Interaction method and device, equipment and storage medium

Abstract

The embodiment of the invention relates to a method and device for interaction, equipment and a storage medium. The method comprises the steps of determining a target machine learning model associated with a predetermined operation from a group of first machine learning models in response to the detected predetermined operation of a user for a session window; switching the configuration parameters of the session window into configuration parameters related to the target machine learning model; based on configuration parameters related to the target machine learning model, processing query content input in the session window by the user into target information; and at least based on the target information, using the target machine learning model to determine reply content for the query content. In this way, a user can easily switch to different first machine learning models in the same chat window, which not only improves the convenience of interaction, but also significantly improves the overall experience of the user.

Description

Technical Field

[0001] Example embodiments of the present disclosure generally relate to the field of computer technology, and more particularly, to a method, apparatus, device, and storage medium for interaction. Background Art

[0002] Chat conversation technology based on machine learning models has been widely used. With its highly intelligent and automated features, this technology has greatly improved the response speed and efficiency of services and reduced the operating costs of enterprises. With the advancement of machine learning and artificial intelligence technology, chat conversation technology has been continuously iterated and optimized, gradually evolving from the initial simple dialogue system to one that can handle complex contexts and understand diverse user needs. Summary of the invention

[0003] In a first aspect of the present disclosure, a method for interaction is provided. The method includes: in response to detecting a predetermined operation of a user on a conversation window, determining a target machine learning model associated with the predetermined operation from a set of first machine learning models; switching the configuration parameters of the conversation window to configuration parameters related to the target machine learning model; based on the configuration parameters related to the target machine learning model, processing the query content input by the user in the conversation window into target information; and determining the reply content for the query content using the target machine learning model based on at least the target information.

[0004] In a second aspect of the present disclosure, a device for interaction is provided. The device includes: a detection module, configured to determine a target machine learning model associated with a predetermined operation from a set of first machine learning models in response to detecting a predetermined operation of a user on a conversation window; a configuration parameter switching module, configured to switch the configuration parameters of the conversation window to configuration parameters related to the target machine learning model; an information processing module, configured to process the query content input by the user in the conversation window into target information based on the configuration parameters related to the target machine learning model; and a reply content determination module, configured to determine the reply content for the query content using the target machine learning model based at least on the target information.

[0005] In a third aspect of the present disclosure, an electronic device is provided. The device includes at least one processing unit; and at least one memory, the at least one memory is coupled to the at least one processing unit and stores instructions for execution by the at least one processing unit. When the instructions are executed by the at least one processing unit, the device executes the method of the first aspect.

[0006] In a fourth aspect of the present disclosure, a computer-readable storage medium is provided, wherein a computer program is stored on the computer-readable storage medium, and the computer program can be executed by a processor to implement the method of the first aspect.

[0007] In a fifth aspect of the present disclosure, a computer program product is provided, which includes computer executable instructions, and when the instructions are executed by a processor, the method according to the first aspect of the present disclosure is implemented.

[0008] It should be understood that the contents described in this content section are not intended to limit the key features or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, wherein:

[0010] Figure 1 A schematic diagram illustrating an example environment according to some embodiments of the present disclosure is shown;

[0011] Figure 2 A flowchart showing an example process of a method for interaction according to some embodiments of the present disclosure;

[0012] Figure 3 A schematic diagram showing an example of a session window according to some embodiments of the present disclosure;

[0013] Figure 4 A schematic diagram showing an example of the interaction of a session window, a target control, and a target machine learning model according to some embodiments of the present disclosure;

[0014] Figure 5 A schematic structural block diagram of an apparatus for interaction according to some embodiments of the present disclosure is shown; and

[0015] Figure 6 A block diagram of an electronic device is shown in which one or more embodiments of the present disclosure may be implemented. DETAILED DESCRIPTION

[0016] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0017] It should be noted that the titles of any sections / subsections provided herein are not restrictive. Various embodiments are described throughout this article, and any type of embodiment may be included under any section / subsection. In addition, the embodiments described in any section / subsection may be combined in any manner with any other embodiments described in the same section / subsection and / or different sections / subsections.

