Interactive dialogue management method and device based on language model, equipment and medium

Abstract

The invention discloses an interactive dialogue management method and device based on a language model, equipment and a medium, belongs to the field of data communication, and considers that a pre-training language model can be used in a dialogue mode and can process various language processing tasks. Therefore, the to-be-sent message of the client and the identity information of the addressee of the to-be-sent message can be received, then whether the addressee of the to-be-sent message belongs to the pre-training language model or not is judged according to the identity information, if not, the to-be-sent message can be directly sent to the addressee, and if yes, the to-be-sent message is sent to the addressee. If yes, the to-be-sent message can be sent to the pre-training language model, and the processing result of the pre-training language model for the to-be-sent message is pushed to the client, so that a user can use the pre-training language model in a dialogue form in chat software, the processing requirements of various language processing tasks of the user can be met, and the user experience is improved. And the user experience is improved.

Description

Technical Field

[0001] The present invention relates to the field of data communication, and in particular to a language model-based interactive dialogue management method, device, equipment and medium. Background Art

[0002] Various types of chat software usually provide users with functions for information exchange between users, such as private chat between two users and group chat within a chat group composed of multiple users. However, the functions that chat software can provide are relatively single and difficult to meet the diverse needs of users. The user experience needs to be improved.

[0003] Therefore, how to provide a solution to the above technical problems is a problem that those skilled in the art need to solve at present. Summary of the invention

[0004] The purpose of the present invention is to provide an interactive dialogue management method, device, equipment and computer-readable storage medium based on a language model. When it is determined that the recipient of the message to be sent belongs to a pre-trained language model, the message to be sent can be sent to the pre-trained language model, and the processing result of the pre-trained language model for the message to be sent can be pushed to the client, so that the user can use the pre-trained language model in the chat software through a dialogue form, which can meet the user's processing needs for various language processing tasks and improve the user experience.

[0005] In order to solve the above technical problems, the present invention provides an interactive dialogue management method based on a language model, which is applied to a server and includes:

[0006] Receive a message to be sent from a client and identity information of a recipient of the message to be sent;

[0007] Determining whether the recipient of the message to be sent belongs to the pre-trained language model according to the identity information;

[0008] If yes, sending the message to be sent to the pre-trained language model;

[0009] Pushing the processing result of the pre-trained language model on the message to be sent to the client;

[0010] If not, the message to be sent is directly sent to the recipient.

[0011] On the other hand, receiving the message to be sent from the client and the identity information of the recipient of the message to be sent includes:

[0012] Receiving a message to be sent from a client through a first channel or a second channel, wherein the first channel refers to a private chat channel and the second channel refers to a group chat channel;

[0013] If the message to be sent comes from a private chat channel, determine the identity information of the recipient of the message to be sent;

[0014] If the message to be sent originates from a group chat channel and the client specifies a recipient of the message to be sent, the identity information of the recipient of the message to be sent is determined, and the message to be sent and the identity information of the recipient are pushed to the group chat interface;

[0015] If the message to be sent originates from a group chat channel and the client does not specify a recipient of the message to be sent, the message to be sent is pushed to the group chat interface.

[0016] On the other hand, the messages to be sent whose recipients belong to the pre-trained language model include:

[0017] Data modality conversion tasks;

[0018] The data modality conversion task refers to: converting the data of the first modality given by the user into the data of the specified second modality, wherein the first modality and the second modality are respectively one of text, picture and video.

[0019] On the other hand, the messages to be sent whose recipients belong to the pre-trained language model also include:

[0020] Function call type tasks, where the function call means: the function module deployed in the server needs to be called;

[0021] Sending the message to be sent to the pre-trained language model includes:

[0022] Sending the function call type task to the pre-trained language model so that the pre-trained language model completes the function call type task by calling the function module deployed in the server;

[0023] The function call type tasks include:

[0024] A task creation task includes a task time, task content and a task executor, and the functional module corresponding to the task creation task is an office system.

[0025] On the other hand, the interactive dialogue management method based on language model also includes:

[0026] In response to a model adding instruction sent by a client, a preset language model list is pushed to the client, wherein the model adding instruction includes a first instruction to add a pre-trained language model as a friend and a second instruction to add the pre-trained language model as a group member of a designated group chat, and the language model list includes a plurality of pre-trained language models;

[0027] In response to a selection instruction from a client, determining a target language model selected by the client from the language model list;

[0028] If the model adding instruction is the first instruction, the target language model is used as a friend of the client;

[0029] If the model adding instruction is the second instruction, the target language model is used as a group member of the designated group chat.

