Multi-agent based information processing method and device, and electronic device

Abstract

The present disclosure provides a multi-agent-based information processing method and device and electronic equipment, relating to the technical field of artificial intelligence. The specific implementation scheme is: receiving an information processing request, wherein the information processing request includes input information; inputting the input information to a first agent, determining one or more second agents from an agent set according to the input information by the first agent, and obtaining output information of the first agent; and obtaining response information corresponding to the input information according to the output information of the first agent and the second agent.

Description

Technical Field

[0001] This disclosure relates to the field of artificial intelligence technology, and in particular to an information processing method, apparatus and electronic device based on multi-agent systems. Background Technology

[0002] Currently, when using intelligent agents as smart customer service representatives to interact with users, traditional agents lack a deep understanding of multi-turn conversations and context, making it difficult to accurately grasp the user's true intentions. Furthermore, traditional agents cannot provide personalized responses based on the user's historical conversations or user profiles, resulting in a poor user experience. Summary of the Invention

[0003] This disclosure provides a method, apparatus, and electronic device for information processing based on multiple agents.

[0004] According to one aspect of this disclosure, a multi-agent-based information processing method is provided, comprising: receiving an information processing request, wherein the information processing request includes input information; inputting the input information to a first agent, wherein the first agent determines one or more second agents from a set of agents based on the input information, and obtains output information of the first agent; and obtaining response information corresponding to the input information based on the output information of the first agent and the second agents.

[0005] According to another aspect of this disclosure, a multi-agent-based information processing apparatus is provided, comprising: a receiving module for receiving an information processing request, wherein the information processing request includes input information; a first generating module for inputting the input information to a first agent, wherein the first agent determines one or more second agents from a set of agents based on the input information, and obtains output information of the first agent; and a second generating module for obtaining response information corresponding to the input information based on the output information of the first agent and the second agents.

[0006] According to another aspect of this disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the multi-agent-based information processing method described in one aspect of the above-described embodiment.

[0007] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided that stores computer instructions thereon, wherein the computer instructions are used to cause the computer to perform the multi-agent-based information processing method described in the above-mentioned embodiment.

[0008] According to another aspect of this disclosure, a computer program product is provided, including a computer program / instructions that, when executed by a processor, implement the multi-agent-based information processing method described in one aspect of the above-described embodiments.

[0009] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0010] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0011] Figure 1 A flowchart illustrating an information processing method based on multiple agents provided in this disclosure embodiment;

[0012] Figure 2 A flowchart illustrating another multi-agent-based information processing method provided in this embodiment of the present disclosure;

[0013] Figure 3 A flowchart illustrating another multi-agent-based information processing method provided in this embodiment of the present disclosure;

[0014] Figure 4 A flowchart illustrating another multi-agent-based information processing method provided in this embodiment of the present disclosure;

[0015] Figure 5 A flowchart illustrating another multi-agent-based information processing method provided in this embodiment of the present disclosure;

[0016] Figure 6 This is a schematic diagram of the structure of the multi-agent-based information processing method provided in the embodiments of this disclosure;

[0017] Figure 7 This is a schematic diagram of the structure of an information processing device based on multiple agents provided in an embodiment of the present disclosure;

[0018] Figure 8 This is a block diagram of an electronic device used to implement the multi-agent-based information processing method of the embodiments of this disclosure. Detailed Implementation

[0019] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0020] The following description, with reference to the accompanying drawings, outlines a multi-agent-based information processing method, apparatus, and electronic device according to embodiments of the present disclosure.

[0021] Artificial intelligence (AI) is the study of enabling computers to simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, and planning). It involves both hardware and software technologies. AI hardware technologies generally include computer vision, speech recognition, natural language processing, and related technologies such as deep learning, big data processing, and knowledge graphs.

[0022] An intelligent agent is a computer system or entity capable of acting autonomously, perceiving its environment, making decisions, and interacting with it. Intelligent agents are systems or machines created by humans that can perform tasks typically requiring human intelligence, such as visual recognition, language understanding, decision-making, and translation. Intelligent agents usually rely on large language models as their core decision-making and processing units, possessing the ability to think independently and invoke tools to gradually achieve given goals.

[0023] The information processing method based on multi-agent provided in this disclosure is applicable to multiple industry scenarios, including e-commerce platforms, financial institutions, medical consultation, and education tutoring. It can provide users with an interactive experience that is low in configuration cost, logically controllable, highly accurate, and personalized.

[0024] The multi-agent-based information processing method disclosed herein is applicable to online platforms requiring large-scale user services, customer-facing enterprise services, and other occasions requiring intelligent and highly automated customer service systems. Through the collaboration of multiple agents, this disclosure can better achieve business demand alignment, improve user conversion rates, customer satisfaction, and service efficiency, and reduce the rate of manual intervention.

[0025] Figure 1 This is a flowchart illustrating an information processing method based on multiple agents, provided in an embodiment of this disclosure.

[0026] like Figure 1 As shown, this multi-agent-based information processing method may include:

[0027] S101, Receive an information processing request, wherein the information processing request includes input information.

[0028] It should be noted that the executing entity of the multi-agent information processing method in this disclosure can be a hardware device with data processing capabilities and / or the necessary software to drive the hardware device. Optionally, the executing entity may include a server, a user terminal, and other intelligent devices. Optionally, the user terminal includes, but is not limited to, mobile phones, computers, and intelligent voice interaction devices. Optionally, the server includes, but is not limited to, a network server, an application server, or a server of a distributed system, or a server combined with blockchain, etc. This disclosure does not impose specific limitations.

[0029] In some implementations, information processing requests can be generated based on user input. That is, user input is obtained, and information processing requests are generated based on that input. Optionally, a query entered by the user can be used as the input.

[0030] Optionally, user input information can be obtained from the agent's data center.

[0031] S102, input information is input to the first intelligent agent, and the first intelligent agent determines one or more second intelligent agents from the set of intelligent agents based on the input information, and obtains the output information of the first intelligent agent.

