Group chat questioning method and device, equipment, storage medium and computer program product

Abstract

The invention discloses a group chat questioning method and device, equipment, a storage medium and a computer program product, and relates to the technical field of intelligent AI, the group chat questioning method comprises the steps that user questions in a discussion group are acquired, the discussion group comprises a user and a target model group, the target model group comprises at least one large model group member, and the user questions in the discussion group are acquired; the large model group members correspond to at least one large model; inputting the user question into a target model group corresponding to the discussion group to obtain large model answer information of at least one large model group member to the user question; and generating and displaying a model reply result according to the large model reply information. Through the mode, the user can obtain collaborative answers and help of a plurality of large models without additional operation, the interaction efficiency with the AI large model is improved, and the comprehensiveness of knowledge acquisition is expanded.

Description

Technical Field

[0001] The present application relates to the field of intelligent AI technology, and in particular to group chat questioning methods, devices, equipment, storage media, and computer program products. Background Art

[0002] With the continuous advancement and iteration of AI technology, numerous large-scale models have emerged that are readily available to ordinary users. Each large-scale model allows for conversational questioning and knowledge search. However, currently, each large-scale model interacts with its own independent operation and conversational interfaces. This makes it inconvenient for users to switch between task windows when they need to consult multiple opinions and knowledge.

[0003] The above content is only used to assist in understanding the technical solution of this application and does not constitute an admission that the above content is prior art. Summary of the Invention

[0004] The main purpose of this application is to provide a group chat questioning method, device, equipment, storage medium and computer program product, aiming to solve the technical problem of not being able to ask questions to multiple AI interactions at the same time.

[0005] To achieve the above objectives, this application proposes a group chat questioning method, which includes:

[0006] Obtaining a user question in a discussion group, wherein the discussion group includes users and a target model group, wherein the target model group includes at least one large model group member, and the large model group member corresponds to at least one large model;

[0007] Inputting the user question into the target model group corresponding to the discussion group, and obtaining large model answer information of at least one large model group member to the user question;

[0008] Generate and display model response results based on the large model response information.

[0009] Optionally, the user question includes at least one of the following:

[0010] The text of the user question, the keywords of the user question, the target model group corresponding to the user question, group member designation information, and the domain information corresponding to the user question; and / or,

[0011] The target model group includes at least one of the following:

[0012] Group naming information, model configuration information, and at least one large model corresponding to the model configuration information.

[0013] Optionally, the steps for obtaining user questions in the discussion group include:

[0014] Determine keywords, question model groups, and domain information of the user question based on the text of the user question triggered in the discussion group;

[0015] Filtering the question model group based on the domain information to obtain a target model group;

[0016] The user question is determined based on the keywords, target model group and domain information of the user question.

[0017] Optionally, before the step of inputting the user question into the target model group corresponding to the discussion group and obtaining large model answer information of at least one large model group member to the user question, the step further includes:

[0018] Obtain a preset model list, wherein the preset model list includes the model configuration information and at least one large model of the model configuration information;

[0019] Constructing an initial model group based on the group naming information;

[0020] Each major model in the preset model list is connected to the initial model group to obtain a target model group, and each major model in the target model group shares the conversation record information in the discussion group corresponding to the target model group.

[0021] Optionally, the step of inputting the user question into the target model group corresponding to the discussion group and obtaining large model answer information of at least one large model group member to the user question includes:

[0022] Inputting the user question into the target group corresponding to the discussion group for semantic analysis to determine the type of the user question;

[0023] The large model group members corresponding to the user question type respond to the user question to obtain large model answer information.

[0024] Optionally, the step of having a large model group member corresponding to the user question type reply to the user question to obtain large model answer information includes:

[0025] Determine, based on the user question and the user question type, from the large models in the target model group a first large model group member who answers the user question, wherein the first large model group member is a large model group member corresponding to the user question type;

[0026] Obtaining context information of the current group chat session in the target model group;

[0027] Inputting the context information and the user question into a target model group and asking the first large model group members a question, thereby obtaining target answer information fed back by each first large model group member;

[0028] The second largest model group member makes a supplementary reply based on the target answer information of the first largest model group member to obtain supplementary answer information;

[0029] Large model answer information is determined based on the target answer information and the supplemental answer information.

[0030] Optionally, the step of determining large model answer information based on the target answer information and the supplementary answer information further includes:

[0031] Determining the types of problems that the members of the first large model group and the members of the second large model group are good at;

[0032] When the proficient question type does not match the user question type, the third largest model group member with the highest matching degree with the user question type in the discussion group is called to supplement and update the large model answer information.

[0033] Optionally, after the step of inputting the context information and the user question into the target model group and asking the first large model group members a question, and obtaining large model answer information fed back by each first large model group member, the step further includes:

[0034] Inputting the large model answer information into the target model group as update context information;

[0035] Inputting the updated context information to the first large model group members to ask additional questions, and obtaining updated answer information corresponding to each first large model group member;

[0036] determining an answer modification item based on the updated answer information and the large model answer information;

[0037] Summarize and count the modified items according to the answers to obtain an outline of modification suggestions;

[0038] Obtaining group chat record correction answer information based on the modification suggestion outline and the answer modification item;

[0039] The large model answer information is updated based on the modified answer information of the group chat record.

[0040] Optionally, the step of correcting the answer information according to the group chat record and updating the large model answer information includes:

[0041] Correct the answer information according to the group chat record to determine the large model opinion statistics corresponding to each large model;

[0042] Obtaining large model opinion unified information and large model opinion distinguishing information according to the large model opinion statistical information;

[0043] Generate modification opinion information and modification summary information corresponding to the large model answer information based on the large model opinion unified information and the large model opinion difference information;

[0044] The large model answer information is updated based on the modification opinion information and the modification summary information.