[0030] On the other hand, sending the message to be sent to the pre-trained language model includes:

[0031] Determining whether the message to be sent is within the processing capability range of the pre-trained language model;

[0032] If yes, sending the message to be sent to the pre-trained language model;

[0033] If not, the client is prompted that the message content cannot be processed.

[0034] On the other hand, the messages to be sent received through the group chat channel and whose recipients belong to the pre-trained language model include:

[0035] Group chat content summary task;

[0036] The group chat content summarizing task refers to summarizing the group chat content of a preset time before the current moment of the group chat where the pre-trained language model is located.

[0037] In order to solve the above technical problems, the present invention also provides an interactive dialogue management device based on a language model, comprising:

[0038] A first receiving module, used for receiving a message to be sent from a client and identity information of a recipient of the message to be sent;

[0039] A first judgment module is used to judge whether the recipient of the message to be sent belongs to the pre-trained language model according to the identity information, and if so, trigger the first sending module; if not, trigger the third sending module;

[0040] The first sending module is used to send the message to be sent to the pre-trained language model;

[0041] A second sending module, used for pushing the processing result of the pre-trained language model on the message to be sent to the client;

[0042] The third sending module is used to send the message to be sent directly to the recipient.

[0043] In order to solve the above technical problems, the present invention also provides an interactive dialogue management device based on a language model, comprising:

[0044] Memory for storing computer programs;

[0045] A processor is used to implement the steps of the interactive dialogue management method based on the language model as described above when executing the computer program.

[0046] In order to solve the above technical problems, the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the interactive dialogue management method based on the language model as described above are implemented.

[0047] Beneficial effect: The present invention provides an interactive dialogue management method based on a language model. Considering that the pre-trained language model can be used in the form of a dialogue and the pre-trained language model can handle a variety of language processing tasks, the present invention can receive the message to be sent from the client and the identity information of the recipient of the message to be sent, and then judge whether the recipient of the message to be sent belongs to the pre-trained language model according to the identity information. If not, the message to be sent can be sent directly to the recipient. If so, the message to be sent can be sent to the pre-trained language model, and the processing result of the pre-trained language model for the message to be sent is pushed to the client, so that the user can use the pre-trained language model in the chat software in the form of a dialogue, which can meet the user's processing requirements for various language processing tasks and improve the user experience.

[0048] The present invention also provides an interactive dialogue management device, equipment and computer-readable storage medium based on a language model, which have the same beneficial effects as the above interactive dialogue management method based on a language model. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the relevant technologies and the drawings required for use in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0050] Figure 1 A flow chart of an interactive dialogue management method based on a language model provided by the present invention;

[0051] Figure 2 A schematic diagram of the structure of an interactive dialogue management system based on a language model provided by the present invention;

[0052] Figure 3A schematic diagram of the structure of an interactive dialogue management device based on a language model provided by the present invention;

[0053] Figure 4 A structural schematic diagram of a language model-based interactive dialogue management device provided by the present invention. DETAILED DESCRIPTION

[0054] The core of the present invention is to provide an interactive dialogue management method, device, equipment and computer-readable storage medium based on a language model. When it is determined that the recipient of the message to be sent belongs to a pre-trained language model, the message to be sent can be sent to the pre-trained language model, and the processing result of the pre-trained language model for the message to be sent can be pushed to the client, so that the user can use the pre-trained language model in the chat software through a dialogue form, which can meet the user's processing needs for various language processing tasks and improve the user experience.

[0055] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] Please refer to Figure 1 , Figure 1 A flow chart of an interactive dialogue management method based on a language model provided by the present invention, the interactive dialogue management method based on a language model comprises:

[0057] S101: receiving a message to be sent from a client and identity information of a recipient of the message to be sent;

[0058] Specifically, considering the technical problems in the above background technology, and considering that the pre-trained language model can be used in the form of a dialogue, and the pre-trained language model can handle a variety of language processing tasks, based on the above considerations, in the embodiment of the present invention, the pre-trained language model is intended to be used as a chat object of the user in the chat software, so that the user can use the pre-trained language model in the chat software through a dialogue. Since the chat software used in the embodiment of the present invention may be a conventional chat software, after the pre-trained language model is incorporated, the chat object may be a real user or a pre-trained language model. Therefore, when the user sends a message, it is first necessary to confirm the identity of the recipient of the message to be sent. Therefore, in this step, the message to be sent by the client and the identity information of the recipient of the message to be sent can be first received.

