Information processing method and device, storage medium and computer equipment

Abstract

The invention discloses an information processing method and device, a storage medium and computer equipment. The method comprises the steps that a main agent conducts intention recognition on user context information to obtain a to-be-executed target task, and the target task comprises at least one subtask; the main agent determines a sub-agent for executing the sub-task and a task execution permission of the sub-agent based on the task content of the sub-task and the description information of the sub-agent; the main agent sets a task identifier for the sub-task, and sends the task identifier, the task content and the task execution permission of the sub-task to the sub-agent; the sub-agent determines a task execution range based on the task content of the sub-task and the task execution permission, and executes the sub-task in the task execution range to obtain a task execution result; and the main agent generates reply information based on the task execution range and the task execution result of the sub-agent and sends the reply information to the client. The method can reduce the training cost of the model and improve the ability of the model to solve problems.

Description

Technical Field

[0001] The present application relates to the field of Internet technology, and in particular to an information processing method, apparatus, storage medium, and computer equipment. Background Art

[0002] In the e-commerce field, large model technology has been widely applied in various scenarios, such as information recommendation, information search, and intelligent customer service, to improve user experience and operational efficiency. Existing technologies primarily utilize pre-trained knowledge bases or invoke preset tools to respond to user questions. However, the questions users ask during actual interactions are often highly diverse and specialized, placing very high demands on the generalization capabilities and service adaptability of large models.

[0003] At present, although large models have a certain knowledge coverage capability through training on massive data, their training costs will increase exponentially as the knowledge coverage expands, and their problem-solving ability is limited by the timeliness of static knowledge bases and the flexibility of tool interfaces, making it difficult to meet the high-frequency update needs of users in e-commerce scenarios. In addition, when user questions involve cross-domain knowledge, large models often cannot effectively associate multi-dimensional information, resulting in deviations between their response results and user expectations, thus affecting user experience. Summary of the Invention

[0004] In view of this, the embodiments of the present application provide an information processing method, apparatus, storage medium and computer equipment, the main purpose of which is to solve the technical problems of high model training cost and limited problem-solving ability.

[0005] According to a first aspect of the present application, there is provided an information processing method, the method comprising: Obtaining user context information, and performing intent recognition on the user context information through a master agent to obtain a target task to be executed, wherein the target task includes at least one subtask; The master agent determines the sub-agent to execute the sub-task and the task execution authority of the sub-agent based on the task content of the sub-task and the description information of the sub-agent, wherein the task execution authority includes full execution authority and partial execution authority; The master agent sets a task identifier for the subtask and sends the task identifier, task content and task execution authority of the subtask to the sub-agent that executes the subtask; The sub-agent determines the task execution scope based on the task content of the sub-task and the task execution authority, and executes the sub-task within the task execution scope to obtain a task execution result; wherein, when the sub-agent executes the sub-task, if it is determined that there are missing task parameters, the parameter information of the missing task parameters is sent to the main agent; the main agent calls a parameter collection tool based on the parameter information, obtains the missing task parameters through the parameter collection tool, and sends the task parameters to the sub-agent; the sub-agent executes the sub-task based on the task parameters to obtain a task execution result; The main agent generates reply information based on the task execution scope and task execution results of the sub-agent, and sends the reply information to the client.

[0006] According to a second aspect of the present application, there is provided an information processing method, the method comprising: In response to receiving user input information, the user input information is sent to the server so that the server obtains user context information corresponding to the user input information, and performs intent recognition on the user context information through the main agent to obtain the target task to be executed, wherein the target task includes at least one subtask; the main agent determines the sub-agent that executes the subtask and the task execution authority of the sub-agent based on the task content of the subtask and the description information of the sub-agent, wherein the task execution authority includes full execution authority and partial execution authority; the main agent sets a task identifier for the subtask, and sends the task identifier, task content and task execution authority of the subtask to the sub-agent that executes the subtask. The sub-agent determines the task execution scope based on the task content of the sub-task and the task execution authority, and executes the sub-task within the task execution scope to obtain the task execution result; wherein, when the sub-agent executes the sub-task, if it is determined that there are missing task parameters, the parameter information of the missing task parameters is sent to the main agent; the main agent calls a parameter collection tool based on the parameter information, obtains the missing task parameters through the parameter collection tool, and sends the task parameters to the sub-agent; the sub-agent executes the sub-task based on the task parameters to obtain the task execution result; the main agent generates a reply message based on the task execution scope and task execution result of the sub-agent; The reply information is received and displayed.

[0007] According to a third aspect of the present application, there is provided an information processing device, the device comprising: A main agent, configured to obtain user context information, perform intent recognition on the user context information, and obtain a target task to be executed, wherein the target task includes at least one subtask; The master agent is further configured to determine a sub-agent to execute the sub-task and the task execution authority of the sub-agent based on the task content of the sub-task and the description information of the sub-agent, wherein the task execution authority includes full execution authority and partial execution authority; The master agent is further configured to set a task identifier for the subtask and send the task identifier, task content, and task execution authority of the subtask to the sub-agent that executes the subtask; The sub-agent is configured to determine a task execution scope based on the task content of the sub-task and the task execution authority, and execute the sub-task within the task execution scope to obtain a task execution result; wherein, when executing the sub-task, if the sub-agent determines that there are missing task parameters, the sub-agent sends parameter information of the missing task parameters to the main agent; the main agent calls a parameter collection tool based on the parameter information, obtains the missing task parameters through the parameter collection tool, and sends the task parameters to the sub-agent; the sub-agent executes the sub-task based on the task parameters to obtain a task execution result; The main agent is further used to generate reply information based on the task execution scope and task execution results of the sub-agent, and send the reply information to the client.

[0008] According to a fourth aspect of the present application, there is provided an information processing device, the device comprising: An information sending module is used to send the user input information to the server in response to receiving the user input information, so that the server obtains the user context information corresponding to the user input information, and performs intention recognition on the user context information through the main agent to obtain the target task to be executed, wherein the target task includes at least one subtask; the main agent determines the sub-agent that executes the subtask and the task execution authority of the sub-agent based on the task content of the subtask and the description information of the sub-agent, wherein the task execution authority includes full execution authority and partial execution authority; the main agent sets a task identifier for the subtask, and sends the task identifier, task content and task execution authority of the subtask to the sub-agent that executes the subtask. The sub-agent of the task; the sub-agent determines the task execution scope based on the task content of the sub-task and the task execution authority, and executes the sub-task within the task execution scope to obtain the task execution result; wherein, when the sub-agent executes the sub-task, if it is determined that there are missing task parameters, the parameter information of the missing task parameters is sent to the main agent; the main agent calls the parameter collection tool based on the parameter information, obtains the missing task parameters through the parameter collection tool, and sends the task parameters to the sub-agent; the sub-agent executes the sub-task based on the task parameters to obtain the task execution result; the main agent generates a reply message based on the task execution scope and task execution result of the sub-agent; The information display module is used to receive and display the reply information.