[0018] In the description of the embodiments of the present disclosure, the term "including" and similar terms should be understood as open inclusion, that is, "including but not limited to". The term "based on" should be understood as "based at least in part on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The term "some embodiments" should be understood as "at least some embodiments". Other explicit and implicit definitions may be included below. The terms "first", "second", etc. may refer to different or the same objects. Other explicit and implicit definitions may be included below.

[0019] The embodiments of the present disclosure may involve user data, data acquisition and / or use, etc. These aspects are subject to the corresponding laws, regulations and relevant provisions. In the embodiments of the present disclosure, all data collection, acquisition, processing, processing, forwarding, use, etc. are carried out on the premise that the user knows and confirms. Accordingly, when implementing each embodiment of the present disclosure, the type, scope of use, usage scenario, etc. of the data or information that may be involved should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with the relevant laws and regulations. The specific notification and / or authorization method can vary according to the actual situation and application scenario, and the scope of the present disclosure is not limited in this respect.

[0020] In this specification and the embodiments, if personal information processing is involved, it will be processed on the premise of having a legal basis (such as obtaining the consent of the subject of personal information, or it is necessary to perform a contract, etc.), and will only be processed within the scope of regulations or agreements. If a user refuses to process personal information other than the necessary information for basic functions, it will not affect the user's use of basic functions.

[0021] As briefly described above, at present, chat session technology based on machine learning models is able to handle complex contexts and understand diverse user needs. Users can choose different machine learning models to interact according to different needs. When users want to interact with different machine learning models, they usually need to access these models through different session windows. Simply put, each machine learning model has its own specific session window, and users need to interact with the machine learning model through this session window. This usually means that if users want to interact with different machine learning models, they need to open multiple different session windows, which not only takes up more screen space, but also complicates the interaction process, which is not conducive to user experience.

[0022] In view of this, an embodiment of the present disclosure provides a scheme for interaction. According to the scheme, first, in response to detecting a predetermined operation of a user on a conversation window, a target machine learning model associated with the predetermined operation is determined from a set of first machine learning models. Then, the configuration parameters of the conversation window are switched to configuration parameters related to the target machine learning model. Subsequently, based on the configuration parameters related to the target machine learning model, the query content input by the user in the conversation window is processed into target information. Thereafter, based at least on the target information, the target machine learning model is used to determine the reply content for the query content.

[0023] It will be more clearly understood through the description below that the core purpose of this solution is to simplify the interaction process between users and multiple machine learning models. Specifically, when a user performs a predetermined operation on a conversation window, this solution determines a target machine learning model associated with the operation from a set of first machine learning models. Subsequently, the configuration parameters of the conversation window will automatically switch to the configuration related to the target model. In this way, the query content entered by the user in the conversation window can be processed into target information adapted to the target machine learning model based on the new configuration parameters. Finally, this solution uses the target machine learning model to generate reply content for the query content. In this way, users do not need to open multiple chat windows to interact with different first machine learning models, thereby greatly simplifying the interaction process. In other words, in this solution, users can easily switch to different first machine learning models in the same chat window through simple operations (such as clicking or selecting operations, etc.). This not only improves the convenience of interaction, but also significantly improves the overall user experience.

[0024] Various example implementations of the solution will be described in detail below in conjunction with the accompanying drawings.

[0025] Figure 1 A schematic diagram of an example environment 100 is shown according to some embodiments of the present disclosure. Figure 1 , the example environment 100 may include an electronic device 120. As an example, the electronic device 120 of the embodiment of the present disclosure may be applied to an artificial intelligence generated content (AIGC) scenario.

[0026] In the example environment 100, an application 130 for interacting with a set of first machine learning models 110 is installed in the electronic device 120. The set of first machine learning models 110 may include N first machine learning models 111-1, 111-2, ..., 111-N, where N is a positive integer. For clarity, the N first machine learning models 111-1, 111-2, ..., 111-N are collectively referred to as first machine learning models 111 below.

[0027] As an example, the electronic device 120 may present an interface 150 of the application 130, in which a conversation window 151 is presented. The user 140 may select the first machine learning model 111 that he wants to use in the conversation window 151. The electronic device 120 may switch the configuration parameters of the conversation window 151 based on the first machine learning model 111 selected by the user 140. Then, the electronic device 120 may process the query content entered by the user 140 in the conversation window 151 based on the switched configuration parameters, and send the processed query content to the first machine learning model 111 selected by the user. Then, after receiving the processed query content sent by the electronic device 120, the first machine learning model 111 processes the query content to generate a reply content for the query content entered by the user 140, and returns it to the electronic device 120. Finally, the electronic device 120 presents the obtained reply content to the user 140 to achieve interaction with the user 140.