[0032] In some implementations, input information is fed into a first intelligent agent, which processes the input information to obtain the user's intent. Based on the user's intent, the first intelligent agent plans the agents in the set and determines one or more second intelligent agents. The first intelligent agent can be a global planning agent.

[0033] Optionally, the input information can be categorized to determine the corresponding intent type, which is then used as the user intent. Alternatively, pre-set matching rules can be used to match the input information to determine the user intent. A pre-trained large model can also be used to identify the user intent based on the input information.

[0034] In some implementations, after receiving input information, the first agent can also generate agent planning status and standard operating procedure (SOP) process information, and use the input information, agent planning status, and SOP process information as output information so that the second agent can respond according to the output information.

[0035] Optionally, after receiving the input information, the first agent can determine whether the input information needs to be adjusted. If adjustment is needed, the adjusted input information is included as an item in the output information; otherwise, the input information is included as an item in the output information. Optionally, the determination of whether the input information needs adjustment can be based on the format and content of the input information.

[0036] Optionally, the set of intelligent agents includes, but is not limited to: speech generation intelligent agents, retrieval enhancement generation intelligent agents, proactive dialogue intelligent agents, dialogue-generated image intelligent agents, dialogue-generated video intelligent agents, image recognition intelligent agents, and information search intelligent agents.

[0037] Understandably, the speech generation agent is used to generate speech, and a large language model (LLM) with hundreds of billions to trillions of parameters can be used as the speech generation agent.

[0038] The retrieval-augmented generation agent is responsible for performing vector retrieval recall and generating responses, clarifications, or rejections based on input information within the data center (which stores semi-structured and structured data). The vector retrieval part employs an arbitrary retrieval-augmented generation (RAG) system. Responses, clarifications, and rejections can all be configured and defined using natural language, typically employing an LLM with tens of billions or fewer parameters.

[0039] The proactive dialogue agent is responsible for generating proactive dialogue scripts based on information such as the user's historical dialogues and user profile when the user is silent. It typically uses an LLM with tens to hundreds of billions of parameters.

[0040] A dialogue-based image agent refers to the process of using artificial intelligence technology to transform textual descriptions into images. This agent can understand the semantic information in the text and convert it into corresponding image content.

[0041] A dialogue-based video agent refers to an intelligent agent that uses artificial intelligence technology to transform descriptions into dynamic video content. This agent needs not only to understand the semantic information in the text, but also to understand the movement and interaction of objects in three-dimensional space in order to generate video information.

[0042] An image recognition intelligent agent is an intelligent system capable of autonomously perceiving image information, performing image analysis and recognition, and making decisions or performing related tasks accordingly. It utilizes advanced image recognition technologies, such as convolutional neural networks, to extract features, classify, and recognize input images.

[0043] An information search intelligent agent is an intelligent system capable of autonomously searching, filtering, and integrating information on the Internet, and providing corresponding information services according to user needs. It utilizes advanced search engine technology, natural language processing technology, and data mining technology to achieve rapid searching and accurate matching of online information.

[0044] For example, if the input information is "Introduce item A", the first agent identifies the script generation agent as the second agent by recognizing the input information, and generates an introduction to item A. If the input information is "Introduce item A in style X", the first agent identifies the script generation agent and the retrieval enhancement generation agent as the second agent by recognizing the input information, and generates an introduction to item A according to style X.

[0045] S103, based on the output information of the first intelligent agent and the second intelligent agent, obtain the response information corresponding to the input information.

[0046] In some implementations, prompts for a second agent can be generated based on the output information of the first agent. The second agent then responds to the input information based on the prompts, obtaining the corresponding response information. The response information can be obtained by inputting the prompts into the second agent.

[0047] Optionally, auxiliary information of the second agent can be obtained, and the auxiliary information and output information can be combined to determine the prompt words for the second agent. That is, the auxiliary information and output information can be used as the prompt words for the second agent.

[0048] Optionally, auxiliary information for the second agent can be obtained from the agent's configuration center. This auxiliary information includes, but is not limited to, dialogue strategies, user profiles, and historical dialogues.

[0049] In some implementations, when there are multiple second agents, the upstream and downstream relationships between the second agents are determined, and the output information of the upstream agent is used as the prompt word for the downstream agent, so that the second agent can determine the response information based on the input information.

[0050] According to the multi-agent information processing method provided in this disclosure, an information processing request containing input information is obtained, and a first agent determines one or more required second agents and the output information of the first agent based on the input information. Further, the second agents determine the response information corresponding to the input information based on the output information, thereby improving the efficiency and accuracy of agents responding to input information and increasing problem-solving efficiency and recall. Responding based on multiple agents effectively increases the number of interactions between the user and the agents and the dialogue effect.

[0051] Figure 2This is a flowchart illustrating an information processing method based on multiple agents, provided in an embodiment of this disclosure.

[0052] like Figure 2 As shown, this multi-agent-based information processing method may include:

[0053] S201, Receive information processing request, wherein the information processing request includes input information.

[0054] The details of step S201 can be found in the above embodiments and will not be repeated here.

[0055] S202, input information is input to the first intelligent agent, and the first intelligent agent determines one or more second intelligent agents from the set of intelligent agents based on the input information, and obtains the output information of the first intelligent agent.

[0056] In some implementations, the user's intent can be determined based on the input information, and one or more agents can be selected based on the user's intent to ensure personalized and differentiated agent responses, thereby providing response information with emotional value.

[0057] In some implementations, the user's intent can be determined by recognizing the input information based on the auxiliary information of the first intelligent agent. Optionally, user profiles and historical behavior data can be obtained from the agent's data center as auxiliary information for the first intelligent agent.

[0058] Furthermore, the first intelligent agent performs intent recognition on the input information based on auxiliary information to obtain an intent information set, which includes at least one user intent, and determines one or more second intelligent agents from the intelligent agent set based on the user intent.

[0059] In some implementations, before determining the second agent, a set of agents can be determined based on the current business scenario. One or more second agents can then be selected from this set, thereby filtering out the set of agents that best matches the business requirements. This helps ensure that the selected second agent can accurately meet the business needs.