[0009] According to a fifth aspect of the present application, a storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the above-mentioned information processing method is implemented.

[0010] According to the sixth aspect of the present application, a computer device is provided, comprising a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor implements the above-mentioned information processing method when executing the program.

[0011] By means of the above technical solution, an information processing method, apparatus, storage medium and computer equipment provided by the embodiments of the present application can decompose the core demands of users into independently executable operation units by using the main agent to identify the intention and decompose the user context information. In addition, by assigning a suitable sub-agent to each sub-task, assigning task execution authority to each sub-agent, and determining the task execution scope through the sub-agent, the execution status of the task can be accurately evaluated before the task is executed, thereby improving the accuracy and controllability of the task execution results. Finally, by generating reply content using the task execution scope and task execution results of the sub-agent, the problem-solving capabilities of different sub-agents in their respective fields can be utilized to enhance the model's cross-domain problem-solving capabilities and the ability to associate multi-dimensional information, thereby reducing the training difficulty of the main agent. The above method can effectively improve the model's problem-solving and task execution capabilities, making the model more flexible and timely, thereby enhancing the user experience.

[0012] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] 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: Figure 1 A flow chart of an information processing method provided by an embodiment of the present application is shown; Figure 2 A flow chart showing another information processing method provided in an embodiment of the present application is shown; Figure 3 A schematic diagram illustrating a scenario of an information processing method provided by an embodiment of the present application is shown; Figure 4 A schematic diagram showing a flow chart of another information processing method provided in an embodiment of the present application is shown; Figure 5 A schematic diagram illustrating another information processing method provided in an embodiment of the present application is shown; Figure 6 A schematic diagram of the interaction process between a master agent and a sub-agent when executing the first embodiment of the information processing method provided by an embodiment of the present application is shown; Figure 7 A schematic diagram of the interaction process between the main agent and the sub-agent when executing the second embodiment of the information processing method provided by the embodiment of the present application is shown; Figure 8 A schematic diagram of the interaction process between the main agent and the sub-agent when executing the third embodiment of the information processing method provided in an embodiment of the present application is shown; Figure 9 A schematic diagram of the interaction process between the main agent and the sub-agent when executing the fourth embodiment of the information processing method provided in an embodiment of the present application is shown; Figure 10 A schematic diagram of the interaction process between the main agent and the sub-agent when executing the fifth embodiment of the information processing method provided in an embodiment of the present application is shown; Figure 11 A schematic diagram of the structure of a memory system provided in an embodiment of the present application is shown; Figure 12 A schematic structural diagram of an information processing device provided in an embodiment of the present application is shown; Figure 13 A schematic structural diagram of another information processing device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0014] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that, unless there is a conflict, the embodiments and features in the embodiments of the present application can be combined with each other.

[0015] In one embodiment, Figure 1 As shown, an information processing method is provided, which is described by taking the application of the method to a computer device such as a server as an example, and includes the following steps: Step 101: In response to receiving user input information, the client sends the user input information to the server.

[0016] Specifically, the user can input information through any interface of the client. For example, the user can send a conversation message through the client by initiating a conversation, or can send search keywords and other information through the client by entering a search keyword. In this embodiment, the purpose of the user inputting information is to solve a certain problem, such as information search problems, query problems, consultation problems, complaint problems, etc. The application scenarios are very broad, and thus the user identity can also be set according to the actual scenario. For example, in a shopping scenario, the user can be a shopping customer; in a merchant service scenario, the user can be a merchant.

[0017] In step 102, the server obtains user context information and performs intent recognition on the user context information through the main agent to obtain a target task to be executed, which includes at least one subtask.

[0018] User context information refers to the historical behavior data and real-time request content generated by users during their interactions with e-commerce platforms. For example, user context information can include multi-dimensional information such as user search keywords, current conversation content, and user tags. The main agent refers to an intelligent decision-making module with global task scheduling capabilities, which can be implemented through pre-trained information processing models. Information processing models refer to artificial intelligence models that can handle a certain amount of tasks and data, such as large language models (LLMs), computer vision models (CVs), large multimodal models (LMMs), and small models formed through information processing model distillation technology.

[0019] Furthermore, intent recognition refers to the process by which the main agent analyzes the user's core needs through the natural language processing capabilities provided by the information processing model. The target task refers to the complete solution to the user's problem, and subtasks refer to the specific operations that must be performed to achieve the target task. The target task can be a single task (i.e., consisting of only one subtask) or can be broken down into multiple subtasks. The subtasks can be parallel or logically sequential.

[0020] Specifically, after receiving user input, the server can use a master agent with natural language processing capabilities to obtain user context and perform intent recognition on this context to determine the user's core needs, specifically the target task to be performed. The master agent can then break down the target task into one or more subtasks based on actual needs. By performing intent recognition on user context and breaking down the identified target tasks, these steps can transform complex user needs into more manageable modular operational units, facilitating a more detailed and accurate understanding of user intent.

[0021] Step 103: The main agent determines the sub-agent that executes the sub-task and the task execution authority of the sub-agent based on the task content of the sub-task and the description information of the sub-agent.

[0022] Among them, a sub-agent refers to an intelligent decision-making module with problem-solving capabilities in a certain professional field. This intelligent decision-making module can also be implemented through a pre-trained information processing model. However, when training the model, the sub-agent focuses more on using data from the corresponding field for vertical training, so that the model has more accurate and efficient problem-solving capabilities in a certain field. Correspondingly, the descriptive information of the sub-agent refers to the pre-defined explanatory information used to describe the functional boundaries and technical parameters of the sub-agent. Furthermore, task execution authority refers to the level of operational authority for processing sub-tasks granted by the main agent to the sub-agent after intent recognition and task decomposition. Specifically, it may include full execution authority and partial execution authority, etc.

[0023] Specifically, after obtaining the task content of each subtask and the descriptive information of each sub-agent, the main agent can match the task content of the subtask with the descriptive information of the sub-agent, and based on the matching results, determine the sub-agent to execute each subtask, as well as the task execution authority of each sub-agent for the subtask. In this embodiment, when the target task is not split (i.e., the target task contains only one subtask), the main agent can directly assign a sub-agent to the target task based on the task content of the target task and determine the task execution authority of the sub-agent to execute the target task; when the target task is split into multiple subtasks, the main agent can assign a sub-agent to each subtask based on the task content of each subtask and determine the task execution authority of each sub-agent to execute the subtask.

[0024] In this embodiment, the main agent can split the target task according to the description information of the sub-agent, that is, the main agent can split the target task according to the field to which the task content belongs, rather than splitting it according to the number of tasks. In this way, the sub-agents can be targeted to execute sub-tasks in the corresponding field and obtain more accurate execution results. In this process, if it is impossible to match a suitable sub-agent for the sub-task, the task can be assigned without the main agent. Instead, the main agent directly completes the execution of the corresponding sub-task and obtains the task execution result. By assigning a suitable sub-agent to each sub-task and assigning task execution authority to each sub-agent, the above steps can accurately evaluate the execution of the task before the task is executed, thereby improving the accuracy and controllability of the task execution results of the sub-task.

