly select or combine them, ultimately determining the user's selected highly relevant prompt fragment as the initial prompt fragment.
[0155] 4. Generate workflow.
[0156] In this embodiment, the intelligent agent automatically breaks down the task execution steps and configures trigger conditions based on task metadata and initial prompt fragments. A visual workflow configuration file is generated, which includes the execution order of the steps, the dependencies between different tasks, and output format rules. Through the process editor, the execution steps in the workflow can be adjusted to achieve personalized task execution logic configuration.
[0157] 5. Start interactive office.
[0158] When a user modifies, adds content, or adds a new page in a PowerPoint presentation, the corresponding target task is automatically triggered. The intelligent agent obtains the initial prompts and content information of the target object, then calls the corresponding target model to execute the operation instructed by the target task, rendering the result in real-time to the corresponding editing area. Based on this, parallel interaction between editing and intelligent tasks is achieved. Users can view and select auxiliary content on the same interface without switching windows, and it supports quick operations such as "one-click insertion of original document" and "marking for optimization."
[0159] During the target task execution phase, the user selects at least one prompt fragment component from the prompt fragment components displayed in the candidate panel (i.e., the first prompt fragment component) and drags it to the editing area corresponding to the target task. The agent automatically parses and updates the logic of the target task after updating the target prompt information, merges it with the target task flow, and updates the execution result. Based on this, the auxiliary capabilities of the editing area dynamically expand with the user's dragging behavior, continuously optimizing interaction efficiency and improving the efficiency and accuracy of the agent in calling the target model to execute the target task.
[0160] This application also provides a data processing apparatus, for reference... Figure 9 , Figure 9 This is a schematic diagram of the structure of a data processing device provided in an embodiment of this application, as shown below. Figure 9 As shown, this device includes:
[0161] The task triggering unit 901 is used to obtain the task information of the target task in response to the triggering condition of the operation on the target editing interface hitting the target task. The task information includes the indication information of the target object and at least one initial prompt fragment.
[0162] The content extraction unit 902 is used to obtain the content information corresponding to the target object on the target editing interface based on the indication information of the target object;
[0163] The prompt information acquisition unit 903 is used to determine the target prompt information of the target task based on at least one of the initial prompt fragments, and the target prompt information is used to guide the target model to generate the execution result of the target task;
[0164] The result generation unit 904 is used to input the content information of the target object and the target prompt information into the target model to obtain the execution result of the target task generated by the target model.
[0165] Preferably, the data processing apparatus further includes a task creation unit, used for:
[0166] In response to the task creation instruction of the target task, a task input box is displayed in the target editing interface;
[0167] Obtain the first task instruction input through the task input box;
[0168] Based on the first task instruction, the task information is determined;
[0169] Based on the first task instruction, determine the triggering conditions for the target task;
[0170] Based on the task information and the triggering conditions, the target task is created;
[0171] The target task is associated and bound to the target editing interface.
[0172] Preferably, when the task creation unit creates the target task based on the task information and the triggering conditions, it is specifically used for:
[0173] Based on the task information and the triggering conditions, an initial task flow for the target task is created;
[0174] In response to receiving an adjustment instruction for the initial task flow, the initial task flow is adjusted according to the adjustment instruction to obtain the target task flow for the target task;
[0175] Based on the target task flow, the target task is created.
[0176] Preferably, the task creation unit, when determining the task information based on the first task instruction, is specifically used for:
[0177] Parse the first task instruction to obtain the task type and the target object of the first task instruction;
[0178] Based on the task type of the first task instruction and the indication information of the target object, at least one first prompt fragment that matches the first task instruction is searched in the prompt fragment template library;
[0179] At least one of the first prompt fragments and corresponding matching information are displayed on the target editing interface, the matching information including matching degree and / or number of times it is used;
[0180] In response to a selection operation for at least one target first prompt fragment, each of the target first prompt fragments is determined as the initial prompt fragment.
[0181] Preferably, the data processing apparatus further includes a region creation unit, used for:
[0182] In response to an editing area creation command, the system obtains indication information of the target editing area for the target task, the indication information including location information.