[0045] Optionally, before the step of obtaining the context information of the current session, the step further includes:

[0046] When a context clearing instruction is received, the historical conversation information of the current group chat session in the target model group is cleared, and the user question is input into the target model group to ask the first large model group members to obtain the large model answer information fed back by each first large model group member.

[0047] Optionally, the model response includes at least one of the following:

[0048] The reasoning process information corresponding to each major model, group chat answer list information, and model reply text.

[0049] Optionally, the step of generating and displaying a model response result based on the large model response information includes:

[0050] Based on the large model answer information, generate the reasoning process information and group chat answer list information corresponding to each large model;

[0051] Generate a model reply text according to the reasoning process information and the group chat answer list information;

[0052] Selecting a target summary model from the target model group;

[0053] Inputting the model reply text into the target summary model to obtain model summary information of the target summary model;

[0054] Generate and display model response results based on the model summary information.

[0055] Optionally, the model response includes at least one of the following:

[0056] Discussion thread information, reply text information, and reply record information.

[0057] Optionally, the method further includes:

[0058] Based on the model reply result, initiating a reply discussion instruction to each available large model in the target model group;

[0059] Sending a group chat discussion instruction to each available large model in turn through the reply discussion instruction, and updating the context information of the current group chat session in real time according to the discussion information replied by each available large model;

[0060] Based on the updated context information of the current group chat session, each available large model is controlled to reply in sequence and the model reply results are analyzed to generate and display the model group chat discussion results.

[0061] Optionally, after the step of controlling each available large model to reply in sequence and analyze the model reply results through the reply discussion instruction, and generating and displaying the model discussion results, the following step is further included:

[0062] When receiving a stop discussion instruction, stop replying to the group chat of each available large model and obtain the historical discussion record of the target model group;

[0063] Generate and display model summary information based on the historical discussion records

[0064] In addition, to achieve the above-mentioned purpose, the present application also proposes a group chat questioning device, which includes:

[0065] An information acquisition module is used to acquire user questions in a discussion group, wherein the discussion group includes users and a target model group, wherein the target model group includes at least one large model group member, and each large model group member corresponds to at least one large model;

[0066] A model input module, configured to input the user question into the target model group corresponding to the discussion group, and obtain a large model answer information to the user question from at least one large model group member;

[0067] The answer generation module is used to generate and display the model answer result based on the large model answer information.

[0068] In addition, to achieve the above-mentioned purpose, the present application also proposes a group chat questioning device, which includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the computer program is configured to implement the steps of the group chat questioning method described above.

[0069] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium. A computer program is stored on the storage medium, and when the computer program is executed by the processor, the steps of the group chat questioning method described above are implemented.

[0070] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the group chat questioning method described above.

[0071] One or more technical solutions proposed in this application have at least the following technical effects:

[0072] This application obtains user questions in a discussion group, wherein the discussion group includes users and a target model group, wherein the target model group includes at least one large model group member, and the large model group member corresponds to at least one large model; the user question is input into the target model group corresponding to the discussion group, and the large model answer information of at least one large model group member to the user question is obtained; and the model answer result is generated and displayed according to the large model answer information. In this way, a discussion group is formed by adding multiple large models to users, and the large model group members of the target model group in the discussion group analyze and answer the user questions, so that the user can get the collaborative answers and help of multiple large models without additional operations, thereby improving the efficiency of interaction with the AI ​​large model and expanding the comprehensiveness of knowledge acquisition. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0074] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0075] Figure 1 A flowchart of the first embodiment of the group chat questioning method of this application is provided;

[0076] Figure 2 This is a schematic diagram of a user asking a question in an embodiment of the group chat questioning method of this application;

[0077] Figure 3 A flowchart of the second embodiment of the group chat questioning method of this application is provided;

[0078] Figure 4 This is a schematic diagram of the module structure of the group chat questioning device according to an embodiment of the present application;

[0079] Figure 5 This is a schematic diagram of the device structure of the hardware operating environment involved in the group chat questioning method in the embodiment of the present application.

[0080] The purpose, features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0081] It should be understood that the specific embodiments described herein are merely used to explain the technical solutions of the present application and are not intended to limit the present application.

[0082] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0083] The main solution of the embodiment of the present application is: by obtaining user questions in a discussion group, the discussion group includes users and a target model group, the target model group includes at least one large model group member, and the large model group member corresponds to at least one large model; the user question is input into the target model group corresponding to the discussion group, and the large model answer information of at least one large model group member to the user question is obtained; and the model reply result is generated and displayed according to the large model answer information.

[0084] In this embodiment, for ease of description, the following description is made with identification of a smart terminal as the execution subject.

[0085] As AI technology continues to evolve and evolve, numerous large-scale models have emerged that are readily available to ordinary users. Each model allows for conversational questioning and knowledge search. However, each model currently interacts with its own independent user interface and conversational interface. This inconveniences users who need to reference multiple opinions and knowledge points by switching between task windows.

[0086] This application provides a solution, through which multiple large models and users are combined to form a discussion group, and the large model group members of the target model group in the discussion group analyze and answer user questions, so that users can get collaborative answers and help from multiple large models without additional operations, thereby improving the efficiency of interaction with AI large models and expanding the comprehensiveness of knowledge acquisition.

[0087] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, mobile phone, etc., or an electronic device or server capable of implementing the above functions. The following uses a smart terminal as an example to illustrate this embodiment and the following embodiments.

[0088] Based on this, the embodiment of the present application provides a group chat questioning method, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the group chat questioning method of this application.

[0089] In this embodiment, the group chat questioning method includes steps S10 to S30:

[0090] Step S10, obtaining user questions in a discussion group, wherein the discussion group includes users and a target model group, wherein the target model group includes at least one large model group member, and the large model group member corresponds to at least one large model;

[0091] It should be noted that user questions refer to questions entered by users in the discussion group through dialog box input or voice input, and can be any questions in any field, such as finance, sports or history, etc. Specific questions can be obtained by direct retrieval, or they can be summarized after thinking and analysis by a large model. This embodiment does not limit this.