[0028] In some embodiments, the electronic device 120 can be any type of mobile terminal, fixed terminal or portable terminal, including mobile phones, desktop computers, laptop computers, notebook computers, netbook computers, tablet computers, media computers, multimedia tablets, personal communication systems (PCS) devices, personal navigation devices, personal digital assistants (PDAs), audio / video players, digital cameras / camcorders, positioning devices, television receivers, radio broadcast receivers, e-book devices, gaming devices, or any combination of the foregoing, including accessories and peripherals of these devices or any combination thereof. In some embodiments, the terminal device 110 can also support any type of interface for the user (such as "wearable" circuits, etc.).

[0029] Multiple first machine learning models 111-1, 111-2, ..., 111-N can be set on one or more service devices, which can be independent physical servers, or server clusters or distributed systems composed of multiple physical servers, or cloud servers that provide basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content distribution networks, and big data and artificial intelligence platforms. These service devices can include, for example, computing systems / servers, such as mainframes, edge computing nodes, computing devices in cloud environments, and the like. These service devices can provide background services for applications 130 that support content presentation in electronic devices 120.

[0030] A communication connection may be established between the electronic device 120 and the service device. The communication connection may be established in a wired manner or a wireless manner. The communication connection may include, but is not limited to, a Bluetooth connection, a mobile network connection, a universal serial bus connection, a wireless fidelity connection, etc., and the embodiments of the present disclosure are not limited in this respect. In the embodiments of the present disclosure, the electronic device 120 and the service device may implement signaling interaction through the communication connection between the two.

[0031] It should be understood that the structure and function of the various elements in the environment 100 are described for exemplary purposes only and do not imply any limitation on the scope of the present disclosure.

[0032] Figure 2 A flowchart of an example process 200 of a method for interaction according to some embodiments of the present disclosure is shown. The process 200 may be implemented at the electronic device 120. Figure 3 A schematic diagram showing an example 300 of a session window 151 according to some embodiments of the present disclosure is shown. Figure 4 A schematic diagram of an example 400 of the interaction between a session window 151, a target control, and a target machine learning model according to some embodiments of the present disclosure is shown. Figure 3 and Figure 4 ,right Figure 2 Process 200 is shown for illustration.

[0033] In box 210, in response to detecting a predetermined operation of the user 140 on the session window 151, the electronic device 120 determines a target machine learning model associated with the predetermined operation from a set of first machine learning models 110.

[0034] In some embodiments, each first machine learning model 111 in a group of first machine learning models 110 can process the query content input by the user 140, but these first machine learning models 111 may tend to perform different tasks, such as image recognition, speech recognition, natural language processing and other tasks. For two first machine learning models 111 that can perform the same task, the structures, parameters, algorithms and training sets used in training of the two first machine learning models 111 are not the same. Therefore, for the same input content, the outputs given by the two first machine learning models 111 will also be different. It should be noted that, according to actual needs, the first machine learning model 111 can perform more tasks, and the embodiments of the present disclosure do not limit this.

[0035] Reference Figure 3 , M tabs 152-1, 152-2, ..., 152-M can be presented in the session window 151, where M is a positive integer. For clarity, the M tabs 152-1, 152-2, ..., 152-M are collectively referred to as tabs 152 below. Each tab 152 corresponds to a first machine learning model 111, and different tabs 152 correspond to different first machine learning models 111. These tabs 152 provide an intuitive interface for the user 140 to select different first machine learning models 111. As an example, the target machine learning model is a first machine learning model 111 selected by the user 140 from a group of first machine learning models 110.

[0036] It should be noted that, according to actual needs, more tabs 152 can be presented in the conversation window 151 to meet different needs. And the arrangement of these tabs 152 is not limited to Figure 3 The method shown can be flexibly adjusted according to actual needs.

[0037] In some embodiments, the electronic device 120 can determine a target machine learning model from a set of first machine learning models 110 based on at least two methods.