[0025] Step 104: The main agent sets a task identifier for the subtask, and sends the task identifier, task content, and task execution authority of the subtask to the sub-agent that executes the subtask.

[0026] Among them, the task identifier refers to the unique identifier generated by the main agent for the subtask, which is used to track the execution status of the subtask throughout the process. Specifically, after the task is disassembled, the main agent can set a unique task identifier for the subtask, and send the set task identifier, the disassembled task content and the task execution authority of the sub-agent to the sub-agent that executes the subtask, so that the sub-agent can perform subsequent processing. In this embodiment, the main agent can assign a task identifier to the subtask executed by the sub-agent to facilitate the tracking of the task execution status, and also facilitate the storage and backtracking of task data; for the subtasks directly executed by the main agent, it can be determined according to the situation, and the task identifier can be set or not.

[0027] In step 105, the sub-agent determines the task execution scope based on the task content and task execution authority of the sub-task, and executes the sub-task within the task execution scope to obtain the task execution result.

[0028] The task execution scope refers to the degree of task execution re-determined by the sub-agent within its task execution authority after evaluation. This can include full execution, partial execution, or no execution. Specifically, after receiving the subtask's task content and task execution authority, the sub-agent can re-evaluate the subtask's content and determine the task execution scope within the task execution authority determined by the main agent. It can then execute the subtask within the task execution scope to obtain the subtask's task execution result.

[0029] In this embodiment, the task execution scope determined by the sub-agent is equal to or lower than the task execution authority determined by the master agent for the sub-agent. For example, when the task execution authority is full, the task execution scope can be full execution, partial execution, or no execution; when the task execution authority is partial, the task execution scope can only be partial execution or no execution. This ensures that the sub-agent can accurately execute the target task or perform parts of the target task within its capabilities.

[0030] Step 106: The main agent generates a reply message based on the task execution scope and task execution result of the sub-agent, and sends the reply message to the client.

[0031] The reply information refers to the final response content generated by the main agent based on the task execution results of the sub-agent, which may include text content, visual charts, interactive components, etc. Specifically, after receiving the task execution scope and task execution results sent by the sub-agent, the main agent can first determine the degree of completion of the sub-agent's task execution results for the target task based on the task execution scope. Then, based on the degree of completion of the target task, the task execution results of the sub-task can be directly output, or the task execution results of the sub-agent can be reintegrated, supplemented, or other tools can be called to execute the target task or supplement the execution content, and then the final reply information is sent to the client.

[0032] In this embodiment, the main agent can process the sub-agent's task execution results in a targeted manner to generate reply content based on the different task execution scopes of the sub-agent. For example, when the sub-agent's task execution scope is full execution, the main agent can directly generate reply content based on the sub-agent's feedback results to reduce the main agent's data processing workload; when the sub-agent's task execution scope is partial execution, the main agent can organize and integrate the sub-agent's feedback results or call other tools to supplement them to generate reply information, thereby improving the reply content; when the sub-agent's task execution scope is non-execution, the main agent can redefine the task execution steps and the tools to be called based on the task content of the target task, and generate reply information to ensure the accuracy of the reply content.

[0033] Step 107: The client receives and displays the reply information.

[0034] Specifically, after receiving the reply information, the client can display the reply information through a variety of components. Specifically, the client can display the reply information in multiple dimensions through text flow, graphic display components, and interactive components, among which a variety of components can be streamed according to actual conditions. For example, when the reply information contains text information, chart information, and action point information at the same time, the client can first display the chart through the graphic display component, then display the text flow information, and finally output the action point card. Based on this, the user can first view the chart, then view the text information, and finally decide whether to adopt the solution provided by the action point card. If the decision is to adopt it, the action point adoption information can be submitted through the action point card.

[0035] The above embodiment uses the main agent to identify the intention of the user context information and decompose the task, which can decompose the core demands of the user into independently executable operation units. In addition, by assigning a suitable sub-agent to each sub-task, assigning task execution authority to each sub-agent, and determining the task execution scope through the sub-agent, the execution status of the task can be accurately evaluated before the task is executed, thereby improving the accuracy and controllability of the task execution results. Finally, by generating reply content based on the task execution scope and task execution results of the sub-agent, the problem-solving capabilities of different sub-agents in their respective fields can be utilized to enhance the model's cross-domain problem-solving capabilities and the ability to associate multi-dimensional information, thereby reducing the training difficulty of the main agent. The above method can effectively improve the model's problem-solving and task execution capabilities, making the model more flexible and timely, thereby enhancing the user experience.

[0036] In the above embodiment, if Figure 2 As shown, the above information processing method further includes the following steps: Step 201: When the sub-agent is executing a sub-task, if it is determined that there are missing task parameters, the parameter information of the missing task parameters will be sent to the main agent.

[0037] In step 202 , the master agent calls a parameter collection tool based on the parameter information of the missing task parameters, and sends the parameter information of the task parameters to the client through the parameter collection tool.

[0038] Step 203: The client displays a parameter collection component, wherein the parameter collection component displays parameter information of the task parameters.

[0039] In step 204, the client receives the task parameters through the parameter collection component and sends the task parameters to the main agent, which then sends the task parameters to the sub-agent.

[0040] In step 205, the sub-agent executes the sub-task based on the task parameters and obtains the task execution result.

[0041] Specifically, the above steps 201 to 205 are a further supplement to the above embodiment. In this embodiment, the sub-agent can also determine whether there are missing task parameters in the task execution process in the process of determining the task execution scope based on the task content of the sub-task and the task execution authority. If it is determined that there are missing task parameters, the parameter collection tool can be called through the main agent, and the task parameters can be obtained from the client through the main agent. After the main agent obtains the task parameters, the task parameters can be sent to the sub-agent so that the sub-agent can continue to execute the sub-task and obtain the task execution result of the sub-task. In this embodiment, in the process of the sub-agent requesting the task parameters from the client through the main agent, it is necessary to carry the task identifier for data transmission, so as to facilitate data backtracking, storage and task status tracking.

[0042] For example, in the "flash sale merchant service scenario", suppose the main agent assigns the subtask "how is the recent positive review rate" to a sub-agent. When the sub-agent performs the subtask, if it determines that the necessary time range information is missing, it can request the time range information from the client in the above way. Figure 3 , the client will display the corresponding parameter collection component to collect the missing task parameters. After the user feedbacks the time range information as "20XX0801-20XX0831" through the client, the main agent can feedback this time range information to the sub-agent so that the sub-agent can perform the query task of the praise rate.

[0043] The above embodiment uses the main agent to interact with the client and sub-agent respectively, and can use the main agent as the entrance and exit for interaction with the client, thereby avoiding repeated development of the interaction interface of each sub-agent, thereby reducing the development workload of the system and reducing the complexity of the system.