[0059] S102: judging whether the recipient of the message to be sent belongs to the pre-trained language model according to the identity information;

[0060] Specifically, the identity information may include a variety of specific contents, such as an identity identifier of a real user, or an identifier of a pre-trained language model, etc., which is not limited in the embodiment of the present invention.

[0061] Specifically, considering that the pre-trained language model is a special existence among chat objects, in order to distinguish it from real users, in an embodiment of the present invention, it is possible to determine whether the recipient of the message to be sent belongs to the pre-trained language model based on the identity information, so as to trigger subsequent different actions through its judgment result.

[0062] Specifically, the specific process of "determining whether the recipient of the message to be sent belongs to the pre-trained language model based on the identity information" can be determined according to the specific content of the identity information, for example, it may include: determining whether the identity information belongs to the identifier of the pre-trained language model, etc., and the embodiments of the present invention are not limited here.

[0063] Among them, the pre-trained language model can be of various types, for example, it can be various reinforcement learning language models, such as ChatGpt (Chat Generative Pre-trained Transformer) and Qwen2_7b_intruct, a reinforcement learning language model, etc., which is not limited in the embodiments of the present invention.

[0064] S103: If yes, sending the message to be sent to the pre-trained language model;

[0065] Specifically, when the recipient belongs to a pre-trained language model, the message to be sent can be sent to the pre-trained language model so that the pre-trained language model processes the message to be sent. "Sending the message to be sent to the pre-trained language model" can be regarded as inputting the message to be sent into the pre-trained language model, and the message to be sent can be pre-processed before input so that the message to be sent meets the input requirements of the pre-trained language model.

[0066] To better illustrate the embodiments of the present invention, please refer to Figure 2 , Figure 2 A structural schematic diagram of a language model-based interactive dialogue management system provided by the present invention, wherein the server can manage between the client and the pre-trained language model, and through the content assembly module, "messages to be sent from the client to the pre-trained language model" or "content returned to the client by the pre-trained language model" can be encapsulated and formatted, while the communication interface is responsible for communicating with the pre-trained language model. When there are multiple pre-trained language models, the communication interface can also realize communication with multiple pre-trained language models.

[0067] The interaction mode between the client and the server may be of various types, such as websocket (a protocol for full-duplex communication on a single TCP (Transmission Control Protocol) connection), etc., which is not limited in the embodiments of the present invention.

[0068] S104: Pushing the processing result of the pre-trained language model for the message to be sent to the client;

[0069] Specifically, after receiving the message to be sent, the pre-trained language model will process the content therein and feed back the processing results to the server. The server can push the processing results of the pre-trained language model for the message to be sent to the client's chat interface, so that the user can receive the processing results through the "chat interface with the pre-trained language model".

[0070] S105: If it does not belong to the specified group, the message to be sent is directly sent to the recipient.

[0071] Specifically, when the recipient of the message to be sent does not belong to the pre-trained language model, the message to be sent can be sent directly to the recipient like a regular chat.

[0072] The present invention provides an interactive dialogue management method based on a language model. Considering that a pre-trained language model can be used in the form of a dialogue and that the pre-trained language model can handle a variety of language processing tasks, the present invention can receive a message to be sent from a client and identity information of a recipient of the message to be sent, and then determine whether the recipient of the message to be sent belongs to the pre-trained language model based on the identity information. If not, the message to be sent can be directly sent to the recipient. If so, the message to be sent can be sent to the pre-trained language model, and the processing result of the pre-trained language model on the message to be sent is pushed to the client, so that the user can use the pre-trained language model in the chat software in the form of a dialogue, which can meet the processing requirements of the user for various language processing tasks and improve the user experience.