[0092] It should be understood that user questions include, but are not limited to, the text of the user question, keywords in the user question, the target model group corresponding to the user question, group member designation information, and domain information corresponding to the user question. Keywords in the user question are keywords automatically identified by the big model or the app, corresponding to the core semantics of the user question. Group member designation information is the big model member designated by the user to answer the question.

[0093] In a specific implementation, the target model group includes the group name pre-built by the user, the configuration of each large model member, and the input port of each large model.

[0094] In a feasible implementation, in order to accurately obtain user questions, step S10 includes: determining the keywords, question model groups and domain information of the user questions based on the text of the user questions triggered in the discussion group; filtering the question model groups based on the domain information to obtain the target model groups; determining the user questions according to the keywords, target model groups and domain information of the user questions.

[0095] It should be noted that first, based on the text of the user question input by the user in the discussion group, the text is analyzed to determine the keywords, question model groups and domain information. Then, based on the domain information, the question model module is searched to obtain the target model group that can answer the corresponding domain of the user question. Then, the user question is determined based on the keywords, target model group and domain information of the user question. It should be emphasized that there can be multiple question model groups in the discussion group, which can answer questions in different fields respectively. Then, at least one of the multiple question model groups is selected as the target model group. For example: there are three question model groups A, B and C in the discussion group. Question model group A can answer questions in fields A and B, question model group B can answer questions in fields A and C, and question model group C can answer questions in field C. If the user question corresponds to field C, question model group B can be used as the target model group, or question model group C can be used as the target model group, or B and C can be merged as the target model group.

[0096] Step S20: input the user question into the target model group corresponding to the discussion group, and obtain a large model answer information of at least one large model group member to the user question;

[0097] It should be understood that if Figure 2 As shown, after obtaining the user question, the user question is input into the target model group corresponding to the discussion group, that is, a preliminary screening is performed to find the target model group that meets the user question field, so that a reply from at least one large model member can be obtained. The reply is displayed in the form of a chat box, and then all the replies are sorted to obtain the large model answer information.

[0098] In a feasible implementation, in order to set the target model group in advance, before step S20, it also includes: obtaining a preset model list, the preset model list including the model configuration information and at least one large model of the model configuration information; constructing an initial model group based on the group naming information; connecting each major model in the preset model list to the initial model group to obtain a target model group, and the major models in the target model group share the conversation record information in the discussion group corresponding to the target model group.

[0099] It should be noted that the discussion group refers to a pre-built conversation group chat, which can be in the form of a group chat, a shared forum, etc., which includes users and a target model group. The target model group includes at least one large model group member, and each large model group member corresponds to at least one large model, that is, the target model group corresponds to multiple large model group members, but each large model group member can correspond to multiple large models, or only one large model, that is, multiple large models can correspond to the same large model group member and use the same account identity to conduct conversations. For example: the target model group includes three large model group members A, B, and C, but A corresponds to large models No. 1 and No. 2, B corresponds to large model No. 3, and C corresponds to large models No. 4, No. 5, and No. 6.

[0100] It should be understood that the preset model list is first obtained, which includes model configuration information and multiple large models that can be called, and then the initial model group is constructed based on the group naming information set by the user, and the large models in the preset model list are added to obtain the target model group, and the target model group is connected to the discussion group, so that the conversation records and chat records therein can be shared.

[0101] Step S30: Generate and display a model response result based on the large model response information.

[0102] In practice, the model response results include the reasoning process information corresponding to each large model, a summary list of all answers in the group chat, and the reply text of each large model under the list. Each large model member's reply includes the thinking and reasoning process and the final answer, which are then summarized to form the group chat answer list information.

[0103] In a feasible implementation, in order to accurately summarize the large model answer information to obtain the model reply result, step S30 includes: based on the large model answer information, generating the reasoning process information and group chat answer list information corresponding to each major model; generating the model reply text according to the reasoning process information and the group chat answer list information; selecting the target summary model from the target model group; inputting the model reply text into the target summary model to obtain the model summary information of the target summary model; generating and displaying the model reply result according to the model summary information.

[0104] It should be noted that first, based on the answer information of the large model, the reasoning process information and group chat answer list corresponding to each large model are generated, that is, the replies of each large model are split, the reasoning process is classified and displayed, and then the final answer is summarized into a table.

[0105] It should be understood that after the inference process information and group chat answer list information are converted into text and semantically analyzed and understood, a summarized model response text is obtained. Then, a model is selected from the target model group as the target summary model. The target summary model then analyzes and summarizes the model response text. Specifically, the target summary model performs semantic extraction on all model response texts, selects the parts with most common opinions, and generates summary information. Finally, the model response result is generated and displayed based on the model summary information.

[0106] In a specific implementation, the model response results include not only the reasoning process information corresponding to each major model, group chat answer list information, and model response text, but also at least one of the discussion topic information, response text information, and response record information.

[0107] In a feasible implementation, in order to discuss and analyze the model reply results through each model in the discussion group after generating the model reply results, the method of this embodiment also includes: based on the model reply results, initiating a reply discussion instruction to each available large model in the target model group; sending a group chat discussion instruction to each available large model in turn through the reply discussion instruction, and updating the context information of the current group chat session in real time according to the discussion information replied by each available large model; based on the updated context information of the current group chat session, controlling each available large model to reply and analyze the model reply results in turn, and generating and displaying the model group chat discussion results.