[0060] Optionally, the current target business scenario can be determined based on the input information, and the intelligent agents associated with the target business scenario can be obtained. Then, based on the intelligent agents associated with the target business scenario, the set of intelligent agents corresponding to the target business scenario can be determined. Optionally, the association between business scenarios and intelligent agents can be established in advance, and after determining the target business scenario, the intelligent agents associated with the target business scenario can be determined by querying this association.

[0061] In some implementations, sets of intelligent agents for different industries can be determined based on the industry to which the intelligent agent belongs and the industry of the user, in order to provide automated and intelligent interaction and improve the service efficiency of different industries. Optionally, the set of intelligent agents can be obtained by acquiring the industry to which the intelligent service belongs and the industry of the user corresponding to the input information, and based on the industry to which the intelligent service belongs and the industry of the user corresponding to the input information.

[0062] Optionally, the set of intelligent agents includes at least one of the following intelligent agents: a speech generation intelligent agent, a retrieval enhancement generation intelligent agent, an active dialogue intelligent agent, a dialogue-generated image intelligent agent, a dialogue-generated video intelligent agent, an image recognition intelligent agent, and an information search intelligent agent.

[0063] S203, determine the auxiliary information required by the second agent, the auxiliary information including at least one of the pre-configured dialogue policy, user profile and historical dialogue.

[0064] In this embodiment of the disclosure, there is one second intelligent agent. That is, the second intelligent agent can be a speech generation intelligent agent, a retrieval enhancement generation intelligent agent, or an active dialogue intelligent agent.

[0065] In some implementations, in order to improve the execution efficiency and accuracy of the second intelligent agent and optimize the user's interaction experience with the second intelligent agent, the auxiliary information required by the second intelligent agent can be determined, and prompt words can be determined based on the auxiliary information.

[0066] Optionally, the target service corresponding to the second intelligent agent can be obtained, and based on the target service, the configuration information of the target service corresponding to the second intelligent agent can be obtained, and the auxiliary information required by the second intelligent agent can be obtained from the configuration information. For example, the configuration information can be determined from the configuration center, and the auxiliary information can be determined from the configuration information. The auxiliary information includes at least one of the following: pre-configured dialogue policy, user profile, and historical dialogue.

[0067] Optionally, the configuration center contains a list of configuration information, and the configuration information in the list corresponds to the target service. The matching information list can be queried based on the target service to determine the configuration information of the target service, and auxiliary information can be determined from the configuration information.

[0068] S204, Based on the auxiliary information and the output information of the first agent, generate the prompt words for the second agent.

[0069] Optionally, a prompt word template can be preset, and auxiliary information and the output information of the first agent can be added to the prompt word template to obtain the prompt words of the second agent.

[0070] S205, obtain the response information corresponding to the input information based on the prompt words of the second agent.

[0071] In some implementations, the prompt word is input into a second agent, which then generates a response based on the prompt word. Optionally, the response can be generated according to a predefined output format.

[0072] According to the multi-agent information processing method provided in this disclosure, an information processing request containing input information is obtained, and a first agent identifies the user intent in the input information to determine one or more second agents and the output information of the first agent based on the user intent. Further, auxiliary information for the second agents is determined, and based on the auxiliary information and the output information, prompt words for the second agents are determined. The second agents then determine the response information corresponding to the input information based on the prompt words, thereby improving the efficiency and accuracy of agents responding to input information and increasing problem-solving efficiency and recall. Responding based on multiple agents effectively increases the number of interactions between the user and the agents and the dialogue effect.

[0073] Figure 3 This is a flowchart illustrating an information processing method based on multiple agents, provided in an embodiment of this disclosure.

[0074] like Figure 3 As shown, this multi-agent-based information processing method may include:

[0075] S301, Receive information processing request, wherein the information processing request includes input information.

[0076] S302, input information is input to the first intelligent agent, and the first intelligent agent determines one or more second intelligent agents from the set of intelligent agents based on the input information, and obtains the output information of the first intelligent agent.

[0077] The details of steps S301-S302 can be found in the above embodiments and will not be repeated here.

[0078] S303, based on the output information of the first agent, determine the execution order of multiple second agents and the dependency relationship between the second agents.

[0079] In some implementations, when it is determined that there are multiple second agents, the execution order of the multiple second agents contained in the output information of the first agent can be determined by parsing the output information of the first agent and the dependency relationship between the second agents.

[0080] For example, the execution order of multiple second agents and the dependencies between second agents can be determined based on the agent planning information and SOP process information in the output information.

[0081] S304: In accordance with the execution order, multiple second agents are invoked, and the input information of the currently invoked second agent is determined based on the dependency relationship.

[0082] In some implementations, the upstream and downstream relationships between second agents can be determined based on dependencies, and the output of the upstream agent can be used as the input of the downstream agent to optimize the task flow and improve the system's collaborative efficiency. In other words, the upstream agent that the currently invoked second agent depends on can be determined based on dependencies; the dependent upstream agent can be at least one of multiple second agents.

[0083] Furthermore, the output information of the upstream agent on which the agent depends is obtained and used as the input information for the currently invoked second agent. For example, the second agent includes a script generation agent and a retrieval enhancement generation agent. The currently invoked second agent is the script generation agent, and its upstream agent is the retrieval enhancement generation agent. That is, retrieval enhancement is performed first, and then the output information of the retrieval enhancement generation agent is used as the input information for the script generation agent.

[0084] S305, Based on the auxiliary information of the second agent and the corresponding input information, generate the prompt word for the currently invoked second agent.

[0085] Optionally, a prompt word template can be preset, and the auxiliary information and corresponding output information of the second agent can be added to the prompt word template to obtain the prompt word of the currently invoked second agent.

[0086] S306, obtain the second output information corresponding to the currently invoked second agent according to the prompt word of the second agent, until the sequential execution of multiple second agents is completed, and obtain the response information corresponding to the input information.