[0073] Based on the above embodiments:

[0074] As an optional embodiment, receiving the message to be sent from the client and the identity information of the recipient of the message to be sent includes:

[0075] Receiving a message to be sent from a client through a first channel or a second channel, wherein the first channel refers to a private chat channel and the second channel refers to a group chat channel;

[0076] If the message to be sent comes from a private chat channel, determine the identity information of the recipient of the message to be sent;

[0077] If the message to be sent originates from a group chat channel and the client specifies a recipient of the message to be sent, the identity information of the recipient of the message to be sent is determined, and the message to be sent and the identity information of the recipient are pushed to the group chat interface;

[0078] If the message to be sent comes from the group chat channel and the client does not specify the recipient of the message to be sent, the message to be sent will be pushed to the group chat interface.

[0079] Specifically, considering that in both private chat scenarios and group chat scenarios of the chat software, interactions with the pre-trained language model can be carried out in the form of chat, in the private chat scenario, the user can directly interact with the pre-trained language model, and in the group chat scenario, the pre-trained language model can be used as a group member, and the user can interact with the "pre-trained language model" as a group member. Therefore, in an embodiment of the present invention, the client's message to be sent can be received through the first channel or the second channel. If the message to be sent comes from the private chat channel, the identity information of the recipient of the message to be sent can be determined. If the message to be sent comes from the group chat channel and the client specifies the recipient of the message to be sent, the identity information of the recipient of the message to be sent is determined, and the message to be sent and its identity information of the recipient are pushed to the group chat interface; if the message to be sent comes from the group chat channel and the client does not specify the recipient of the message to be sent, the message to be sent is pushed to the group chat interface; the use of the pre-trained language model in private chat and group chat scenarios is realized.

[0080] As an optional embodiment, the message to be sent to which the recipient belongs to the pre-trained language model includes:

[0081] Data modality conversion tasks;

[0082] The data modality conversion task refers to converting the data of the first modality given by the user into the data of the specified second modality, where the first modality and the second modality are respectively one of text, picture and video.

[0083] Specifically, data modality conversion is a common demand, and the pre-trained language model can complete this task efficiently and accurately. Therefore, the message to be sent in the embodiment of the present invention may include a data modality conversion task. For example, the data of the first modality given by the user is text data, and the specified second modality is a picture. The embodiment of the present invention is not limited here.

[0084] As an optional embodiment, the message to be sent to which the recipient belongs to the pre-trained language model further includes:

[0085] Function call type tasks, function call means: the function module deployed in the server needs to be called;

[0086] Sending the message to be sent to the pre-trained language model includes:

[0087] Sending the function call type task to the pre-trained language model so that the pre-trained language model completes the function call type task by calling the function module deployed in the server;

[0088] Function call tasks include:

[0089] Task creation task, task creation task includes task time, task content and task executor, and the functional module corresponding to task creation task is office system.

[0090] Specifically, considering that the functional modules pre-deployed on the server side can also be connected to the pre-trained language model for use, so that users can achieve more functions through interaction with the pre-trained language model, the message to be sent in the embodiment of the present invention can also include function call-type tasks, and the functional modules can include multiple types. In addition to the above-mentioned office system, the functional modules can also include weather queries, etc., which are not limited in the embodiment of the present invention.

[0091] Among them, the function call type task may include various types, such as weather query tasks and the above-mentioned task creation tasks, etc., which are not limited in the embodiment of the present invention.

[0092] As an optional embodiment, the interactive dialogue management method based on the language model further includes:

[0093] In response to a model adding instruction sent by the client, a preset language model list is pushed to the client, wherein the model adding instruction includes a first instruction to add a pre-trained language model as a friend and a second instruction to add the pre-trained language model as a group member of a designated group chat, and the language model list includes a plurality of pre-trained language models;

[0094] In response to a selection instruction from the client, determining a target language model selected by the client from the language model list;

[0095] If the model adding instruction is the first instruction, the target language model is used as a friend of the client;

[0096] If the model adding instruction is the second instruction, the target language model is used as a group member of the specified group chat.

[0097] Specifically, in order to facilitate users to use and manage pre-trained language models in chat software, the embodiment of the present invention first provides multiple pre-trained language models, and provides users with the function of adding pre-trained language models as friends or adding pre-trained language models as group members. Of course, the pre-trained language models can also be deleted as friends or "removed from group members", etc., which is not limited in the embodiment of the present invention.

[0098] Specifically, in an embodiment of the present invention, in response to a model adding instruction sent by the client, a preset language model list can be pushed to the client, and in response to a selection instruction of the client, a target language model selected by the client from the language model list can be determined. If the model adding instruction is a first instruction, the target language model is used as a friend of the client; if the model adding instruction is a second instruction, the target language model is used as a group member of a specified group chat.