[0183] Based on the indication information of the target editing area, the target editing area is created in the target editing interface; the task information also includes the indication information of the target editing area.
[0184] Preferably, the data processing device further includes a result rendering unit, used to output the execution result of the target task to the target editing area after obtaining the execution result of the target task generated by the target model.
[0185] Preferably, the prompt information acquisition unit, when determining the target prompt information of the target task based on at least one of the initial prompt fragments, is specifically used for:
[0186] At least one first prompt component is displayed on the target editing interface, and the first prompt component is associated with a second prompt fragment;
[0187] In response to an add operation for a target first prompt component in the at least one first prompt component, determine whether the add operation matches the target task;
[0188] If a match is found, the second prompt fragment associated with the target first prompt component is obtained;
[0189] Add the second prompt fragment associated with the first prompt component of the target to the target prompt information.
[0190] Preferably, when the prompt information acquisition unit determines the target prompt information of the target task based on at least one of the initial prompt fragments, it is further specifically used for:
[0191] At least one second prompt component is displayed in the target editing interface, and each second prompt component is associated with one of the initial prompt fragments;
[0192] In response to a removal operation of a target second prompt component in the at least one second prompt component, an initial prompt fragment associated with the target second prompt component is obtained;
[0193] Remove the initial prompt fragment associated with the target second prompt component from the target prompt information.
[0194] This application also provides an electronic device, for reference. Figure 10 , Figure 10 This is a schematic diagram of the structure of an electronic device provided for an embodiment of this application, illustrating a structure suitable for implementing the electronic device in the embodiment of this application. For example... Figure 10 As shown, the electronic devices in the embodiments of this application may include, but are not limited to, fixed terminals such as mobile phones, laptops, PDAs (personal digital assistants), PADs (tablet computers), desktop computers, etc. Figure 10The 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.
[0195] like Figure 10 As shown, the electronic device may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 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 storage device 1008 into a random access memory (RAM) 1003. When the electronic device is powered on, the RAM 1003 also stores various programs and data required for the operation of the electronic device. The processing unit 1001, ROM 1002, and RAM 1003 are interconnected via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.
[0196] Typically, the following devices can be connected to the I / O interface 1005: input devices 1006 including, for example, a touchscreen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices 1007 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1008 including, for example, memory card, hard disk, etc.; and communication devices 1009. Communication device 1009 allows electronic devices to exchange data via wireless or wired communication with other devices. Although Figure 10 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown. More or fewer devices may be implemented or have alternatively.
[0197] This application also provides a computer program product including computer-readable instructions, which, when executed on an electronic device, cause the electronic device to implement the various steps of any data processing method provided in this application.
[0198] This application also provides a computer-readable storage medium that carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement the various steps of any of the data processing methods provided in this application.
[0199] It should also be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. In addition, in the device embodiment drawings provided in this application, the connection relationship between modules indicates that they have a communication connection, which can be implemented as one or more communication buses or signal lines.
[0200] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware, or it can be implemented by special-purpose hardware including application-specific integrated circuits, special-purpose CPUs, special-purpose memory, special-purpose components, etc. Generally, any function performed by a computer program can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be diverse, such as analog circuits, digital circuits, or special-purpose circuits. However, for this application, software program implementation is more often the preferred implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium, such as a computer floppy disk, USB flash drive, mobile hard disk, ROM, RAM, magnetic disk, or optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, training equipment, or network device, etc.) to execute the methods described in the various embodiments of this application.
[0201] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product.
[0202] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, training device, or data center to another website, computer, training device, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives (SSDs)).
[0203] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.
[0204] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0205] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.
[0206] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A data processing method, comprising: In response to the triggering condition of the target task being hit by an operation on the target editing interface, the task information of the target task is obtained, the task information including the indication information of the target object and at least one initial prompt fragment; Based on the indication information of the target object, obtain the content information corresponding to the target object on the target editing interface; Based on at least one of the initial prompt fragments, target prompt information for the target task is determined, and the target prompt information is used to guide the target model to generate the execution result of the target task; The content information of the target object and the target prompt information are input into the target model to obtain the execution result of the target task generated by the target model.