[0044] In one embodiment, the above-mentioned step 106 can be implemented as follows: the main agent calls at least one information display tool based on the reply information, and sends the reply information to the client through the information display tool, wherein the information display tool may include at least one of a graphic display tool, a text flow display tool, and an action point display tool. Correspondingly, in step 107, the client can display the reply information fed back by the main agent through at least one preset component, wherein the preset component may include at least one of a graphic display component, a text flow display component, and an action point display component.

[0045] In the above embodiment, after generating the reply information, the main agent can call the corresponding information display tool based on the content contained in the reply information to display the reply information in multiple dimensions. Figure 5The reply message can include the query result, the reply text, and the action point information. In this case, the main agent can first call the graphic display tool to send the query result, then call the text flow display tool to send the text information, and finally call the action point display tool to send the action point information. Accordingly, after receiving the above information, the client can first display the query information through the graphic display component, then display the text information through the text flow display component, and finally display the action point information through the action point display component.

[0046] The above embodiment uses a variety of information display tools to send reply information, and uses a variety of preset components to display reply information, so that users can obtain multi-dimensional information such as charts, text flows, and action points. At the same time, other tools and components can be nested in these tools and components to further enrich the form of information display. For example, in the graphic display component, multiple charts can be displayed at the same time, and the corresponding explanatory text of the charts can be displayed; in the text flow display component, in addition to displaying copy information, other content such as links and flow charts can also be displayed; in the action point display component, solutions can be directly provided to users, so that users can send action point adoption information by just clicking a button, or an action point setting entrance can be provided to users, so that users can directly set functions through action point cards, etc. In the above manner, on the one hand, the way of information display can be enriched, and on the other hand, the efficiency of problem solving can be improved.

[0047] In one embodiment, Figure 4 As shown, the above information processing method further includes the following steps: Step 301: The client displays action point information through an action point display component.

[0048] Step 302: The client receives action point adoption information through the action point display component and sends the action point adoption information to the server.

[0049] Step 303: In response to receiving the action point adoption information, the server processes the action point based on the action point adoption information and sends a message indicating successful execution of the action point to the client.

[0050] Step 304: The client receives and displays a message indicating that the action point has been successfully executed.

[0051] In the above embodiment, when the reply message generated by the master agent includes action point information, the action point display tool can be called to send the action point information to the client, so that the client can display the action point information through the action point display component. Action point information refers to the executable action plan provided by the master agent or sub-agent to the user based on the target task, which is part of the task execution result. Based on this information, the user can decide whether to accept it. If so, the action point acceptance information is sent to the server through the action point display component, causing the server to perform the corresponding operation. The server then sends a message indicating that the action point has been successfully executed to the client, which displays the message on the client side, allowing the user to promptly understand the execution status of the action point.

[0052] In this embodiment, after the server sends a message confirming the successful execution of an action point to the client, the client can first forward this message to the main agent. The main agent then invokes an information display tool based on this information and uses the information display tool to send the successful execution message to the client. This allows the client to display the successful execution message of the action point in the form of a component, thereby improving the user experience. Furthermore, for some action points that require changes to the user's original settings, the information display tool can be invoked to resend a confirmation message to the client before execution. After receiving the user's confirmation message, the server then processes the action point, avoiding user inconvenience caused by incorrect action point processing.

[0053] For example, in the "Flash Sale Merchant Service Scenario", refer to Figure 5 Assuming that the reply message generated by the main agent contains the action point information of "turning on order reminders and automatic order acceptance", after receiving the above information, the main agent can display the two action point information of "turning on order reminders" and "turning on automatic order acceptance" respectively through the action point display component. Users can choose the action point they want to execute according to their needs. When the user wants to turn on "automatic order acceptance", they can click the corresponding turn-on button of the "automatic order acceptance" component to send the action point adoption information with one click. The server can turn on "automatic order acceptance" based on this information and send a message to the client indicating that the action point has been successfully executed, so that the user can understand the execution status of the action point.

[0054] In one embodiment, the task execution authority determined by the main agent for the sub-agent may include full execution authority and partial execution authority. The task execution scope determined by the sub-agent based on the task execution authority is lower than or equal to the task execution authority assigned by the main agent, and may specifically include full execution, partial execution, and no execution. Based on the above-mentioned task allocation and execution methods, five task processing methods can be generated for different task execution authorities and task execution scopes, namely "full execution authority and full execution", "full execution authority but partial execution", "full execution authority but no execution", "partial execution authority and partial execution", and "partial execution authority but no execution". Taking the "flash sale merchant service scenario" as an example, the implementation methods of the above five task processing methods are specifically explained.

[0055] Implementation method 1: All execution permissions and all execution. Figure 6 As shown in the figure, in this task processing method, the interaction between the main agent and the sub-agent is as follows: In step 1031, the main agent determines the sub-agent to execute the sub-task based on the task content of the sub-task and the description information of the sub-agent, and determines that the task execution authority of the sub-agent is full execution authority.

[0056] Step 1041: The main agent sets a task identifier for the subtask, and sends the task identifier, task content, and all execution permissions of the subtask to the sub-agent that executes the subtask.

[0057] Step 1051: The sub-agent determines that the task execution scope is full execution based on the task content and all execution permissions of the sub-task, and executes all the contents of the sub-task to obtain the task execution result.

[0058] Step 1061: The main agent generates a reply message based on the task execution result of the sub-agent.

[0059] For example, in the "Flash Sale Merchant Service Scenario," when a user enters the message "Traffic has been slow recently, how can I improve it?", the main agent, based on the user's context, determines the target task to be "Check recent traffic and explain how to improve it." The main agent then determines, based on the sub-agent's description, that a sub-agent can fully address the issue and grants it "Full Execution Permissions." The main agent then sends the sub-agent its task ID, task details, and execution permissions. After analysis, the sub-agent determines it can directly answer the question, but lacks the time range for checking traffic. It then sends a confirmation of the time to the main agent. The main agent then invokes a tool to request the time range from the client and sends the requested time range, "20XX0801-20XX0831," to the sub-agent. The sub-agent then performs a traffic query based on this time range, generates a response to the merchant, and provides action points. The results of this processing are then sent to the main agent. After receiving the above information, the main intelligent agent can call multiple tools to send the above information to the client, so that the client can display traffic cards, reply texts to merchants and action point cards in sequence through multiple components, so that merchants can understand the cause of the problem and get corresponding solutions.

[0060] In the above implementation, when the master agent determines that a sub-agent can fully resolve a problem, it can grant it "full execution permissions." Upon further evaluation, the sub-agent, upon determining it can indeed resolve the problem, will provide the master agent with a task execution scope, within which it will continue to execute the task. Upon receiving the "full execution" task execution scope from the sub-agent, the master agent can grant full resolution authority to the sub-agent and directly output a response message based on the sub-agent's processing results. This avoids duplicate task processing, conserves computing resources, and enables more professional task processing.