[0087] In some implementations, a prompt word is input into the currently invoked second agent, which then generates a second output message based on the prompt word. This second output message is used as input to the downstream agent corresponding to the currently invoked second agent, which in turn generates a corresponding prompt word based on the input message. Furthermore, the downstream agent can generate output information based on the prompt word, and this process continues until multiple second agents are invoked according to the execution order. The output message of the last invoked second agent is then used as the response message corresponding to the input message.

[0088] According to the multi-agent information processing method provided in this disclosure, an information processing request containing input information is obtained, and a first agent determines one or more required second agents and their output information based on the input information. Further, the execution order and dependencies of the second agents are determined, and the input information of the currently invoked second agent is determined based on the dependencies. A prompt word for the second agent is determined based on the input information, and the second output information corresponding to the currently invoked second agent is determined based on the prompt word. The invocation of the second agents is then completed according to the execution order to obtain the response information corresponding to the input information. This improves the efficiency and accuracy of the agents responding to input information, and enhances problem-solving efficiency and recall. Responding based on multiple agents effectively increases the number of interactions between the user and the agents and improves the dialogue effect.

[0089] Figure 4 This is a flowchart illustrating an information processing method based on multiple agents, provided in an embodiment of this disclosure.

[0090] like Figure 4 As shown, this multi-agent-based information processing method may include:

[0091] S401, Receive information processing request, wherein the information processing request includes input information.

[0092] S402, input information is input to the first intelligent agent, and the first intelligent agent determines one or more second intelligent agents from the set of intelligent agents based on the input information, and obtains the output information of the first intelligent agent.

[0093] The details of steps S401-S402 can be found in the above embodiments and will not be repeated here.

[0094] S403, for each second agent, determine the auxiliary information required by the second agent.

[0095] In some implementations, the target service for each second agent is determined, and the corresponding configuration information is also determined, in order to obtain the auxiliary information required by the second agent from the configuration information. For example, the configuration information can be determined from the configuration center based on the target service, and the auxiliary information can be determined from the configuration information.

[0096] S404, determine the associated output information that matches the second agent from the output information.

[0097] In some implementations, the output information can be filtered to determine the associated output information that matches the second agent. Alternatively, the associated output information that matches the second agent can be filtered from the output information based on the second agent's characteristic information, such as its features, requirements, and functions.

[0098] S405, Based on the auxiliary information of the second agent and the matched associated output information, generate prompt words for each second agent.

[0099] Optionally, for each second agent, the prompt words can be determined based on a pre-set prompt word template. The prompt words for each second agent can be obtained by adding the auxiliary information of the second agent and the matched associated output information to the prompt word template.

[0100] S406, obtain the first output information corresponding to the associated output information based on the prompt word of the second agent.

[0101] S407, based on the first output information of each second agent, generate response information corresponding to the input information.

[0102] In some implementations, a prompt word is input into a second agent, which then generates a first output message corresponding to the input information based on the prompt word. Furthermore, by concatenating the first output messages corresponding to each second agent, the response message corresponding to the input information can be obtained.

[0103] Optionally, the first output information can also be used as response information, that is, one input information can correspond to multiple response information.

[0104] According to the multi-agent information processing method provided in this disclosure, an information processing request containing input information is obtained, and a first agent determines one or more required second agents and their output information based on the input information. Further, auxiliary information required by each second agent is determined, and associated output information matching the second agent is determined from the output information to identify the prompt words for the second agents. The second agents then determine the response information corresponding to the input information based on the prompt words, thereby improving the efficiency and accuracy of agents responding to input information and increasing problem-solving efficiency and recall. Responding based on multiple agents effectively increases the number of interactions between the user and the agents and the dialogue effect.

[0105] Figure 5 This is a flowchart illustrating an information processing method based on multiple agents, provided in an embodiment of this disclosure.

[0106] like Figure 5 As shown, this multi-agent-based information processing method may include:

[0107] S501, Receive information processing request, wherein the information processing request includes input information.

[0108] S502, input information is input to the first intelligent agent, and the first intelligent agent determines one or more second intelligent agents from the set of intelligent agents based on the input information, and obtains the output information of the first intelligent agent.

[0109] S503: Based on the output information of the first intelligent agent and the second intelligent agent, obtain the response information corresponding to the input information.

[0110] The relevant content of steps S501-S503 can be found in the above embodiments, and will not be repeated here.

[0111] S504, activate at least one third agent for dialogue monitoring and acquire the dialogue content during the dialogue process.

[0112] Optionally, the third intelligent agent includes a dialogue analysis agent and a dialogue quality inspection agent. By activating the third intelligent agent and acquiring the dialogue content during the conversation between the user and the second intelligent agent, the third intelligent agent can analyze the dialogue content to promptly identify problems encountered by the user in the dialogue, and quickly respond to and resolve them, thereby improving user satisfaction. Furthermore, the agent can be optimized based on the analysis results, thereby enhancing its learning ability.

[0113] S505 uses a third-party intelligent agent to intelligently analyze the dialogue content.

[0114] In some implementations, if the third agent is a dialogue analysis agent, it can extract information from the dialogue content to obtain key information, which is then used to optimize subsequent dialogues. The third agent can also clarify and reflect on the dialogue content to obtain dialogue reflection results. These reflection results can then be used to optimize the agent, thereby improving its learning ability and enhancing its adaptability.

[0115] Optionally, by identifying key information and dialogue reflection results, the analysis results of the dialogue analysis agent are obtained, and the analysis results of the dialogue analysis agent are subjected to structured processing to obtain first structured data. Further, the first structured data is determined to be shared data and stored in a data center. The first structured data can be accessed by at least one of the first and second agents.

[0116] In other words, the first and second intelligent agents can optimize themselves by accessing structured data in the data center, thereby improving their learning ability and enhancing their adaptability.

[0117] In some implementations, if the third agent is a dialogue quality control agent, it can perform illusion detection and compliance monitoring on the dialogue content, obtain quality detection results, and fine-tune the agent based on the quality detection results, thereby correcting the agent's errors and illusions and improving the accuracy of the agent's responses.