[0108] It should be noted that, first, based on the model reply result, a reply discussion instruction is initiated to all available large models that can be activated in the target model group. Thus, a group chat discussion instruction can be sent to each available large model in turn through the reply discussion instruction. Then, each available large model will update the context information in real time in the discussion group, that is, the model reply result is input into the available large model as the updated context information. The available large model then analyzes the context information for discussion and analysis of the model reply result, and finally obtains the model group chat discussion result. Among them, when no user stop instruction is received, each available large model will continue to receive updated context information and then reply to the discussion result.

[0109] In a feasible implementation, in order to stop the group chat discussion after the discussion is completed, the reply discussion instruction is used to control each available large model to reply in turn and analyze the model reply results. After the step of generating and displaying the model discussion results, it also includes: when receiving the stop discussion instruction, stopping the group chat reply of each available large model, and obtaining the historical discussion record of the target model group; generating and displaying model summary information based on the historical discussion record.

[0110] It should be understood that when a user-triggered instruction to stop discussion is received, each available large model is first notified to stop replying, and all discussion records after the model reply result until the reply is stopped are sorted out. Then, text analysis is performed based on the historical discussion records to generate model summary information, which includes the evaluation, analysis and opinion discussion of the model reply results by each available large model.

[0111] This embodiment provides a group chat questioning method, which obtains user questions in a discussion group, wherein the discussion group includes users and a target model group, wherein the target model group includes at least one large model group member, and the large model group member corresponds to at least one large model; the user question is input into the target model group corresponding to the discussion group, and large model answer information to the user question from at least one large model group member is obtained; and a model answer result is generated and displayed based on the large model answer information. In this way, a discussion group is formed by adding multiple large models to users, and the user questions are analyzed and answered by the large model group members of the target model group in the discussion group, so that users can obtain collaborative answers and help from multiple large models without additional operations, thereby improving the efficiency of interaction with AI large models and expanding the comprehensiveness of knowledge acquisition.

[0112] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above embodiment 1 can be referred to the above introduction and will not be described in detail later. Figure 2 , step S20 includes steps S201 to S202:

[0113] Step S201: input the user question into the target group corresponding to the discussion group for semantic analysis to determine the type of the user question;

[0114] It should be noted that the user question is first input into the target group corresponding to the discussion group for semantic analysis, so that the type of user question can be determined. The target group can be a large model group composed of any large model of group members, or one or more large models, and then the user question type can be analyzed, that is, whether the subdivision field and question type corresponding to the user question require model analysis and answer.

[0115] In step S202, a large model group member corresponding to the user question type responds to the user question to obtain large model answer information.

[0116] It should be understood that after determining the type of user question, the corresponding large model group members are determined to respond to the user question, and then the large model response information is aggregated. For example, first determine that the user question type is type A, then determine the types that each large model group member can answer, and then use the types that each large model group member can answer, including type A, as the large model group members corresponding to the user question type.

[0117] In a feasible implementation, in order to accurately generate large model answer information, step S202 includes: determining the first large model group member who answers the user question from the large model in the target model group based on the user question and the user question type, wherein the first large model group member is the large model group member corresponding to the user question type; obtaining the context information of the current group chat session in the target model group; inputting the context information and the user question into the target model group to ask the first large model group member, and obtaining the target answer information fed back by each first large model group member; obtaining the supplementary answer information by the second large model group member based on the target answer information of the first large model group member; and determining the large model answer information based on the target answer information and the supplementary answer information.

[0118] In a specific implementation, the first large model group member to respond is first determined from the large models of each group member in the target model group based on the user question and the user question type. The first large model group member includes at least one large model group member corresponding to the user question type.

[0119] It should be noted that the context information of the current group chat session is obtained. This context information includes the conversation records between all group member macro models and the user since the discussion group was established. This context information and the user's question are then input into the first macro model group member, allowing each macro model to process and analyze the information and obtain the target answer information.

[0120] It should be understood that after obtaining the target answer information, the second largest model group member continues to supplement and reply to the target answer information of the first largest model group member, obtaining supplementary answer information, and finally integrating and summarizing the target answer information and supplementary answer information to obtain the large model answer information. Among them, the second largest model group member refers to other large models in the target model group or in the discussion group that are not selected. They supplement and reply to the target answer information of the first largest model group member, obtaining supplementary answer information, and finally combining the supplementary answer information and the target answer information to obtain the large model answer information.

[0121] In a specific implementation, the first large model group member can be a member of the large model group specified by the user, and the second large model group member is a member of the large model group that can answer the user's question type but is not specified by the user. For example, if the large model group members that can answer the user's question include A, B, C, and D, and the user specifies A and B as the first large model group members, then C and D will serve as the second large model group members to provide supplementary responses.

[0122] In a feasible implementation, in order to automatically match and link a model that is more proficient in the current field to reply when the user does not select a large model that matches the field, the step of determining the large model answer information based on the target answer information and the supplementary answer information also includes: determining the types of questions that the first large model group members and the second large model group members are proficient in; when the proficient question types do not match the user question type, calling the third large model group member in the discussion group who has the highest matching degree with the user question type to supplement and update the large model answer information.

[0123] It should be noted that the types of problems that the members of the first and second model groups are good at are first determined, that is, the areas they are good at, the depth of the problems, whether deep thinking is required, etc.

[0124] It should be understood that if the question types that the first and second largest model group members are good at do not match the type of the user question asked by the current user, that is, the question types are different, then the large model group members in the discussion group who are good at the type closest to the user question type and most suitable for the discussion group but not designated by the user to answer will be called to supplement and update. The large model group member in the discussion group who is good at the type closest to the user question type and most suitable for the discussion group but not designated by the user to answer will be the third large model group member.

[0125] In a feasible implementation, in order to be able to update the context in real time and provide supplementary answers based on the updated context, the context information and the user question are input into the target model group to ask questions to the first large model group members, and the step of obtaining large model answer information fed back by each first large model group member also includes: inputting the large model answer information into the target model group as updated context information; inputting the updated context information into the first large model group members to ask additional questions to obtain updated answer information corresponding to each first large model group member; determining answer modification entries based on the updated answer information and the large model answer information; summarizing and counting the answer modification entries to obtain a modification opinion outline; obtaining group chat record correction answer information based on the modification opinion outline and the answer modification entries; and updating the large model answer information based on the group chat record correction answer information.