[0061] Implementation method 2: Full execution authority but partial execution. Figure 7 As shown in the figure, in this task processing method, the interaction between the main agent and the sub-agent is as follows: In step 1032, the main agent determines the sub-agent to execute the sub-task based on the task content of the sub-task and the description information of the sub-agent, and determines that the task execution authority of the sub-agent is full execution authority.

[0062] Step 1042: The main agent sets a task identifier for the subtask, and sends the task identifier, task content, and all execution permissions of the subtask to the sub-agent that executes the subtask.

[0063] In step 1052, the sub-agent determines that the task execution scope is partial execution based on the task content and all execution permissions of the sub-task, and executes at least part of the content of the sub-task to obtain the task execution result.

[0064] Step 1062: The main agent processes the task execution results of the sub-agent and generates a reply message, or determines the task execution tool based on the task execution results and target task of the sub-agent, and generates a reply message based on the task execution results of the sub-agent and the execution results of the task execution tool.

[0065] For example, in the "Flash Sale Merchant Service Scenario," when a user enters the question "How are my recent positive reviews?", the main agent, based on the user context, determines the target task to be "Query and evaluate recent positive reviews." The main agent then determines, based on the sub-agent's description, that a sub-agent is fully capable of solving the problem and grants it "full execution permissions." The main agent then sends the sub-agent the task identifier, task content, and execution permissions. After analysis, the sub-agent determines that it cannot directly answer the question, but can provide a partial answer. However, it lacks the necessary timeframe information. The sub-agent then sends the timeframe confirmation information to the main agent. The main agent then invokes a tool to request the timeframe information from the client and sends the requested timeframe, "20XX0801-20XX0831," to the sub-agent. The sub-agent then performs a positive review query based on this timeframe information, generates a positive review query result, and sends this result to the main agent. After receiving the above information, the main intelligent agent can further evaluate the user's praise rate based on the "query results of the praise rate" and generate a reply message to "How is the recent praise rate?" Finally, the tool is called to send the generated reply message to the client, so that the client can display the praise rate card and reply to the merchant through the preset component, so that the merchant can understand the issues they want to consult in a timely manner and get corresponding solutions.

[0066] In the above implementation, for the "full execution authority" granted by the main agent, if the sub-agent, after re-evaluation, believes that it cannot completely solve the current problem, it will also feedback a task execution scope to the main agent and continue to execute the task within this scope. When the main agent receives the "partial execution" task execution scope feedback from the sub-agent, it can give partial authority to solve the problem to the sub-agent, and then generate reply content based on the sub-agent's processing results, or call other tools based on the sub-agent's processing results to obtain the final reply information. In this way, the processing results of multiple agents can be integrated, saving the main agent's computing power, and at the same time, ensuring the professionalism of task execution.

[0067] Implementation method three: All execution permissions but not execution. Figure 8 As shown in the figure, in this task processing method, the interaction between the main agent and the sub-agent is as follows: In step 1033, the main agent determines the sub-agent to execute the sub-task based on the task content of the sub-task and the description information of the sub-agent, and determines that the task execution authority of the sub-agent is full execution authority.

[0068] Step 1043: The main agent sets a task identifier for the subtask, and sends the task identifier, task content, and all execution permissions of the subtask to the sub-agent that executes the subtask.

[0069] In step 1053, when the sub-agent determines that the task execution scope is not to be executed based on the task content and all execution permissions of the sub-task, it generates a rejection execution message as the task execution result.

[0070] In step 1063, the master agent determines a task execution tool based on the target task, and generates reply information based on the execution result of the task execution tool.

[0071] For example, in the "Flash Sale Merchant Service Scenario," when a user enters the question "What is my most recent store rating?", the main agent, based on the user context, determines the target task to be "What is the most recent store rating?" The main agent then determines, based on the sub-agent's description, that a sub-agent is capable of fully resolving the issue and grants it "full execution permissions." The main agent then sends the sub-agent the task identifier, task content, and execution permissions. After analysis, the sub-agent determines it cannot directly answer the question or provide a partial answer, and then returns a rejection as a task execution result to the main agent. Based on this rejection, the main agent considers invoking other tools to solve the current task. The main agent then uses the store rating query tool to query the current store rating based on the target task's content. Based on the query results, it generates a response to the question "What is the most recent store rating?" The tool then sends the generated response to the client, which displays the response to the merchant through a pre-set component, allowing the merchant to promptly understand the inquiry and obtain a solution.

[0072] In the above implementation, if a sub-agent, after reassessing its inability to resolve the current problem or provide a partial answer to the "full execution permission" granted by the master agent, determines that it is unable to do so, it can send back a rejection message to the master agent. Upon receiving the sub-agent's "non-execution" response, the master agent can then reselect other tools based on the task content (at least one tool can be invoked in the order of execution) and receive a final response. This approach allows sub-agents to accurately assess their task execution capabilities, avoid hallucinations, and ensure professional execution.

[0073] Implementation method 4: Partial execution authority and partial execution. Figure 9 As shown in the figure, in this task processing method, the interaction between the main agent and the sub-agent is as follows: In step 1034, the main agent determines the sub-agent to execute the sub-task based on the task content of the sub-task and the description information of the sub-agent, and determines that the task execution authority of the sub-agent is a partial execution authority.

[0074] Step 1044: The main agent sets a task identifier for the subtask, and sends the task identifier, task content, and partial execution authority of the subtask to the sub-agent that executes the subtask.

[0075] In step 1054, the sub-agent determines that the task execution scope is partial execution based on the task content and partial execution authority of the sub-task, and executes at least part of the content of the sub-task to obtain the task execution result.

[0076] Step 1064: The main agent processes the task execution results of the sub-agent and generates a reply message, or determines the task execution tool based on the task execution results and target task of the sub-agent, and generates a reply message based on the task execution results of the sub-agent and the execution results of the task execution tool.

[0077] For example, in a flash sale merchant service scenario, when a user enters the question "Why has my positive review rate recently dropped?", the main agent can, based on the user context, decompose the target task into two subtasks: "What are my recent positive review rates?" and "Why has my positive review rate dropped?" It then assigns the "What are my recent positive review rates?" subtask to a subagent and grants it "partial execution permissions." The main agent then sends the subagent the task identifier, task content, and execution permissions for the subtask. After analysis, the subagent determines that it cannot directly answer the question, but can provide a partial answer. However, it lacks the necessary timeframe information. It then sends the timeframe confirmation information to the main agent. The main agent then invokes a tool to request the timeframe information from the client and sends the requested timeframe information, "20XX0801-20XX0831," to the subagent. The subagent then performs a positive review query based on this timeframe information and generates a "positive review query result." Subsequently, the sub-agent sends the "query results of the praise rate" to the main agent. After receiving the above information, the main agent calls the knowledge query retrieval tool to query the "reasons for the decline in the praise rate" and obtains the knowledge retrieval results of the "reasons for the decline in the praise rate". Finally, based on the "query results of the praise rate" and the knowledge retrieval results of the "reasons for the decline in the praise rate", it generates a reply message to "Why has the recent praise rate declined?" and calls the tool to send the generated reply message to the client, so that the client can display the praise rate card and the reply text to the merchant through the preset components, so that the merchant can understand the issues he wants to consult in a timely manner and get the corresponding solutions.