[0118] Optionally, based on the quality detection results, abnormal dialogue content can be identified and collected as sample data for fine-tuning the agent. The sample data is used to fine-tune the relevant agent to obtain a fine-tuned agent, and the fine-tuned agent is used for the next dialogue.

[0119] According to the multi-agent information processing method provided in this disclosure, an information processing request containing input information is obtained, and a first agent determines one or more required second agents and the output information of the first agent based on the input information. Further, the second agents determine the response information corresponding to the input information based on the output information, thereby improving the efficiency and accuracy of the agents' responses to the input information. Responses based on multiple agents effectively increase the number of interactions between the user and the agents and the dialogue effect. By obtaining the dialogue content and using a third agent to analyze the dialogue content, problems encountered by the user in the dialogue can be identified in a timely manner and quickly responded to and resolved, thereby improving user satisfaction. Furthermore, the agents can be optimized based on the analysis results, thereby improving the learning ability of the agents.

[0120] The following example illustrates the information processing method based on multi-agent systems provided in this disclosure, using a pre-sales scenario in the marketing field as an example for customer acquisition and lead collection:

[0121] Among them, the first intelligent agent is the global planning intelligent agent, the second intelligent agent is the speech generation intelligent agent, the retrieval enhancement generation intelligent agent, and the proactive dialogue intelligent agent, and the third intelligent agent is the dialogue analysis intelligent agent and the dialogue quality inspection intelligent agent.

[0122] By inputting the information into the global planning agent, the second agent can be identified as the script generation agent, the retrieval enhancement generation agent, and the proactive dialogue agent. The planning strategy, ultimate goal, and secondary goals are then generated as the output information of the first agent. Specifically, the planning strategy is a strategic planner for the marketing scenario, the ultimate goal is customer acquisition, and the secondary goal is to obtain customer lead information.

[0123] By determining the execution order of the second intelligent agent as the script generation intelligent agent, the retrieval enhancement generation intelligent agent, and the proactive dialogue intelligent agent, further, based on the auxiliary information of the script generation intelligent agent, the prompt words for the script generation intelligent agent are determined: a pre-sales service role in marketing scenarios, a specific description of obtaining customer lead information, and providing a high-quality customer service dialogue experience. By inputting the prompt words into the script generation intelligent agent, its corresponding second output information is obtained, and then based on the second output information, prompt words for the retrieval enhancement generation intelligent agent are generated: for answering questions in sales scenarios. By inputting the prompt words into the retrieval enhancement generation intelligent agent, its corresponding second output information is obtained, and then based on the second output information, prompt words for the proactive dialogue intelligent agent are generated: a personal assistant in private domain marketing scenarios, combining historical conversations to re-establish contact with users.

[0124] Furthermore, by acquiring the dialogue content during the dialogue process and using a dialogue quality inspection agent and a dialogue analysis agent, the dialogue content is analyzed in order to optimize the agents based on the analysis results.

[0125] like Figure 6 The diagram shown is a structural schematic of a multi-agent information processing method. Figure 6 It includes a data center, a configuration center, and multiple intelligent agents: a global planning agent, a dialogue generation agent, a retrieval enhancement generation agent, a proactive dialogue agent, a dialogue analysis agent, and a dialogue quality inspection agent.

[0126] The data center stores user data in structured and semi-structured formats, such as historical dialogues and dialogue content analysis results; the configuration center stores auxiliary information for intelligent agents, such as dialogue strategies, triggering strategies, user profiles, business descriptions, and business SOPs.

[0127] The global planning agent generates the planning logic and execution list of the second agent as output information based on the user's input information, and calls one or more second agents. Taking the second agent as the dialogue generation agent as an example, it obtains dialogue strategy and user profile as auxiliary information from the configuration center, and generates prompt words based on the auxiliary information and the output information of the global planning agent, so as to determine the response information corresponding to the input information according to the prompt words.

[0128] Taking the second agent as the retrieval-augmented generation agent as an example, the dialogue strategy and the retrieval-augmented generation (RAG) system are obtained from the configuration center as auxiliary information. Based on the auxiliary information and the output information of the global planning agent, prompt words are generated to determine the response information corresponding to the input information according to the prompt words.

[0129] Taking the second agent as the active dialogue agent as an example, the dialogue strategy and trigger strategy are obtained from the configuration center as auxiliary information, and prompt words are generated based on the auxiliary information and the output information of the global planning agent, so as to determine the response information corresponding to the input information according to the prompt words.

[0130] In agent-based dialogue, at least one agent—either a dialogue analysis agent or a dialogue quality control agent—can be used to monitor and analyze the dialogue content. Taking the dialogue analysis agent as an example, it extracts key information from the dialogue content, which is then used to optimize subsequent dialogue. It can also clarify and reflect on the dialogue content to obtain dialogue reflection results.

[0131] Taking a dialogue quality inspection agent as an example, the agent obtains quality inspection results by performing illusion monitoring and compliance monitoring on the dialogue content, and then fine-tunes the training of the relevant agent based on the quality inspection results.

[0132] Corresponding to the multi-agent-based information processing methods provided in the above embodiments, one embodiment of this disclosure also provides a multi-agent-based information processing device. Since the multi-agent-based information processing device provided in this disclosure corresponds to the multi-agent-based information processing methods provided in the above embodiments, the implementation methods of the above-mentioned multi-agent-based information processing methods are also applicable to the multi-agent-based information processing device provided in this disclosure, and will not be described in detail in the following embodiments.

[0133] Figure 7 This is a schematic diagram of the structure of an information processing device based on multiple agents provided in an embodiment of this disclosure.

[0134] like Figure 7 As shown, the information processing device 700 based on multi-agent in this embodiment of the present disclosure includes a receiving module 701, a first generating module 702, and a second generating module 703.