[0126] In a specific implementation, the existing answer text, i.e., the large model answer information, is first input into the target model group and marked as updated context information. The updated context information is then input into the first large model group members for additional questions. In other words, each first large model group member provides an additional answer based on the already generated large model answer information, resulting in updated answer information.

[0127] It should be noted that after obtaining the updated answer information, the updated answer information is compared with the large model answer information to determine the answer modification items, that is, the answer content that needs to be modified. After obtaining the answer modification items, the answer modification items are summarized and counted to obtain a modification suggestion outline, that is, the integration and statistics of all modification suggestions. Finally, based on the modification suggestion outline and the answer modification items, the completed large model reply in the chat record is revised to obtain the group chat record modified answer information, and then the large model answer information is updated.

[0128] In a feasible implementation manner, in order to accurately update the big model answer information, the steps of updating the big model answer information according to the group chat record correction answer information include: determining the big model opinion statistical information corresponding to each major model according to the group chat record correction answer information; obtaining the big model opinion unified information and the big model opinion difference information according to the big model opinion statistical information; generating the modification opinion information and the modification summary information corresponding to the big model answer information based on the big model opinion unified information and the big model opinion difference information; and updating the big model answer information based on the modification opinion information and the modification summary information.

[0129] It should be understood that first, based on the group chat record, the answer information is corrected to determine the big model opinion statistics corresponding to each big model group member, that is, the opinion statistics of each big model on the big model answer information, including unified opinions and opposing opinions. Then, based on the big model opinion statistics, the unified opinion information and opinion difference information of each big model group member are determined, that is, for the same point of view or information, each big model group member holds an opinion in favor, against, or neutral. For example: for a certain opinion in the big model answer information, opinion statistics are performed to obtain the big model unified opinion information and the big model opinion difference information. The big model unified opinion information shows that the big models that agree with the opinion include A, B, and C, and the big model opinion difference information shows that the big models that oppose the opinion include D and E. At the same time, the big model F that maintains a neutral opinion can be separately counted.

[0130] In the specific implementation, after obtaining the unified opinion information and the distinguished opinion information of the big model, the modification opinions of the big model answer information are collected to determine the modification opinion statistics for the content of the big model answer information, obtain the modification opinion information, and then summarize and summarize the modification opinion information through the big model to obtain the modification summary information.

[0131] It should be noted that after obtaining the modification opinion information and the modification summary information, the large model answer information is updated in the form of annotations based on the modification opinion information and the modification summary information.

[0132] In a feasible implementation, in order to clear the context answer in time when the user needs to clear it, the step of obtaining the context information of the current session also includes: when receiving the context clearing instruction, clearing the historical conversation information of the current group chat session in the target model group, and inputting the user question into the target model group to ask the first large model group members, and obtaining the large model answer information fed back by each first large model group member.

[0133] It should be understood that when a context clearing instruction triggered by a user is received, the historical conversation information of the current group chat session in the discussion group is cleared, for example, the chat history is cleared or the reference database is emptied.

[0134] In a specific implementation, after clearing, the user question is input into the target model group again to ask questions, and the large model answer information without contextual information thinking fed back by the members of the first large model group can be obtained.

[0135] This embodiment performs semantic analysis by inputting the user's question into the target group corresponding to the discussion group to determine the user's question type. Then, members of the large model group corresponding to the user's question type respond to the user's question, generating large model answer information. This method selects the corresponding large model group member to respond based on the user's question type, improving the quality of the response.

[0136] It should be noted that the above examples are only used to understand this application and do not constitute a limitation on the group chat question-asking method of this application. More simple transformations based on this technical concept are all within the scope of protection of this application.

[0137] This application also provides a group chat questioning device, please refer to Figure 4 , the group chat questioning device includes:

[0138] The information acquisition module 10 is used to obtain user questions in a discussion group, wherein the discussion group includes users and a target model group, wherein the target model group includes at least one large model group member, and the large model group member corresponds to at least one large model.

[0139] The model input module 20 is used to input the user question into the target model group corresponding to the discussion group, and obtain the large model answer information of at least one large model group member to the user question.

[0140] The answer generation module 30 is used to generate and display the model answer result according to the large model answer information.

[0141] This embodiment obtains user questions from a discussion group, wherein the discussion group includes users and a target model group, wherein the target model group includes at least one large model group member, and the large model group member corresponds to at least one large model; the user question is input into the target model group corresponding to the discussion group, and large model answer information to the user question from at least one large model group member is obtained; and a model answer result is generated and displayed based on the large model answer information. In this way, a discussion group is formed by combining multiple large models with users, and the large model group members of the target model group in the discussion group analyze and answer user questions, so that users can obtain collaborative answers and help from multiple large models without additional operations, thereby improving the efficiency of interaction with AI large models and expanding the comprehensiveness of knowledge acquisition.

[0142] In one embodiment, the information acquisition module 10 is also used to determine the keywords, question model group and domain information of the user question based on the text of the user question triggered in the discussion group; filter the question model group based on the domain information to obtain the target model group; and determine the user question based on the keywords, target model group and domain information of the user question.

[0143] In one embodiment, the model input module 20 is also used to obtain a preset model list, which includes the model configuration information and at least one large model of the model configuration information; construct an initial model group based on the group naming information; connect each major model in the preset model list to the initial model group to obtain a target model group, and the major models in the target model group share the conversation record information in the discussion group corresponding to the target model group.

[0144] In one embodiment, the model input module 20 is also used to input the user question into the target group corresponding to the discussion group for semantic analysis to determine the type of user question; the large model group members corresponding to the user question type reply to the user question to obtain large model answer information.