[0038] In the first manner, the electronic device 120 determines a first machine learning model 111 associated with the selected object as a target machine learning model from a group of first machine learning models 110 in response to detecting a selection by the user 140 of at least one tab 152 presented in the session window 151.

[0039] In some embodiments, each tab 152 may be presented as a label or icon representing the first machine learning model 111 or function associated with the tab 152. The user 140 may select the first machine learning model 111 they want to use by clicking or touching the tab 152.

[0040] In some embodiments, when the user 140 selects a tab 152 in the conversation window 151, the electronic device 120 can detect this action with the help of the conversation window 151. Once the user 140 selects the tab 152, the electronic device 120 determines the first machine learning model 111 selected by the user based on the selected object (i.e., the selected tab 152), and thus uses the first machine learning model 111 as the target machine learning model.

[0041] In some embodiments, the electronic device 120 may determine the first machine learning model 111 selected by the user based on the association between the tab 152 and the first machine learning model 111. This process may be completed through a pre-set mapping relationship. As an example, each tab 152 may correspond to an operation value, and each operation value is associated with a first machine learning model 111. After the user 140 selects a tab 152, the session window 151 may obtain the corresponding operation value based on the selected tab 152. Then, the electronic device 120 may determine the first machine learning model 111 selected by the user from a group of first machine learning models 110 based on the operation value.

[0042] In this way, the user 140 is able to intuitively select the first machine learning model 111 they need.

[0043] In the second method, in response to detecting that the user 140 enters content in the input box 155 of the session window 151, the electronic device 120 uses a third machine learning model to determine a first machine learning model 111 related to the input content from a group of first machine learning models 110 as a target machine learning model.

[0044] In some embodiments, the electronic device 120 is able to identify and determine, with the help of a third machine learning model, a first machine learning model 111 that is most relevant to the input content of the user 140 from a group of first machine learning models 110. The selected first machine learning model 111 is then used as a target machine learning model for further processing the input of the user 140.

[0045] As an example, the user 140 can input the content they want to query or process in the input box 155 of the conversation window 151. The content can be in multimedia forms such as text, images, audio, etc. The third machine learning model can be trained to recognize the features of the input content, such as semantics and themes, and recommend the first machine learning model 111 that is most suitable for processing the input based on these features and historical data.

[0046] Compared with the first method, the second method saves the user 140 from selecting the tab 152 , thereby simplifying the thinking process of the user 140 .

[0047] As an example, in order to ensure that the second method can determine an accurate target machine learning model for the user 140, after the third machine learning model determines the first machine learning model 111 that is most relevant to the content input by the user 140, the first machine learning model 111 can be presented to the user 140, and a confirmation request can be issued to the user 140. After the user 140 confirms that he agrees to use the first machine learning model 111, the electronic device 120 uses the first machine learning model 111 as the target machine learning model. Otherwise, the electronic device 120 can redetermine the first machine learning model 111 that is most relevant to the content input by the user 140 with the help of the third machine learning model, or prompt the user 140 to determine the target machine learning model through the first method described above.

[0048] As an example, after the user 140 confirms and agrees to use the first machine learning model 111, the electronic device may determine an operation value based on the first machine learning model 111, and the operation value is used to indicate which first machine learning model 111 the user has specifically selected in subsequent steps.

[0049] It should be noted that in the embodiments of the present disclosure, the first method and the second method can be used alone or in combination. When used in combination, different weights can be assigned to the two methods according to actual needs to facilitate more effective integration of the results of the two methods.

[0050] In box 220, the electronic device 120 switches the configuration parameters of the session window 151 to configuration parameters related to the target machine learning model.

[0051] In some embodiments, after the electronic device 120 determines the target machine learning model, in order to make the session window 151 better adapt to the functions and requirements of the target machine learning model, the electronic device 120 adjusts the configuration parameters of the session window 151. These configuration parameters may include, for example, the layout of the window, display elements, and interaction methods. It should be noted that, according to actual needs, the configuration parameters may include more content, and the embodiments of the present disclosure do not limit this.

[0052] As an example, the electronic device 120 can sense the predetermined operation of the user 140 (e.g., the selection of the tab 152 by the user 140) by means of the conversation window 151, and then determine the corresponding operation value (e.g., the source value of the tab 152) according to the sensed predetermined operation. As described above, this operation value can establish a corresponding relationship with a first machine learning model 111 in a set of first machine learning models 110, so that the operation value can indicate the first machine learning model 111 (i.e., the target machine learning model) selected by the user 140 in subsequent steps.