[0135] The receiving module 701 is used to receive an information processing request, wherein the information processing request includes input information;

[0136] The first generation module 702 is used to input the input information to the first intelligent agent, and the first intelligent agent determines one or more second intelligent agents from the set of intelligent agents based on the input information, and obtains the output information of the first intelligent agent;

[0137] The second generation module 703 is used to obtain the response information corresponding to the input information based on the output information of the first intelligent agent and the second intelligent agent.

[0138] In one embodiment of this disclosure, the second generation module 703 is further configured to: determine the auxiliary information required by the second intelligent agent, the auxiliary information including at least one of a pre-configured dialogue strategy, a user profile, and historical dialogues; generate prompt words for the second intelligent agent based on the auxiliary information and the output information of the first intelligent agent; and obtain response information corresponding to the input information based on the prompt words of the second intelligent agent.

[0139] In one embodiment of this disclosure, the second generation module 703 is further configured to: determine the execution order of a plurality of second agents and the dependency relationship between the second agents based on the output information of the first agent; call the plurality of second agents according to the execution order, and determine the input information of the currently called second agent based on the dependency relationship; generate a prompt word for the currently called second agent based on the auxiliary information of the second agent and the corresponding input information; obtain the second output information corresponding to the currently called second agent based on the prompt word of the second agent, until the sequential execution of the plurality of second agents is completed, and obtain the response information corresponding to the input information.

[0140] In one embodiment of this disclosure, the second generation module 703 is further configured to: determine, based on the dependency relationship, the upstream intelligent agent on which the currently invoked second intelligent agent depends, wherein the upstream intelligent agent is at least one of the plurality of second intelligent agents; obtain the output information of the upstream intelligent agent on which it depends, and determine it as the input information of the currently invoked second intelligent agent.

[0141] In one embodiment of this disclosure, the second generation module 703 is further configured to: determine the auxiliary information required by each second agent; determine the associated output information matching the second agent from the output information; generate a prompt word for each second agent based on the auxiliary information of the second agent and the matched associated output information; obtain the first output information corresponding to the associated output information based on the prompt word of the second agent; and generate response information corresponding to the input information based on the first output information of each second agent.

[0142] In one embodiment of this disclosure, the apparatus further includes: an acquisition module, configured to activate at least one third intelligent agent for dialogue monitoring and acquire dialogue content during the dialogue process; and an analysis module, configured to perform intelligent analysis of the dialogue content through the third intelligent agent.

[0143] In one embodiment of this disclosure, the analysis module is further configured to: extract information from the dialogue content through the third intelligent agent to obtain key information, the key information being used to optimize subsequent dialogue; or, clarify and reflect on the dialogue content through the third intelligent agent to obtain dialogue reflection results.

[0144] In one embodiment of this disclosure, the analysis module is further configured to: determine the key information and the dialogue reflection result as the analysis result of the dialogue analysis agent; perform structured processing on the analysis result of the dialogue analysis agent to obtain first structured data; determine that the first structured data is shared data and store it in a data center, wherein the first structured data can be invoked by at least one of the first agent and the second agent.

[0145] In one embodiment of this disclosure, the analysis module is further configured to: perform illusion detection and compliance monitoring on the dialogue content through the third intelligent agent, and obtain quality detection results.

[0146] In one embodiment of this disclosure, the analysis module is further configured to: determine abnormal dialogue content based on the quality detection result, and collect the abnormal dialogue content as sample data for fine-tuning the agent, wherein the sample data is used to fine-tune the relevant agent.

[0147] In one embodiment of this disclosure, the second generation module 703 is further configured to: obtain the target service corresponding to the second intelligent agent; obtain the configuration information of the target service corresponding to the second intelligent agent, and obtain the auxiliary information required by the second intelligent agent from the configuration information.

[0148] In one embodiment of this disclosure, the first generation module 702 is further configured to: acquire user profiles and historical behavior data as auxiliary information for the first intelligent agent; have the first intelligent agent perform intent recognition on the input information based on the auxiliary information to acquire an intent information set, wherein the intent information set includes at least one user intent; and determine the one or more second intelligent agents from the intelligent agent set according to the user intent.

[0149] In one embodiment of this disclosure, the first generation module 702 is further configured to: determine the current target business scenario based on the input information; obtain the intelligent agent associated with the target business scenario; and determine the intelligent agent set corresponding to the target business scenario based on the intelligent agent associated with the target business scenario.

[0150] In one embodiment of this disclosure, the first generation module 702 is further configured to: obtain the industry to which the intelligent service belongs and the industry of the user corresponding to the input information; and obtain a set of intelligent agents based on the industry to which the intelligent service belongs and the industry of the user corresponding to the input information.

[0151] In one embodiment of this disclosure, the set of intelligent agents includes at least one of the following intelligent agents: a speech generation intelligent agent, a retrieval enhancement generation intelligent agent, an active dialogue intelligent agent, a dialogue-generated image intelligent agent, a dialogue-generated video intelligent agent, an image recognition intelligent agent, and an information search intelligent agent.

[0152] According to the multi-agent information processing apparatus provided in this disclosure, an information processing request containing input information is acquired, and a first agent determines one or more required second agents and the output information of the first agent based on the input information. Further, the second agents determine the response information corresponding to the input information based on the output information, thereby improving the efficiency and accuracy of agents responding to input information, and increasing problem-solving efficiency and recall. Responding based on multiple agents effectively increases the number of interactions between the user and the agents and the dialogue effect.

[0153] The acquisition, storage, and application of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0154] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0155] Figure 8 A schematic block diagram of an example electronic device 800 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0156] like Figure 8As shown, device 800 includes a computing unit 801, which can perform various appropriate actions and processes based on computer programs / instructions stored in read-only memory (ROM) 802 or loaded from storage unit 806 into random access memory (RAM) 803. RAM 803 may also store various programs and data required for the operation of device 800. The computing unit 801, ROM 802, and RAM 803 are interconnected via bus 804. Input / output (I / O) interface 805 is also connected to bus 804.