[0145] In one embodiment, the model input module 20 is further used to determine the first large model group member among the group members who responds to the user question from the large model in the target model group based on the user question and the user question type, wherein the first large model group member is the large model group member corresponding to the user question type; obtain the context information of the current group chat session in the target model group; input the context information and the user question into the target model group to ask the first large model group member, and obtain target answer information fed back by each first large model group member; obtain supplementary answer information by having the second large model group member make a supplementary reply based on the target answer information of the first large model group member; and determine the large model answer information based on the target answer information and the supplementary answer information.

[0146] In one embodiment, the model input module 20 is also used to determine the types of questions that the first large model group member and the second large model group member are good at; when the types of questions that the members are good at do not match the user question type, the third large model group member in the discussion group with the highest matching degree with the user question type is called to supplement and update the large model answer information.

[0147] In one embodiment, the model input module 20 is also used to input the large model answer information into the target model group as updated context information; input the updated context information into the first large model group members to ask additional questions to obtain updated answer information corresponding to each first large model group member; determine the answer modification entries based on the updated answer information and the large model answer information; summarize and count the answer modification entries to obtain a modification opinion outline; obtain group chat record modified answer information based on the modification opinion outline and the answer modification entries; and update the large model answer information based on the group chat record modified answer information.

[0148] In one embodiment, the model input module 20 is also used to determine the big model opinion statistical information corresponding to each big model based on the modified answer information of the group chat record; obtain the big model opinion unified information and the big model opinion difference information based on the big model opinion statistical information; generate the modification opinion information and modification summary information corresponding to the big model answer information based on the big model opinion unified information and the big model opinion difference information; and update the big model answer information based on the modification opinion information and the modification summary information.

[0149] In one embodiment, the model input module 20 is also used to clear the historical conversation information of the current group chat session in the target model group when receiving a context clearing instruction, and input the user question into the target model group to ask the first large model group members, and obtain the large model answer information fed back by each first large model group member.

[0150] In one embodiment, the reply generation module 30 is further used to generate reasoning process information and group chat answer list information corresponding to each major model based on the large model answer information; generate a model reply text according to the reasoning process information and the group chat answer list information; select a target summary model from the target model group; input the model reply text into the target summary model to obtain the model summary information of the target summary model; generate and display the model reply result according to the model summary information.

[0151] In one embodiment, the reply generation module 30 is further used to initiate a reply discussion instruction to each available large model in the target model group based on the model reply result; send a group chat discussion instruction to each available large model in turn through the reply discussion instruction, and update the context information of the current group chat session in real time according to the discussion information replied by each available large model; based on the updated context information of the current group chat session, control each available large model to reply and analyze the model reply result in turn, and generate and display the model group chat discussion result.

[0152] In one embodiment, the reply generation module 30 is further used to stop the group chat reply of each available large model when receiving a stop discussion instruction, and obtain the historical discussion record of the target model group; generate and display model summary information based on the historical discussion record.

[0153] The group chat questioning device provided in this application, using the group chat questioning method described in the above-mentioned embodiment, can resolve the technical issue of being unable to simultaneously ask questions to multiple AI interactions. Compared to the prior art, the beneficial effects of the group chat questioning device provided in this application are the same as those of the group chat questioning method described in the above-mentioned embodiment. Other technical features of the group chat questioning device are the same as those disclosed in the above-mentioned embodiment and are not further elaborated here.

[0154] The present application provides a group chat questioning device, which includes: 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, and the instructions are executed by the at least one processor to enable the at least one processor to execute the group chat questioning method in the above-mentioned embodiment 1.

[0155] Reference below Figure 5, which shows a schematic diagram of the structure of a group chat questioning device suitable for implementing the embodiments of the present application. The group chat questioning device in the embodiments of the present application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), in-vehicle terminals (such as in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 5 The group chat questioning device shown is merely an example and should not limit the functions and scope of use of the embodiments of the present application.

[0156] like Figure 5 As shown, the group chat questioning device may include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes based on programs stored in ROM (Read Only Memory) 1002 or programs loaded from storage device 1003 into RAM (Random Access Memory) 1004. RAM 1004 also stores various programs and data required for the operation of the group chat questioning device. Processing device 1001, ROM 1002, and RAM 1004 are interconnected via bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input device 1007 including, for example, a touch screen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output device 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage device 1003 including, for example, a magnetic tape, hard disk, etc.; and communication device 1009. The communication device 1009 can allow the group chat questioning device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows a group chat questioning device with various systems, it should be understood that it is not required to implement or have all of the systems shown. More or fewer systems can be implemented or provided instead.

[0157] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program comprising program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0158] The group chat questioning device provided in this application, using the group chat questioning method described in the above embodiment, can resolve the technical issue of being unable to simultaneously ask questions to multiple AI interactions. Compared to the prior art, the beneficial effects of the group chat questioning device provided in this application are the same as those of the group chat questioning method described in the above embodiment. Other technical features of the group chat questioning device are the same as those disclosed in the above embodiment and are not further elaborated here.

[0159] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0160] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

[0161] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the group chat questioning method in the above embodiment.

[0162] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0163] The computer-readable storage medium may be included in the group chat questioning device; or may exist independently without being assembled into the group chat questioning device.

[0164] The above-mentioned computer-readable storage medium carries one or more programs. When the above-mentioned one or more programs are executed by the group chat questioning device, the group chat questioning device: obtains user questions in the discussion group, the discussion group includes users and target model groups, the target model group includes at least one large model group member, and the large model group member corresponds to at least one large model; inputs the user question into the target model group corresponding to the discussion group, and obtains the large model answer information of at least one large model group member to the user question; generates and displays the model reply result according to the large model answer information.