[0078] In the above implementation, for the "partial execution authority" granted by the main agent, if the sub-agent, after re-evaluation, believes that it can provide a partial answer, it will feedback a task execution scope to the main agent and continue to execute the task within this scope. When the main agent receives the "partial execution" task execution scope feedback from the sub-agent, it can transfer partial authority to solve the problem to the sub-agent and then generate reply content based on the sub-agent's processing results, or call other tools based on the sub-agent's processing results to obtain the final reply information. In this way, the processing results of multiple agents can be integrated, saving the main agent's computing power, while ensuring the professionalism of task execution.

[0079] Implementation method five: Partial execution of permissions but not execution. Figure 10 As shown in the figure, in this task processing method, the interaction between the main agent and the sub-agent is as follows: In step 1035, the main agent determines the sub-agent to execute the sub-task based on the task content of the sub-task and the description information of the sub-agent, and determines that the task execution authority of the sub-agent is a partial execution authority.

[0080] Step 1045: The main agent sets a task identifier for the subtask, and sends the task identifier, task content, and partial execution authority of the subtask to the sub-agent that executes the subtask.

[0081] In step 1055, when the sub-agent determines that the task execution scope is not to be executed based on the task content and partial execution authority of the sub-task, it generates execution refusal information as the task execution result.

[0082] In step 1065 , the master agent determines a task execution tool based on the target task, and generates reply information based on the execution result of the task execution tool.

[0083] For example, in the "Flash Sale Merchant Service Scenario," when a user enters the question "How can I improve my store's food delivery speed?", the main agent, based on the user's context, breaks the target task into two subtasks: "What is the store's food delivery time?" and "How can I improve food delivery speed?" It then assigns the "What is the store's food delivery time?" subtask to a subagent and grants it "partial execution permissions." The main agent then sends the subagent the task identifier, task content, and execution permissions for the subtask. After analysis, the subagent determines that it cannot directly answer the question or provide a partial answer. It then sends a rejection message as a task execution result back to the main agent. Based on the refusal to execute information fed back by the sub-agent, the main agent chooses to call other tools to solve the current task. The main agent first queries the "current serving time of the store" through the serving time query tool, and then calls the knowledge query retrieval tool to query "how to improve the serving speed" to obtain the knowledge retrieval results of "how to improve the serving speed". Finally, based on the query results of "current serving time of the store" and the knowledge retrieval results of "how to improve the serving speed", it generates a reply information to "how to improve the serving speed of the store", and calls the tool to send the generated reply information to the client, so that the client can display the reply text to the merchant through the preset component, so that the merchant can understand the questions he wants to consult in time and get the corresponding solutions.

[0084] In the above implementation, if a sub-agent, after reassessing its inability to resolve the problem or provide a partial answer, determines that it is unable to perform the task granted by the master agent, it can send a rejection message back to the master agent. Upon receiving the sub-agent's "non-execution" response, the master agent can then reselect other tools based on the task content (at least one tool can be invoked in the order of execution) and receive a final response. This approach allows sub-agents to accurately assess their task execution capabilities, avoids hallucinations, and ensures professional execution.

[0085] In one embodiment, the above-mentioned information processing method further includes the following steps: in response to an information update request, storing the updated information in a memory system, wherein the information covered by the information update request includes at least one of user context information, task parameters, reply information, action point adoption information, and a message indicating successful execution of the action point.

[0086] In the above embodiments, during the execution of the information processing methods described in the above embodiments, the updated information on the task execution status generated by each execution body can be automatically stored in the memory system. The premise of information storage is to obtain the user's prior authorization, and the stored information may include user input information received by the client, user context information obtained by the main agent, task parameters requested by the main agent or sub-agent to the client, reply information generated by the main agent, action point adoption information received by the client, and action point execution success message sent by the server, etc. By updating the above information to the memory system in a timely manner, it is convenient to update the user's demand information in a timely manner and have a more detailed understanding, so that it is convenient to provide targeted services to users, thereby improving the user experience.

[0087] In one embodiment, Figure 11 As shown, the memory system includes a session data layer, the session data layer includes at least one target data layer, the target data layer includes at least one task data layer, the task data layer includes at least one model data layer and at least one behavior data layer, user context information is stored in the task data layer, task parameters are stored in a model data layer of the task data layer, and reply information, action point adoption information, and action point execution success message are stored in a behavior data layer of the task data layer.

[0088] In the above embodiment, by storing different information in different data layers of the memory system in a hierarchical manner, it is easy to classify and organize the data, thereby improving the efficiency and accuracy of data query. For example, in the current round of dialogue, the sub-agent requests the time range information of the information query from the client through the main agent during the execution of the task. Then in the next dialogue, if the user makes an information query request again, the main agent or the sub-agent can directly query the time range information just obtained in the previous round in the corresponding data layer of the memory system, and directly perform information query based on the time range information. In this way, on the one hand, it can avoid users from repeatedly providing the same information and improve the user experience. On the other hand, it can improve the efficiency of data query and the efficiency of problem solving.

[0089] The technical solution of the present invention can be applied to the transaction and delivery services of instant e-commerce platforms, such as Taobao flash sales, Taoxianda, Ele.me takeout and retail, etc.

[0090] 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, storage, and display, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. In addition, the numbers corresponding to the various steps in the above embodiments serve only as identifiers and do not limit the order in which the steps are executed. The order in which the steps are executed in each embodiment can be set according to actual circumstances.

[0091] Further, as Figures 1 to 11 The specific implementation of the method, the embodiment of the present application provides an information processing device, such as Figure 12 As shown, the device includes a main agent and at least one sub-agent, wherein: The main agent 41 may be configured to obtain user context information and perform intent recognition on the user context information to obtain a target task to be executed, wherein the target task includes at least one subtask; The main agent 41 may also be used to determine the sub-agent that executes the sub-task and the task execution authority of the sub-agent based on the task content of the sub-task and the description information of the sub-agent; The main agent 41 may also be used to set a task identifier for the subtask, and send the task identifier, task content and task execution authority of the subtask to the sub-agent that executes the subtask; The sub-agent 42 may be configured to determine a task execution scope based on the task content of the sub-task and the task execution authority, and execute the sub-task within the task execution scope to obtain a task execution result; The main agent 41 can also be used to generate reply information based on the task execution scope and task execution results of the sub-agent, and send the reply information to the client.