2. The data processing method according to claim 1 further includes: In response to the task creation instruction of the target task, a task input box is displayed in the target editing interface; Obtain the first task instruction input through the task input box; Based on the first task instruction, the task information is determined; Based on the first task instruction, determine the triggering conditions for the target task; Based on the task information and the triggering conditions, the target task is created; The target task is associated and bound to the target editing interface.
3. The data processing method according to claim 2, wherein creating the target task based on the task information and the triggering condition includes: Based on the task information and the triggering conditions, an initial task flow for the target task is created; In response to receiving an adjustment instruction for the initial task flow, the initial task flow is adjusted according to the adjustment instruction to obtain the target task flow for the target task; Based on the target task flow, the target task is created.
4. The data processing method according to claim 2, wherein determining the task information based on the first task instruction includes: Parse the first task instruction to obtain the task type and the target object of the first task instruction; Based on the task type of the first task instruction and the indication information of the target object, at least one first prompt fragment that matches the first task instruction is searched in the prompt fragment template library; At least one of the first prompt fragments and corresponding matching information are displayed on the target editing interface, the matching information including matching degree and / or number of times it is used; In response to a selection operation for at least one target first prompt fragment, each of the target first prompt fragments is determined as the initial prompt fragment.
5. The data processing method according to claim 1, further comprising: In response to an editing area creation command, the system obtains indication information of the target editing area for the target task, the indication information including location information. Based on the indication information of the target editing area, the target editing area is created in the target editing interface; The task information also includes indication information for the target editing area.
6. The data processing method according to claim 5, after obtaining the execution result of the target task generated by the target model, the data processing method further includes: The execution result of the target task is output to the target editing area.
7. The data processing method according to claim 1, wherein determining the target prompt information of the target task based on at least one of the initial prompt fragments includes: At least one first prompt component is displayed on the target editing interface, and the first prompt component is associated with a second prompt fragment; In response to an add operation for a target first prompt component in the at least one first prompt component, determine whether the add operation matches the target task; If a match is found, the second prompt fragment associated with the target first prompt component is obtained; Add the second prompt fragment associated with the first prompt component of the target to the target prompt information.
8. The data processing method according to claim 7, wherein determining the target prompt information of the target task based on at least one of the initial prompt fragments further comprises: At least one second prompt component is displayed in the target editing interface, and each second prompt component is associated with one of the initial prompt fragments; In response to a removal operation of a target second prompt component in the at least one second prompt component, an initial prompt fragment associated with the target second prompt component is obtained; Remove the initial prompt fragment associated with the target second prompt component from the target prompt information.
9. A data processing apparatus, comprising: The task triggering unit is used to obtain the task information of the target task in response to the triggering condition of the operation on the target editing interface hitting the target task. The task information includes the indication information of the target object and at least one initial prompt fragment. The content extraction unit is used to obtain the content information corresponding to the target object on the target editing interface based on the indication information of the target object; The prompt information acquisition unit is used to determine the target prompt information of the target task based on at least one of the initial prompt fragments, and the target prompt information is used to guide the target model to generate the execution result of the target task; The result generation unit is used to input the content information of the target object and the target prompt information into the target model to obtain the execution result of the target task generated by the target model.
10. An electronic device, comprising: A memory, a processor, and a computer program stored in the memory, wherein an intelligent agent is deployed in the processor, and the processor invokes the computer program to perform: In response to the triggering condition of the target task being hit by an operation on the target editing interface, the task information of the target task is obtained, the task information including the indication information of the target object and at least one initial prompt fragment; Based on the indication information of the target object, obtain the content information corresponding to the target object on the target editing interface; Based on at least one of the initial prompt fragments, target prompt information for the target task is determined, and the target prompt information is used to guide the target model to generate the execution result of the target task; The content information of the target object and the target prompt information are input into the target model to obtain the execution result of the target task generated by the target model.