[0053] In some embodiments, the electronic device 120 uses a target control corresponding to the window frame of the session window 151 to switch the configuration parameters of the session window 151 to configuration parameters related to the target machine learning model.

[0054] As an example, the session window 151 and the window frame can be constructed based on the Document Object Model (DOM). DOM is a programming interface for representing and processing HTML or XML documents, which represents elements such as the session window 151 and the window frame as nodes in a tree structure. As an example, the window frame can be the basic structure of the session window 151. The window frame can define the layout, appearance, and behavior of the session window 151. For example, a session window 151 may be represented by a specific element (such as DOM) in an HTML document. ), which can be considered as the window frame of the session window 151. The target control can be one or more specific DOM nodes within the window frame, and can be conveniently accessed and styled through JavaScript and CSS.

[0055] As described above, the traditional session window management mode requires users to switch session windows at the same time when switching machine learning models. This frequent window switching is not only cumbersome, but also interrupts the user's thinking continuity. In the embodiment of the present disclosure, with the help of the target control of the window frame, the user does not need to be subject to this restriction. As a management module of the session window 151, the window frame can uniformly manage and schedule the configuration parameters of the session window, so that users can interact smoothly with multiple machine learning models without leaving the current session window, thereby maintaining the continuity and efficiency of work.

[0056] In some embodiments, in block 410 , the session window 151 sends a switching instruction to the target control when the operation value is determined. In block 420 , in response to receiving the switching instruction, the electronic device 120 obtains the current configuration parameters of the session window 151 using the target control.

[0057] In box 430, the target control adjusts the current configuration parameters of the session window based on the information transmission method corresponding to the target machine learning model. For example, if the target machine learning model needs to receive data in JSON format, the target control will adjust the configuration parameters to ensure that the form, input box and other controls in the session window 151 can generate data in JSON format.

[0058] In addition, the adjustment of configuration parameters may also include the adjustment of window size, the adjustment of layout, the addition or removal of specific UI elements (such as buttons, text boxes, etc.), the update of styles and themes, etc. All these adjustments are to make the session window 151 more in line with the requirements of the first machine learning model 111 selected by the user 140.

[0059] It should be noted that, according to actual needs, the configuration parameters may include more content, and the session window 151 may be adjusted in more ways, which are not limited in the embodiments of the present disclosure.

[0060] In box 440, the target control sends the adjusted configuration parameters to the session window 151 to switch the configuration parameters of the session window 151 to the configuration parameters related to the target machine learning model.

[0061] As an example, after receiving the adjusted configuration parameters, the session window 151 applies them to its own configuration, so that the session window 151 can process the query content according to the requirements of the target machine learning model.

[0062] In some embodiments, in addition to the switching tasks of the configuration parameters described above, the target control is also configured to be presented as an outer frame 153 and / or a switching sub-control 154 near the session window 151, and the target control can be switched between the session window 151 and other windows through the user's operation of the switching sub-control 154.

[0063] It should be noted that the embodiments of the present disclosure do not limit the content of other windows. For example, other windows may also be session windows 151, but the difference from the currently displayed session window 151 is that the tabs 152 presented in other windows may be different. That is, the embodiments of the present disclosure may present different combinations of tabs 152 (or first machine learning models 111) in different session windows 151.

[0064] In some embodiments, the target control may also be configured to manage the content presented in the session window 151. As an example, the target control may execute tasks such as login, authentication, and authorization initiated by the user 140 in the session window 151 for a first machine learning model 111.

[0065] In some embodiments, in addition to displaying the above-mentioned tab 152, the conversation window 151 may also display "Reset conversation", "Send", date in the form of "yyyy / mm / dd", and time in the form of "aa:bb:cc", etc. These contents may be managed by the target control to implement editing functions for these contents.

[0066] In some embodiments, the electronic device 120 locally stores configuration parameters of at least one first machine learning model 111 in a set of first machine learning models 110 to obtain a set of preloaded configuration parameters.