[0092] In a specific application scenario, the task execution authority includes full execution authority and partial execution authority; the sub-agent 42 can be specifically used to, under the full execution authority, when the task execution scope is determined to be full execution based on the task content, execute the entire content of the subtask to obtain the task execution result; and / or, under the full execution authority, when the task execution scope is determined to be partial execution based on the task content, execute at least part of the content of the subtask to obtain the task execution result; and / or, under the full execution authority, when the task execution scope is determined to be non-execution based on the task content, generate a refusal to execute information as the task execution result; and / or, under the partial execution authority, when the task execution scope is determined to be partial execution based on the task content, execute at least part of the content of the subtask to obtain the task execution result; and / or, under the partial execution authority, when the task execution scope is determined to be non-execution based on the task content, generate a refusal to execute information as the task execution result.

[0093] In a specific application scenario, the device also includes a parameter collection tool 43, and the sub-agent 42 is also used to send the parameter information of the missing task parameters to the main agent when executing the sub-task if it is determined that there are missing task parameters; the main agent 41 is also used to call the parameter collection tool 43 based on the parameter information, and obtain the missing task parameters through the parameter collection tool 43, and send the task parameters to the sub-agent; the sub-agent 42 is also used to execute the sub-task based on the task parameters to obtain the task execution result.

[0094] In a specific application scenario, the device also includes a task execution tool 44, and the main agent 41 is also used to generate a reply message based on the task execution result of the sub-agent when the task execution scope of the sub-agent is full execution; and / or, when the task execution scope of the sub-agent is partial execution, process the task execution result of the sub-agent to generate a reply message; and / or, when the task execution scope of the sub-agent is partial execution, determine the task execution tool 44 based on the task execution result of the sub-agent and the target task, and generate a reply message based on the task execution result of the sub-agent and the execution result of the task execution tool 44; and / or, when the task execution scope of the sub-agent is non-execution, determine the task execution tool 44 based on the target task, and generate a reply message based on the execution result of the task execution tool 44.

[0095] In a specific application scenario, the device also includes an information display tool 45, and the main intelligent entity 41 is also used to call at least one information display tool 45 based on the reply information, and send the reply information to the client through the information display tool 45, wherein the information display tool includes at least one of a graphic display tool, a text flow display tool, and an action point display tool.

[0096] In a specific application scenario, the information processing device also includes an action point server 46, wherein the action point server 46 can be used to respond to receiving action point adoption information when the action point display tool is included in the called information display tool, perform action point processing based on the action point adoption information, and send a message of successful action point execution to the client.

[0097] In a specific application scenario, the information processing device also includes a memory system 47, wherein the information processing device can store the updated information in the memory system 47 in response to an information update request, wherein the information covered by the information update request includes at least one of user context information, task parameters, reply information, action point adoption information, and a message indicating successful execution of the action point.

[0098] In a specific application scenario, the memory system 47 includes a session data layer, the session data layer includes at least one target data layer, the target data layer includes at least one task data layer, the task data layer includes at least one model data layer and at least one behavior data layer, wherein the user context information is stored in the task data layer, the task parameters are stored in a model data layer of the task data layer, and the reply information, action point adoption information, and action point execution success message are stored in a behavior data layer of the task data layer.

[0099] It should be noted that for other corresponding descriptions of the functional units involved in the information processing device provided in the embodiment of the present application, please refer to Figures 1 to 11 The corresponding description in the method will not be repeated here.

[0100] Further, as Figures 1 to 11 The specific implementation of the method, the embodiment of the present application provides an information processing device, such as Figure 13 As shown, the device includes: The information sending module 51 can be used to send the user input information to the server in response to receiving the user input information, so that the server obtains the user context information corresponding to the user input information, and performs intent recognition on the user context information through the main agent to obtain the target task to be executed, wherein the target task includes at least one subtask; the main agent determines the sub-agent that executes the subtask and the task execution authority of the sub-agent based on the task content of the subtask and the description information of the sub-agent; the main agent sets a task identifier for the subtask and sends the task identifier, task content and task execution authority of the subtask to the sub-agent that executes the subtask; the sub-agent determines the task execution scope based on the task content of the subtask and the task execution authority, and executes the subtask within the task execution scope to obtain the task execution result; the main agent generates a reply message based on the task execution scope and task execution result of the sub-agent; The information display module 52 is configured to receive and display the reply information.

[0101] In a specific application scenario, the information display module 52 can be specifically used to display the reply information through at least one preset component, wherein the preset component includes at least one component among a graphic display component, a text flow display component and an action point display component.

[0102] In a specific application scenario, when the action point display component is included in the displayed preset components, the information sending module 51 can also be used to receive action point adoption information through the action point display component and send the action point adoption information to the server; the information display module 52 can also be used to receive and display a message indicating that the action point has been successfully executed.

[0103] In a specific application scenario, the information display module 52 can also be used to display the parameter collection component, wherein the parameter collection component displays the parameter information to be collected; the information sending module 51 can also be used to receive task parameters through the parameter collection component and send the task parameters to the server.

[0104] It should be noted that for other corresponding descriptions of the functional units involved in the information processing device provided in the embodiment of the present application, please refer to Figures 1 to 11 The corresponding description in the method will not be repeated here.

[0105] The embodiment of the present application also provides a computer device, which can be specifically a personal computer, a server, a network device, etc. The computer device includes a bus, a processor, a memory and a communication interface, and may also include an input and output interface and a display device. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store location information. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, the steps in each method embodiment are implemented.

[0106] Those skilled in the art will understand that the structure of the above-mentioned computer device is only a partial structure related to the solution of the present application and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components, or combine certain components, or have a different component arrangement.

[0107] In one embodiment, a computer-readable storage medium is provided. The computer-readable storage medium may be non-volatile or volatile, and stores a computer program thereon. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0108] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0109] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the various embodiments provided herein may be, but are not limited to, general-purpose processors, graphics processors, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, and the like.

[0110] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0111] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. An information processing method, characterized in that: The method comprises: Obtaining user context information, and performing intent recognition on the user context information through a master agent to obtain a target task to be executed, wherein the target task includes at least one subtask; The master agent determines the sub-agent to execute the sub-task and the task execution authority of the sub-agent based on the task content of the sub-task and the description information of the sub-agent, wherein the task execution authority includes full execution authority and partial execution authority; The master agent sets a task identifier for the subtask and sends the task identifier, task content and task execution authority of the subtask to the sub-agent that executes the subtask; The sub-agent determines the task execution scope based on the task content of the sub-task and the task execution authority, and executes the sub-task within the task execution scope to obtain a task execution result; wherein, when the sub-agent executes the sub-task, if it is determined that there are missing task parameters, the parameter information of the missing task parameters is sent to the main agent; the main agent calls a parameter collection tool based on the parameter information, obtains the missing task parameters through the parameter collection tool, and sends the task parameters to the sub-agent; the sub-agent executes the sub-task based on the task parameters to obtain a task execution result; The main agent generates reply information based on the task execution scope and task execution results of the sub-agent, and sends the reply information to the client.