[0165] Computer program code for performing the operations of the present application may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0166] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, program segment or a part of code, and the module, program segment or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the specified function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.

[0167] The modules described in the embodiments of the present application may be implemented in software or hardware, wherein the name of a module does not necessarily limit the unit itself.

[0168] The readable storage medium provided in this application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., a computer program) for executing the above-mentioned group chat questioning method, which can solve the technical problem of not being able to ask questions to multiple AI interactions simultaneously. Compared with the existing technology, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the group chat questioning method provided in the above embodiment, and will not be repeated here.

[0169] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned group chat questioning method when executed by a processor.

[0170] The computer program product provided in this application can solve the technical problem of not being able to ask questions to multiple AI interactions at the same time. Compared with the existing technology, the beneficial effects of the computer program product provided in this application are the same as the beneficial effects of the group chat questioning method provided in the above embodiment, and will not be repeated here.

[0171] The above description is only part of the embodiments of the present application and does not limit the patent scope of the present application. All equivalent structural transformations made by using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect application in other related technical fields are included in the patent protection scope of the present application.

[0172] The present invention discloses A1. a group chat questioning method, the method comprising:

[0173] Obtaining a user question in a discussion group, wherein the discussion group includes users and a target model group, wherein the target model group includes at least one large model group member, and the large model group member corresponds to at least one large model;

[0174] Inputting the user question into the target model group corresponding to the discussion group, and obtaining large model answer information of at least one large model group member to the user question;

[0175] Generate and display model response results based on the large model response information.

[0176] A2. The method described in A1, wherein the user question includes at least one of the following:

[0177] The text of the user question, the keywords of the user question, the target model group corresponding to the user question, group member designation information, and the domain information corresponding to the user question; and / or,

[0178] The target model group includes at least one of the following:

[0179] Group naming information, model configuration information, and at least one large model corresponding to the model configuration information.

[0180] A3. As described in the method of A2, the step of obtaining user questions in the discussion group includes:

[0181] Determine keywords, question model groups, and domain information of the user question based on the text of the user question triggered in the discussion group;

[0182] Filtering the question model group based on the domain information to obtain a target model group;

[0183] The user question is determined based on the keywords, target model group and domain information of the user question.

[0184] A4. The method of A1, wherein before inputting the user question into the target model group corresponding to the discussion group and obtaining a large model answer to the user question from at least one large model group member, the method further comprises:

[0185] Obtain a preset model list, wherein the preset model list includes the model configuration information and at least one large model of the model configuration information;

[0186] Constructing an initial model group based on the group naming information;

[0187] Each major model in the preset model list is connected to the initial model group to obtain a target model group, and each major model in the target model group shares the conversation record information in the discussion group corresponding to the target model group.

[0188] A5. The method described in A1, wherein the step of inputting the user question into the target model group corresponding to the discussion group and obtaining a large model answer to the user question from at least one large model group member comprises:

[0189] Inputting the user question into the target group corresponding to the discussion group for semantic analysis to determine the type of the user question;

[0190] The large model group members corresponding to the user question type respond to the user question to obtain large model answer information.

[0191] A6. The method described in A5, wherein the step of having a large model group member corresponding to the user question type respond to the user question to obtain large model answer information includes:

[0192] Determine, based on the user question and the user question type, from the large models in the target model group a first large model group member who answers the user question, wherein the first large model group member is a large model group member corresponding to the user question type;

[0193] Obtaining context information of the current group chat session in the target model group;

[0194] Inputting the context information and the user question into a target model group and asking the first large model group members a question, thereby obtaining target answer information fed back by each first large model group member;

[0195] The second largest model group member makes a supplementary reply based on the target answer information of the first largest model group member to obtain supplementary answer information;

[0196] Large model answer information is determined based on the target answer information and the supplemental answer information.

[0197] A7. The method according to A6, wherein the step of determining the large model answer information based on the target answer information and the supplemental answer information further comprises:

[0198] Determining the types of problems that the members of the first large model group and the members of the second large model group are good at;

[0199] When the proficient question type does not match the user question type, the third largest model group member with the highest matching degree with the user question type in the discussion group is called to supplement and update the large model answer information.

[0200] A8. The method of A6, wherein the step of inputting the context information and the user question into the target model group and asking the first large model group members a question, and obtaining large model answer information fed back by each of the first large model group members, further comprises:

[0201] Inputting the large model answer information into the target model group as update context information;

[0202] Inputting the updated context information to the first large model group members to ask additional questions, and obtaining updated answer information corresponding to each first large model group member;

[0203] determining an answer modification item based on the updated answer information and the large model answer information;

[0204] Summarize and count the modified items according to the answers to obtain an outline of modification suggestions;

[0205] Obtaining group chat record correction answer information based on the modification suggestion outline and the answer modification item;

[0206] The large model answer information is updated based on the modified answer information of the group chat record.

[0207] A9. The method as described in A8, wherein the step of updating the large model answer information based on the group chat record correcting the answer information includes:

[0208] Correct the answer information according to the group chat record to determine the large model opinion statistics corresponding to each large model;

[0209] Obtaining large model opinion unified information and large model opinion distinguishing information according to the large model opinion statistical information;

[0210] Generate modification opinion information and modification summary information corresponding to the large model answer information based on the large model opinion unified information and the large model opinion difference information;

[0211] The large model answer information is updated based on the modification opinion information and the modification summary information.

[0212] A10. The method of A6 further includes the following steps before obtaining the context information of the current session:

[0213] When a context clearing instruction is received, the historical conversation information of the current group chat session in the target model group is cleared, and the user question is input into the target model group to ask the first large model group members to obtain the large model answer information fed back by each first large model group member.

[0214] A11. The method described in A1, wherein the model response includes at least one of the following:

[0215] The reasoning process information corresponding to each major model, group chat answer list information, and model reply text.