[0067] In some embodiments, if there are configuration parameters related to the target machine learning model in a set of preloaded configuration parameters, the electronic device 120 obtains the configuration parameters related to the target machine learning model from the set of preloaded configuration parameters. Then, the electronic device 120 switches the configuration parameters of the session window 151 to configuration parameters that match the target machine learning model. In some embodiments, if there are no configuration parameters related to the target machine learning model in a set of preloaded configuration parameters, the electronic device 120 can obtain the configuration parameters related to the target machine learning model based on a predetermined server or other remote location.

[0068] As an example, assume that a set of first machine learning models 110 includes a first machine learning model 111-1 capable of performing image recognition tasks, a first machine learning model 111-2 capable of performing speech recognition tasks, and a first machine learning model 111-M capable of performing natural language processing tasks. When the user 140 chooses to use the first machine learning model 111-1 capable of performing image recognition tasks, the electronic device 120 checks whether the preloaded configuration parameters of the first machine learning model 111-1 are stored locally. If so, the electronic device 120 quickly applies these configuration parameters to the session window 151, such as adjusting the window size, adding an image upload area, etc., so that the user 140 can immediately start using the image recognition function.

[0069] In this way, when there is a need to quickly switch to a first machine learning model 111, these preloaded configuration parameters can be used directly without re-downloading or generating from a server or other remote location.

[0070] In some embodiments, the electronic device 120 determines at least one first machine learning model 111 in a set of first machine learning models 110 as a preloading target based on the editable preloading configuration information. Then, the electronic device 120 locally stores the configuration parameters of the first machine learning model 111 determined as the preloading target in the set of first machine learning models 110 to obtain a set of preloading configuration parameters.

[0071] As an example, the preloading configuration information indicates which configuration parameters of the first machine learning model 111 should be preloaded. This preloading configuration information is editable, so the user 140 can adjust which first machine learning models 111 are considered as preloading targets according to their needs.

[0072] In some embodiments, the editable preloaded configuration information can be dynamically updated based on information such as the user's 140 usage and evaluation of the first machine learning model 111, so as to better meet the actual needs of the user 140.

[0073] In this way, preloading of configuration parameters of the low-usage first machine learning model 111 can be omitted, which is beneficial to reducing bandwidth occupancy.

[0074] In box 230, the electronic device 120 processes the query content entered by the user 140 in the conversation window 151 into target information based on the configuration parameters related to the target machine learning model.

[0075] In some embodiments, the electronic device 120 processes the query content into a form that matches the information transmission method based on configuration parameters related to the target machine learning model to obtain the target information.

[0076] In some embodiments, the processing of the query content may include steps such as packaging the data into a specific message format, adding necessary request headers or metadata, and selecting an appropriate communication protocol for transmission to ensure that the query content meets the input requirements of the target machine learning model. The processed target information will be used as input data for the target machine learning model. This process ensures the accurate delivery and effective processing of the query content, thereby supporting smooth interaction between the user and the target machine learning model.

[0077] It should be noted that, according to actual needs, the processing of the query content may also include more content, and the embodiments of the present disclosure do not limit this.

[0078] In box 240, based on at least the target information, a target machine learning model is used to determine the answer content for the query content. As an example, the query content can be determined by the electronic device 120 based on the content input by the user in the input box 155.

[0079] In some embodiments, the electronic device 120 does not directly transmit the query content in the session window 151 to the target machine learning model, but instead implements the scheduling of a group of first machine learning models 110 through a second machine learning model.

[0080] In some embodiments, the electronic device 120 sends the operation value and target information corresponding to the predetermined operation to the second machine learning model 460, wherein the operation value is configured to instruct the second machine learning model 460 to determine the target machine learning model as the processing object of the target information. Then, the electronic device 120 uses the second machine learning model 460 to send the target information to the target machine learning model. Subsequently, the electronic device 120 uses the second machine learning model 460 to obtain the reply content generated by the target machine learning model based on the target information for the query content.

[0081] In some embodiments, the operation value corresponding to the predetermined operation can be obtained in the manner described above. For example, in the process of the electronic device 120 sending the target information to the second machine learning model 460 through the session window 151, the session window 151 will simultaneously transmit the operation value determined based on the predetermined operation. This operation value actually indicates the first machine learning model 111 selected by the user 140. Therefore, the second machine learning model 460 can use this operation value to accurately identify the processing object of the target information, that is, the target machine learning model, from a group of first machine learning models 110, and send the target information to the model for processing.