[0099] Specifically, it is worth mentioning that a user can add multiple pre-trained language models as friends, and can also add multiple pre-trained language models as group members, which is not limited in the embodiment of the present invention.

[0100] As an optional embodiment, the language model list includes several third-party language models.

[0101] Specifically, the use of third-party language models can reduce model training workload and improve the accuracy of language models.

[0102] The specific type of the third-party language model can be set independently, and the embodiment of the present invention does not limit it here.

[0103] Of course, in addition to the third-party language model, the language model list may also include a self-trained pre-trained language model, which is not limited in the embodiment of the present invention.

[0104] As an optional embodiment, sending the message to be sent to the pre-trained language model includes:

[0105] Determine whether the message to be sent is within the processing capability of the pre-trained language model;

[0106] If yes, the message to be sent is sent to the pre-trained language model;

[0107] If not, the client is prompted that the message content cannot be processed.

[0108] Specifically, considering the differences in the capabilities of different language models, the message to be sent by the user to the pre-trained language model may not be processed by the pre-trained language model. Therefore, in an embodiment of the present invention, it can be determined whether the message to be sent is within the processing capability of the pre-trained language model. If not, the client can be prompted that the message content cannot be processed, or the client can be prompted that it is within the processing capability range of the pre-trained language model so that the user can send the message content that can be processed by the pre-trained language model.

[0109] As an optional embodiment, the message to be sent received through the group chat channel and the recipient belongs to the pre-trained language model includes:

[0110] Group chat content summary task;

[0111] Among them, the group chat content summary task refers to: summarizing the group chat content of a preset time before the current moment of the group chat where the pre-trained language model is located.

[0112] Specifically, considering that in group chats (especially work-related group chats), there is a need to summarize the content of group chats within a period of time, and the language model itself can complete the content summary work, the message to be sent in the embodiment of the present invention may include a group chat content summary task. After receiving the task, the pre-trained language model can summarize the group chat content of a preset length before the current moment of the group chat where the pre-trained language model is located, which is conducive to improving the user's work efficiency and further enhancing the user experience.

[0113] The preset duration may be set independently, for example, one week, etc., and the embodiment of the present invention does not limit this.

[0114] Please refer to Figure 3 , Figure 3 A schematic diagram of the structure of an interactive dialogue management device based on a language model provided by the present invention, the interactive dialogue management device based on a language model comprises:

[0115] The first receiving module 31 is used to receive the message to be sent from the client and the identity information of the recipient of the message to be sent;

[0116] A first judging module 32 is used to judge whether the recipient of the message to be sent belongs to the pre-trained language model according to the identity information, and if so, trigger the first sending module 33; if not, trigger the third sending module 35;

[0117] A first sending module 33, used to send the message to be sent to the pre-trained language model;

[0118] The second sending module 34 is used to push the processing result of the pre-trained language model for the message to be sent to the client;

[0119] The third sending module 35 is used to send the message to be sent directly to the recipient.

[0120] For an introduction to the language model-based interactive dialogue management device provided in an embodiment of the present invention, please refer to the aforementioned embodiment of the language model-based interactive dialogue management method, and the embodiment of the present invention will not be described in detail here.

[0121] Please refer to Figure 4 , Figure 4 A schematic diagram of the structure of an interactive dialogue management device based on a language model provided by the present invention, the interactive dialogue management device based on a language model comprises:

[0122] A memory 41, used for storing computer programs;

[0123] The processor 42 is used to implement the steps of the interactive dialogue management method based on the language model in the above-mentioned embodiment when executing the computer program.

[0124] For an introduction to the language model-based interactive dialogue management device provided in an embodiment of the present invention, please refer to the aforementioned embodiment of the language model-based interactive dialogue management method, and the embodiment of the present invention will not be described in detail here.

[0125] The present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the interactive dialogue management method based on the language model in the aforementioned embodiment are implemented.

[0126] For an introduction to the computer-readable storage medium provided in an embodiment of the present invention, please refer to the aforementioned embodiment of the interactive dialogue management method based on the language model, and the embodiment of the present invention will not be described in detail here.