[0157] Multiple components in device 800 are connected to I / O interface 805, including: input units 806 such as keyboard, mouse, etc.; output units 807 such as various types of displays, speakers, etc.; storage units 808 such as disks, optical disks, etc.; and communication units 809 such as network cards, modems, wireless transceivers, etc. Communication unit 809 allows device 800 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0158] The computing unit 801 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 801 performs the various methods and processes described above, such as multi-agent-based information processing methods. For example, in some embodiments, the multi-agent-based information processing method can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 806. In some embodiments, part or all of the computer program / instructions can be loaded and / or installed on device 800 via ROM 802 and / or communication unit 809. When the computer program / instructions are loaded into RAM 803 and executed by the computing unit 801, one or more steps of the multi-agent-based information processing method described above can be performed. Alternatively, in other embodiments, computing unit 801 may be configured to perform a multi-agent-based information processing method by any other suitable means (e.g., by means of firmware).

[0159] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include: implementations in one or more computer programs / instructions that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transferring data and instructions to the storage system, the at least one input device, and the at least one output device.

[0160] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0161] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0162] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0163] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), the Internet, and blockchain networks.

[0164] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. The client-server relationship is created by computer programs / instructions running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0165] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in the disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this document does not impose any restrictions.

[0166] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A multi-agent-based information processing method, wherein, The method includes: Receive an information processing request, wherein the information processing request includes input information; The input information is input to the first intelligent agent, which then determines one or more second intelligent agents from the set of intelligent agents based on the input information, and obtains the output information of the first intelligent agent. Based on the output information of the first intelligent agent and the second intelligent agent, obtain the response information corresponding to the input information; The step of obtaining the response information corresponding to the input information based on the output information of the first agent and the second agent includes: The configuration center determines the auxiliary information required by the second agent, which includes pre-configured dialogue policies, user profiles, and historical dialogues. Based on the auxiliary information and the output information of the first agent, the prompt words for the second agent are generated; The prompt word is input to the second intelligent agent, which then generates the response information corresponding to the input information based on the prompt word.

2. The method according to claim 1, wherein, When there are multiple second intelligent agents, the step of obtaining the response information corresponding to the input information based on the output information of the first intelligent agent and the second intelligent agents includes: Based on the output information of the first intelligent agent, the execution order of multiple second intelligent agents and the dependencies between the second intelligent agents are determined; According to the execution order, the plurality of second agents are invoked, and the input information of the currently invoked second agent is determined according to the dependency relationship; Based on the auxiliary information of the second agent and the corresponding input information, generate the prompt word for the currently invoked second agent; Based on the prompt words of the second agent, obtain the second output information corresponding to the currently invoked second agent, until the sequential execution of multiple second agents is completed, and obtain the response information corresponding to the input information.

3. The method according to claim 2, wherein, Determining the input information of the currently invoked second agent based on the dependency relationship includes: Based on the dependency relationship, determine the upstream agent that the currently invoked second agent depends on, wherein the upstream agent depends on is at least one of the plurality of second agents; Obtain the output information of the upstream agent on which it depends, and determine it as the input information of the second agent currently being invoked.

4. The method according to claim 1, wherein, When there are multiple second intelligent agents, the step of obtaining the response information corresponding to the input information based on the output information of the first intelligent agent and the second intelligent agents includes: For each second agent, determine the auxiliary information required by the second agent; Determine the associated output information that matches the second agent from the output information; Based on the auxiliary information of the second agent and the matched associated output information, generate prompt words for each second agent; Obtain the first output information corresponding to the associated output information based on the prompt words of the second intelligent agent; Based on the first output information of each of the second intelligent agents, response information corresponding to the input information is generated.

5. The method according to any one of claims 1-4, wherein, The method further includes: Launch at least one third agent for dialogue monitoring and acquire the dialogue content during the dialogue process; The third intelligent agent performs intelligent analysis on the dialogue content.

6. The method according to claim 5, wherein, The third intelligent agent is a dialogue analysis intelligent agent, and the intelligent analysis of the dialogue content by the third intelligent agent includes: The third intelligent agent extracts information from the dialogue content to obtain key information, which is used to optimize subsequent dialogue; or... The third intelligent agent clarifies and reflects on the dialogue content to obtain a dialogue reflection result.

7. The method according to claim 6, wherein, The method further includes: The key information and the dialogue reflection results are determined as the analysis results of the dialogue analysis agent; The analysis results of the dialogue analysis agent are processed in a structured manner to obtain the first structured data; The first structured data is determined to be shared data and stored in the data center. The first structured data can be accessed by at least one of the first and second intelligent agents.

8. The method according to any one of claims 5, wherein, The third intelligent agent is a dialogue quality inspection intelligent agent. The intelligent analysis of the dialogue content by the third intelligent agent includes: The third intelligent agent performs illusion detection and compliance monitoring on the dialogue content to obtain quality inspection results.

9. The method according to claim 8, wherein, After obtaining the quality inspection results, the process also includes: Based on the quality detection results, abnormal dialogue content is identified and collected as sample data for fine-tuning the agent. The sample data is used to fine-tune the relevant agent.

10. The method according to any one of claims 1-4, wherein, The auxiliary information required to determine the second agent includes: Obtain the target service corresponding to the second intelligent agent; Obtain the configuration information of the target service corresponding to the second intelligent agent, and obtain the auxiliary information required by the second intelligent agent from the configuration information.

11. The method according to claim 1, wherein, The step of determining one or more second agents from the set of agents by the first agent based on the input information includes: Acquire user profiles and historical behavior data as auxiliary information for the first intelligent agent; The first intelligent agent performs intent recognition on the input information based on the auxiliary information to obtain an intent information set, wherein the intent information set includes at least one user intent; The one or more second agents are determined from the set of agents based on the user intent.

12. The method according to claim 1, wherein, Before the first agent determines one or more second agents from the set of agents based on the input information, the method further includes: The current target business scenario is determined based on the input information; Obtain the intelligent agent associated with the target business scenario; Based on the intelligent agents associated with the target business scenario, determine the set of intelligent agents corresponding to the target business scenario.