2. The information processing method according to claim 1, wherein: The sub-agent determines the task execution scope based on the task content of the sub-task and the task execution authority, and executes the sub-task within the task execution scope to obtain the task execution result, including: The sub-agent, under the full execution authority, determines that the task execution scope is full execution based on the task content, executes the entire content of the sub-task to obtain a task execution result; and / or When the sub-agent determines that the task execution scope is partial execution based on the task content under the full execution authority, the sub-agent executes at least part of the sub-task to obtain a task execution result; and / or When the sub-agent determines that the task execution scope is not to be executed based on the task content under all the execution permissions, it generates execution refusal information as the task execution result; and / or The sub-agent, under the partial execution authority, determines that the task execution scope is partial execution based on the task content, and executes at least part of the content of the sub-task to obtain a task execution result; and / or, When the sub-agent determines that the task execution scope is not to be executed based on the task content under the partial execution authority, it generates execution refusal information as the task execution result.

3. The information processing method according to claim 1 or 2, characterized in that: The master agent generates reply information based on the task execution scope and task execution results of the sub-agent, including: The main agent generates a reply message based on the task execution result of the sub-agent when the task execution scope of the sub-agent is full execution; and / or When the task execution scope of the sub-agent is partially executed, the main agent processes the task execution result of the sub-agent and generates a reply message; and / or When the task execution scope of the sub-agent is partially executed, the main agent determines the task execution tool based on the task execution result of the sub-agent and the target task, and generates a reply message based on the task execution result of the sub-agent and the execution result of the task execution tool; and / or, When the task execution scope of the sub-agent is not to execute, the main agent determines a task execution tool based on the target task, and generates reply information based on the execution result of the task execution tool.

4. The information processing method according to claim 1, wherein: The master agent sends the reply information to the client, including: The main intelligent body calls at least one information display tool based on the reply information, and sends the reply information to the client through the information display tool, wherein the information display tool includes at least one tool among a graphic display tool, a text flow display tool and an action point display tool.

5. The information processing method according to claim 4, characterized in that When the called information display tool includes the action point display tool, the method further includes: In response to receiving the action point adoption information, performing action point processing based on the action point adoption information, and sending a message indicating that the action point has been successfully executed to the client.

6. The information processing method according to claim 1, wherein: The method further comprises: In response to an information update request, updated information is stored in a memory system, wherein the information covered by the information update request includes at least one of user context information, task parameters, reply information, action point adoption information, and a message indicating successful execution of the action point.

7. The information processing method according to claim 6, characterized in that: The memory system includes a session data layer, the session data layer includes at least one target data layer, the target data layer includes at least one task data layer, the task data layer includes at least one model data layer and at least one behavior data layer, wherein the user context information is stored in the task data layer, the task parameters are stored in a model data layer of the task data layer, and the reply information, action point adoption information, and action point execution success message are stored in a behavior data layer of the task data layer.

8. An information processing method, characterized in that: The method comprises: In response to receiving user input information, the user input information is sent to the server so that the server obtains user context information corresponding to the user input information, and performs intent recognition on the user context information through the main agent to obtain the target task to be executed, wherein the target task includes at least one subtask; the main agent determines the sub-agent that executes the subtask and the task execution authority of the sub-agent based on the task content of the subtask and the description information of the sub-agent, wherein the task execution authority includes full execution authority and partial execution authority; the main agent sets a task identifier for the subtask, and sends the task identifier, task content and task execution authority of the subtask to the sub-agent that executes the subtask. The sub-agent determines the task execution scope based on the task content of the sub-task and the task execution authority, and executes the sub-task within the task execution scope to obtain the task execution result; wherein, when the sub-agent executes the sub-task, if it is determined that there are missing task parameters, the parameter information of the missing task parameters is sent to the main agent; the main agent calls a parameter collection tool based on the parameter information, obtains the missing task parameters through the parameter collection tool, and sends the task parameters to the sub-agent; the sub-agent executes the sub-task based on the task parameters to obtain the task execution result; the main agent generates a reply message based on the task execution scope and task execution result of the sub-agent; The reply information is received and displayed.

9. The information processing method according to claim 8, characterized in that The receiving and displaying the reply information includes: The reply information is displayed through at least one preset component, wherein the preset component includes at least one component among a graphic display component, a text flow display component and an action point display component.

10. The information processing method according to claim 9, wherein: When the displayed preset components include the action point display component, the method further includes: receiving action point adoption information through the action point display component, and sending the action point adoption information to the server; Receive and display a message indicating that the action point was successfully executed.

11. The information processing method according to claim 8, wherein: The method further comprises: Display parameter collection component, wherein the parameter collection component displays parameter information of task parameters; The task parameters are received by the parameter collection component, and the task parameters are sent to the server.

12. An information processing device, characterized in that: The device comprises: A main agent, configured to obtain user context information, perform intent recognition on the user context information, and obtain a target task to be executed, wherein the target task includes at least one subtask; The master agent is further configured to determine a sub-agent to execute the sub-task and the task execution authority of the sub-agent based on the task content of the sub-task and the description information of the sub-agent, wherein the task execution authority includes full execution authority and partial execution authority; The master agent is further configured to set a task identifier for the subtask and send the task identifier, task content, and task execution authority of the subtask to the sub-agent that executes the subtask; The sub-agent is configured to determine a task execution scope based on the task content of the sub-task and the task execution authority, and execute the sub-task within the task execution scope to obtain a task execution result; wherein, when executing the sub-task, if the sub-agent determines that there are missing task parameters, the sub-agent sends parameter information of the missing task parameters to the main agent; the main agent calls a parameter collection tool based on the parameter information, obtains the missing task parameters through the parameter collection tool, and sends the task parameters to the sub-agent; the sub-agent executes the sub-task based on the task parameters to obtain a task execution result; The main agent is further used to generate reply information based on the task execution scope and task execution results of the sub-agent, and send the reply information to the client.

13. An information processing device, characterized in that: The device comprises: An information sending module is used to send the user input information to the server in response to receiving the user input information, so that the server obtains the user context information corresponding to the user input information, and performs intention recognition on the user context information through the main agent to obtain the target task to be executed, wherein the target task includes at least one subtask; the main agent determines the sub-agent that executes the subtask and the task execution authority of the sub-agent based on the task content of the subtask and the description information of the sub-agent, wherein the task execution authority includes full execution authority and partial execution authority; the main agent sets a task identifier for the subtask, and sends the task identifier, task content and task execution authority of the subtask to the sub-agent that executes the subtask. The sub-agent of the task; the sub-agent determines the task execution scope based on the task content of the sub-task and the task execution authority, and executes the sub-task within the task execution scope to obtain the task execution result; wherein, when the sub-agent executes the sub-task, if it is determined that there are missing task parameters, the parameter information of the missing task parameters is sent to the main agent; the main agent calls the parameter collection tool based on the parameter information, obtains the missing task parameters through the parameter collection tool, and sends the task parameters to the sub-agent; the sub-agent executes the sub-task based on the task parameters to obtain the task execution result; the main agent generates a reply message based on the task execution scope and task execution result of the sub-agent; The information display module is used to receive and display the reply information.

14. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 11 is implemented.

15. A computer device comprising a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 11 is implemented.

Images