[0082] In this way, the second machine learning model 460 acts as an intermediary between the session window 151 and a group of first machine learning models 110. The second machine learning model 460 can identify the content transmitted by the session window 151 and accurately route this information to the first machine learning model 111 selected by the user 140 from a group of first machine learning models 110. This design enables flexible scheduling of messages based on the same session window 151 between multiple first machine learning models 111. In addition, the second machine learning model 460 can also receive the reply content output by the target machine learning model and return the reply content to the session window 151, thereby ensuring the smoothness and efficiency of the entire interaction process.

[0083] In some embodiments, a group of first machine learning models 110 may be configured with a scheduling module (or model engine). The scheduling module is responsible for managing, scheduling, and optimizing the operation of a group of first machine learning models 110, thereby forming an efficient and collaborative working system.

[0084] As an example, the scheduling module can intelligently allocate computing resources and processing time based on system requirements, model characteristics, and real-time load conditions to ensure that each first machine learning model 111 in a group of first machine learning models 110 can be used reasonably.

[0085] As an example, when processing a complex task, a single first machine learning model 111 may not be able to meet all requirements. Therefore, the scheduling module is responsible for calling other related tools to work with the first machine learning model 111 to complete the task together.

[0086] In some embodiments, a second machine learning model 460 may be added to the scheduling module to work in conjunction with other parts of the scheduling module.

[0087] In some embodiments, after the second machine learning model 460 sends the target information to the target machine learning model, the target machine learning model may try to understand the query content of the user 140 based on the target information, so as to know the true intention of the user 140. Then, the target machine learning model retrieves relevant information from the internal knowledge base or external data source according to the determined true intention of the user 140. This information can be facts, statistics, rules, and previously learned knowledge, etc. Then, the target machine learning model can use the retrieved information and its internal reasoning mechanism to generate the reply content. This can include steps such as logical reasoning, context understanding, and language generation.

[0088] In some embodiments, after the reply content to the query content is determined, the reply content is presented in the conversation window 151 .

[0089] After obtaining the reply content for the query content returned by the target machine learning model, the second machine learning model 460 can pass the reply content to the session window 151. In the session window 151, the reply content can be rendered according to a preset format and presented to the user 140. This design ensures that the entire interaction process between the user 140 and the target machine learning model is carried out smoothly in the same session window 151. When the user 140 wants to use a different first machine learning model 111, switching can be achieved in the same session window 151. This seamless switching process greatly improves the user 140 experience, allowing the user 140 to communicate with various first machine learning models 111 easily and conveniently.

[0090] According to the various embodiments described above, it can be clearly understood that in an embodiment of the present disclosure, when the user 140 performs a predetermined operation on the conversation window 151, the embodiment of the present disclosure determines the target machine learning model associated with the operation from a set of first machine learning models 110. Subsequently, the configuration parameters of the conversation window 151 will automatically switch to the configuration related to the target model. In this way, the query content entered by the user 140 in the conversation window 151 can be processed into target information adapted to the target machine learning model based on the new configuration parameters. Finally, the embodiment of the present disclosure uses the target machine learning model to generate reply content for the query content. In this way, the user 140 does not need to open multiple chat windows to interact with different models, which greatly simplifies the interaction process. In other words, in an embodiment of the present disclosure, the user 140 can easily switch to different machine learning models in the same chat window through simple operations (such as clicking or selecting operations, etc.). This not only improves the convenience of interaction, but also significantly improves the overall experience of the user 140.

[0091] The embodiments of the present disclosure also provide corresponding devices for implementing the above methods or processes. Figure 5 A schematic structural block diagram of an apparatus 500 for interaction according to some embodiments of the present disclosure is shown. The apparatus 500 may be implemented as or included in the electronic device 120. Each module / component in the apparatus 500 may be implemented by hardware, software, firmware or any combination thereof.

[0092] Reference Figure 5 The device 500 includes a detection module 510, a configuration parameter switching module 520, an information processing module 530 and a reply content determination module 540.