13. The method according to claim 1, wherein, Before the first agent determines one or more second agents from the set of agents based on the input information, the method further includes: Obtain the industry to which the intelligent service belongs and the industry of the user corresponding to the input information; A set of intelligent agents is obtained based on the industry to which the intelligent service belongs and the industry of the user corresponding to the input information.

14. The method according to any one of claims 1-4 and 11-13, wherein the set of intelligent agents includes at least one of the following intelligent agents: a speech generation intelligent agent, a retrieval enhancement generation intelligent agent, an active dialogue intelligent agent, a dialogue-generated image intelligent agent, a dialogue-generated video intelligent agent, an image recognition intelligent agent, and an information search intelligent agent.

15. An information processing device based on multi-agent systems, wherein, The device includes: A receiving module is used to receive an information processing request, wherein the information processing request includes input information; The first generation module is used to input the input information into the first intelligent agent, and the first intelligent agent determines one or more second intelligent agents from the set of intelligent agents based on the input information, and obtains the output information of the first intelligent agent; The second generation module is used to obtain the response information corresponding to the input information based on the output information of the first intelligent agent and the second intelligent agent; The second generation module is further configured to: The configuration center determines the auxiliary information required by the second agent, which includes pre-configured dialogue policies, user profiles, and historical dialogues. Based on the auxiliary information and the output information of the first agent, the prompt words for the second agent are generated; The prompt word is input to the second intelligent agent, which then generates the response information corresponding to the input information based on the prompt word.

16. The apparatus according to claim 15, wherein, The second generation module is further configured to: Based on the output information of the first intelligent agent, the execution order of multiple second intelligent agents and the dependencies between the second intelligent agents are determined; According to the execution order, the plurality of second agents are invoked, and the input information of the currently invoked second agent is determined according to the dependency relationship; Based on the auxiliary information of the second agent and the corresponding input information, generate the prompt word for the currently invoked second agent; Based on the prompt words of the second agent, obtain the second output information corresponding to the currently invoked second agent, until the sequential execution of multiple second agents is completed, and obtain the response information corresponding to the input information.

17. The apparatus according to claim 16, wherein, The second generation module is further configured to: Based on the dependency relationship, determine the upstream agent that the currently invoked second agent depends on, wherein the upstream agent depends on is at least one of the plurality of second agents; Obtain the output information of the upstream agent on which it depends, and determine it as the input information of the second agent currently being invoked.

18. The apparatus according to claim 15, wherein, The second generation module is further configured to: For each second agent, determine the auxiliary information required by the second agent; Determine the associated output information that matches the second agent from the output information; Based on the auxiliary information of the second agent and the matched associated output information, generate prompt words for each second agent; Obtain the first output information corresponding to the associated output information based on the prompt words of the second intelligent agent; Based on the first output information of each of the second intelligent agents, response information corresponding to the input information is generated.

19. The apparatus according to any one of claims 15-18, wherein, The device further includes: The acquisition module is used to initiate at least one third-party intelligent agent for dialogue monitoring and to acquire the dialogue content during the dialogue process. The analysis module is used to perform intelligent analysis of the dialogue content through the third intelligent agent.

20. The apparatus according to claim 19, wherein, The analysis module is also used for: The third intelligent agent extracts information from the dialogue content to obtain key information, which is used to optimize subsequent dialogue; or... The third intelligent agent clarifies and reflects on the dialogue content to obtain a dialogue reflection result.

21. The apparatus according to claim 20, wherein, The third intelligent agent is a dialogue analysis intelligent agent, and the analysis module is further used for: The key information and the dialogue reflection results are determined as the analysis results of the dialogue analysis agent; The analysis results of the dialogue analysis agent are processed in a structured manner to obtain the first structured data; The first structured data is determined to be shared data and stored in the data center. The first structured data can be accessed by at least one of the first and second intelligent agents.

22. The apparatus according to any one of claims 19, wherein, The third intelligent agent is a dialogue quality inspection intelligent agent, and the analysis module is further used for: The third intelligent agent performs illusion detection and compliance monitoring on the dialogue content to obtain quality inspection results.

23. The apparatus according to claim 22, wherein, The analysis module is also used for: Based on the quality detection results, abnormal dialogue content is identified and collected as sample data for fine-tuning the agent. The sample data is used to fine-tune the relevant agent.

24. The apparatus according to any one of claims 15-18, wherein, The second generation module is further configured to: Obtain the target service corresponding to the second intelligent agent; Obtain the configuration information of the target service corresponding to the second intelligent agent, and obtain the auxiliary information required by the second intelligent agent from the configuration information.

25. The apparatus according to claim 15, wherein, The first generation module is further configured to: Acquire user profiles and historical behavior data as auxiliary information for the first intelligent agent; The first intelligent agent performs intent recognition on the input information based on the auxiliary information to obtain an intent information set, wherein the intent information set includes at least one user intent; The one or more second agents are determined from the set of agents based on the user intent.

26. The apparatus according to claim 15, wherein, The first generation module is further configured to: The current target business scenario is determined based on the input information; Obtain the intelligent agent associated with the target business scenario; Based on the intelligent agents associated with the target business scenario, determine the set of intelligent agents corresponding to the target business scenario.

27. The apparatus according to claim 15, wherein, The first generation module is further configured to: Obtain the industry to which the intelligent service belongs and the industry of the user corresponding to the input information; A set of intelligent agents is obtained based on the industry to which the intelligent service belongs and the industry of the user corresponding to the input information.

28. The apparatus according to any one of claims 15-18 and 25-27, wherein the set of intelligent agents includes at least one of the following intelligent agents: a speech generation intelligent agent, a retrieval enhancement generation intelligent agent, an active dialogue intelligent agent, a dialogue-generated image intelligent agent, a dialogue-generated video intelligent agent, an image recognition intelligent agent, and an information search intelligent agent.

29. An electronic device comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method as described in any one of claims 1-14.

30. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-14.

31. A computer program product, comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the method of any one of claims 1-14.

Images