[0127] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same and similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description. It should also be noted that in this specification, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. Moreover, the term "include", "comprise" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements includes not only those elements, but also includes other elements that are not explicitly listed, or also includes elements inherent to such process, method, article or equipment. In the absence of more restrictions, the elements defined by the sentence "including one..." do not exclude the existence of other identical elements in the process, method, article or equipment including the element.

[0128] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An interactive dialogue management method based on a language model, characterized in that: Applied to the server, including: Receive a message to be sent from a client and identity information of a recipient of the message to be sent; Determining whether the recipient of the message to be sent belongs to the pre-trained language model according to the identity information; If yes, sending the message to be sent to the pre-trained language model; Pushing the processing result of the pre-trained language model on the message to be sent to the client; If not, the message to be sent is directly sent to the recipient.

2. The interactive dialogue management method based on language model according to claim 1, characterized in that: The message to be sent from the receiving client and the identity information of the recipient of the message to be sent include: Receiving a message to be sent from a client through a first channel or a second channel, wherein the first channel refers to a private chat channel and the second channel refers to a group chat channel; If the message to be sent comes from a private chat channel, determine the identity information of the recipient of the message to be sent; If the message to be sent originates from a group chat channel and the client specifies a recipient of the message to be sent, the identity information of the recipient of the message to be sent is determined, and the message to be sent and the identity information of the recipient are pushed to the group chat interface; If the message to be sent originates from a group chat channel and the client does not specify a recipient of the message to be sent, the message to be sent is pushed to the group chat interface.

3. The interactive dialogue management method based on language model according to claim 2, characterized in that: The messages to be sent to the recipients of the pre-trained language model include: Data modality conversion tasks; The data modality conversion task refers to: converting the data of the first modality given by the user into the data of the specified second modality, wherein the first modality and the second modality are respectively one of text, picture and video.

4. The interactive dialogue management method based on language model according to claim 3, characterized in that: Messages to be sent whose recipients belong to the pre-trained language model also include: Function call type tasks, where the function call means: the function module deployed in the server needs to be called; Sending the message to be sent to the pre-trained language model includes: Sending the function call type task to the pre-trained language model so that the pre-trained language model completes the function call type task by calling the function module deployed in the server; The function call type tasks include: A task creation task includes a task time, task content and a task executor, and the functional module corresponding to the task creation task is an office system.

5. The interactive dialogue management method based on language model according to claim 2, characterized in that: The interactive dialogue management method based on the language model also includes: In response to a model adding instruction sent by a client, a preset language model list is pushed to the client, wherein the model adding instruction includes a first instruction to add a pre-trained language model as a friend and a second instruction to add the pre-trained language model as a group member of a designated group chat, and the language model list includes a plurality of pre-trained language models; In response to a selection instruction from a client, determining a target language model selected by the client from the language model list; If the model adding instruction is the first instruction, the target language model is used as a friend of the client; If the model adding instruction is the second instruction, the target language model is used as a group member of the designated group chat.

6. The interactive dialogue management method based on language model according to claim 1, characterized in that: Sending the message to be sent to the pre-trained language model includes: Determining whether the message to be sent is within the processing capability of the pre-trained language model; If yes, sending the message to be sent to the pre-trained language model; If not, the client is prompted that the message content cannot be processed.

7. The interactive dialogue management method based on a language model according to any one of claims 2 to 6, characterized in that: Messages to be sent through the group chat channel and whose recipients belong to the pre-trained language model include: Group chat content summary task; The group chat content summarizing task refers to summarizing the group chat content of a preset time before the current moment of the group chat where the pre-trained language model is located.

8. An interactive dialogue management device based on a language model, characterized in that: include: A first receiving module, used for receiving a message to be sent from a client and identity information of a recipient of the message to be sent; A first judgment module is used to judge whether the recipient of the message to be sent belongs to the pre-trained language model according to the identity information, and if so, trigger the first sending module; if not, trigger the third sending module; The first sending module is used to send the message to be sent to the pre-trained language model; A second sending module, used for pushing the processing result of the pre-trained language model on the message to be sent to the client; The third sending module is used to send the message to be sent directly to the recipient.

9. An interactive dialogue management device based on a language model, characterized in that: include: Memory for storing computer programs; A processor, configured to implement the steps of the language model-based interactive dialogue management method as claimed in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the language model-based interactive dialogue management method according to any one of claims 1 to 7 are implemented.

Images