[0216] A12. The method described in A11, wherein the step of generating and displaying a model response result based on the large model response information includes:

[0217] Based on the large model answer information, generate the reasoning process information and group chat answer list information corresponding to each large model;

[0218] Generate a model reply text according to the reasoning process information and the group chat answer list information;

[0219] Selecting a target summary model from the target model group;

[0220] Inputting the model reply text into the target summary model to obtain model summary information of the target summary model;

[0221] Generate and display model response results based on the model summary information.

[0222] A13. The method according to any one of A1 to A12, wherein the model response comprises at least one of the following:

[0223] Discussion thread information, reply text information, and reply record information.

[0224] A14. The method according to A13, further comprising:

[0225] Based on the model reply result, initiating a reply discussion instruction to each available large model in the target model group;

[0226] Sending a group chat discussion instruction to each available large model in turn through the reply discussion instruction, and updating the context information of the current group chat session in real time according to the discussion information replied by each available large model;

[0227] Based on the updated context information of the current group chat session, each available large model is controlled to reply in sequence and the model reply results are analyzed to generate and display the model group chat discussion results.

[0228] A15. The method of A14, wherein the steps of controlling each available large model to reply sequentially and analyzing the model reply results through the reply discussion instruction, and generating and displaying the model discussion results further include:

[0229] When receiving a stop discussion instruction, stop replying to the group chat of each available large model and obtain the historical discussion record of the target model group;

[0230] Generate and display model summary information based on the historical discussion records.

[0231] The present invention also discloses B16. A group chat questioning device, comprising:

[0232] An information acquisition module is used to acquire user questions in a discussion group, wherein the discussion group includes users and a target model group, wherein the target model group includes at least one large model group member, and each large model group member corresponds to at least one large model;

[0233] A model input module, configured to input the user question into the target model group corresponding to the discussion group, and obtain a large model answer information to the user question from at least one large model group member;

[0234] The answer generation module is used to generate and display the model answer result based on the large model answer information.

[0235] B17. In the group chat questioning device as described in B16, the model input module is further used to determine at least one first large model group member from the large models in the target model group based on the user question; obtain context information of the current group chat session in the target model group; input the context information and the user question into the target model group to ask questions to the first large model group members, and obtain large model answer information fed back by each first large model group member.

[0236] The present invention also discloses C18. A group chat questioning device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the group chat questioning method described above.

[0237] The present invention also discloses D19. A storage medium, which is a computer-readable storage medium and stores a computer program. When the computer program is executed by a processor, the steps of the group chat questioning method described above are implemented.

[0238] The present invention also discloses E20. A computer program product, comprising a computer program, which implements the steps of the group chat questioning method described above when executed by a processor.

Claims

1. A group chat questioning method, characterized in that: The method includes: Obtaining a user question in a discussion group, wherein the discussion group includes users and a target model group, wherein the target model group includes at least one large model group member, and the large model group member corresponds to at least one large model; Inputting the user question into the target model group corresponding to the discussion group, and obtaining large model answer information of at least one large model group member to the user question; Generate and display model response results based on the large model response information.

2. The method according to claim 1, wherein The user question includes at least one of the following: The text of the user question, the keywords of the user question, the target model group corresponding to the user question, group member designation information, and the domain information corresponding to the user question; and / or, The target model group includes at least one of the following: Group naming information, model configuration information, and at least one large model corresponding to the model configuration information.

3. The method according to claim 2, wherein The steps to obtain user questions in the discussion group include: Determine keywords, question model groups, and domain information of the user question based on the text of the user question triggered in the discussion group; Filtering the question model group based on the domain information to obtain a target model group; The user question is determined based on the keywords, target model group and domain information of the user question.

4. The method according to claim 1, wherein Before the step of inputting the user question into the target model group corresponding to the discussion group and obtaining the large model answer information of at least one large model group member to the user question, the step further includes: Obtain a preset model list, wherein the preset model list includes the model configuration information and at least one large model of the model configuration information; Constructing an initial model group based on the group naming information; Each major model in the preset model list is connected to the initial model group to obtain a target model group, and each major model in the target model group shares the conversation record information in the discussion group corresponding to the target model group.

5. The method according to claim 1, wherein The step of inputting the user question into the target model group corresponding to the discussion group and obtaining large model answer information of at least one large model group member to the user question includes: Inputting the user question into the target group corresponding to the discussion group for semantic analysis to determine the type of the user question; The large model group members corresponding to the user question type respond to the user question to obtain large model answer information.

6. The method according to claim 5, wherein The step of having a large model group member corresponding to the user question type reply to the user question to obtain large model answer information includes: Determine, based on the user question and the user question type, from the large models in the target model group a first large model group member who answers the user question, wherein the first large model group member is a large model group member corresponding to the user question type; Obtaining context information of the current group chat session in the target model group; Inputting the context information and the user question into a target model group and asking the first large model group members a question, thereby obtaining target answer information fed back by each first large model group member; The second largest model group member makes a supplementary reply based on the target answer information of the first largest model group member to obtain supplementary answer information; Large model answer information is determined based on the target answer information and the supplemental answer information.

7. A group chat questioning device, characterized in that: The device comprises: An information acquisition module is used to acquire user questions in a discussion group, wherein the discussion group includes users and a target model group, wherein the target model group includes at least one large model group member, and each large model group member corresponds to at least one large model; A model input module, configured to input the user question into the target model group corresponding to the discussion group, and obtain a large model answer information to the user question from at least one large model group member; The answer generation module is used to generate and display the model answer result based on the large model answer information.

8. A group chat questioning device, characterized in that: The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the group chat questioning method according to any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the group chat questioning method according to any one of claims 1 to 6 are implemented.

10. A computer program product, characterized in that The computer program product includes a computer program, and when the computer program is executed by a processor, the steps of the group chat questioning method according to any one of claims 1 to 6 are implemented.

Images