[0093] In some embodiments, the detection module 510 is configured to determine a target machine learning model associated with a predetermined operation of the user on the conversation window from a set of first machine learning models in response to detecting the predetermined operation of the user on the conversation window. The configuration parameter switching module 520 is configured to switch the configuration parameters of the conversation window to configuration parameters related to the target machine learning model. The information processing module 530 is configured to process the query content input by the user in the conversation window into target information based on the configuration parameters related to the target machine learning model. The reply content determination module 540 is configured to determine the reply content for the query content using the target machine learning model based on at least the target information.

[0094] In some embodiments, the reply content determination module 540 is also configured to: send the operation value and target information corresponding to the predetermined operation to the second machine learning model, wherein the operation value is configured to instruct the second machine learning model to determine the target machine learning model as the processing object of the target information; use the second machine learning model to send the target information to the target machine learning model; and use the second machine learning model to obtain the reply content for the query content generated by the target machine learning model based on the target information.

[0095] In some embodiments, the configuration parameter switching module 520 is further configured to: utilize a target control corresponding to the window frame of the session window to switch the configuration parameters of the session window to configuration parameters related to the target machine learning model.

[0096] In some embodiments, the configuration parameter switching module 520 is further configured to: use the target control to obtain the current configuration parameters of the session window; adjust the current configuration parameters of the session window based on the information transmission method corresponding to the target machine learning model; and send the adjusted configuration parameters to the session window to switch the configuration parameters of the session window to configuration parameters related to the target machine learning model.

[0097] In some embodiments, the information processing module 530 is further configured to: based on configuration parameters related to the target machine learning model, process the query content into a form that matches the information transmission method to obtain target information.

[0098] In some embodiments, the target control is further configured to switch between the session window and other windows by presenting a switching sub-control near the session window.

[0099] In some embodiments, the detection module 510 is also configured to perform at least one of the following: in response to detecting a user selection of at least one tab presented in the session window, determining a first machine learning model associated with the selected object from a group of first machine learning models as a target machine learning model; or in response to detecting a user inputting content in the input box 155 of the session window, using a third machine learning model to determine a first machine learning model related to the input content from a group of first machine learning models as a target machine learning model.

[0100] In some embodiments, the device 500 further includes a preloading module configured to: locally store configuration parameters of at least one first machine learning model in a set of first machine learning models to obtain a set of preloaded configuration parameters. And the configuration parameter switching module 520 is further configured to: if there are configuration parameters related to the target machine learning model in the set of preloaded configuration parameters, obtain the configuration parameters related to the target machine learning model from the set of preloaded configuration parameters; and switch the configuration parameters of the session window to configuration parameters that match the target machine learning model.

[0101] In some embodiments, the preloading module is further configured to: determine at least one first machine learning model in a group of first machine learning models as a preloading target based on editable preloading configuration information; and locally store configuration parameters of the first machine learning model in a group of first machine learning models determined as a preloading target to obtain a set of preloading configuration parameters.

[0102] In some embodiments, the apparatus 500 further includes a content presentation module configured to present the reply content in the conversation window after determining the reply content to the query content.

[0103] Figure 6 1 is a block diagram of an electronic device 600 in which one or more embodiments of the present disclosure may be implemented. The electronic device 600 may be used to implement, for example, Figure 1 The electronic device 120 shown. It should be understood that Figure 6 The electronic device 600 shown is merely exemplary and should not constitute any limitation on the functionality and scope of the embodiments described herein.

[0104] Reference Figure 6 , the electronic device 600 is in the form of a general electronic device. The components of the electronic device 600 may include, but are not limited to, one or more processors or processing units 610, a memory 620, a storage device 630, one or more communication units 640, one or more input devices 650, and one or more output devices 660. The processing unit 610 may be an actual or virtual processor and is capable of performing various processes according to a program stored in the memory 620. In a multi-processor system, multiple processing units execute computer executable instructions in parallel to improve the parallel processing capability of the electronic device 600.

[0105] The electronic device 600 typically includes a plurality of computer storage media. Such media can be any available media accessible to the electronic device 600, including but not limited to volatile and non-volatile media, removable and non-removable media. The memory 620 can be a volatile memory (e.g., registers, caches, random access memory (RAM)), a non-volatile memory (e.g., a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), flash memory) or some combination thereof. The storage device 630 can be a removable or non-removable medium, and can include a machine-readable medium, such as a flash drive, a disk, or any other medium, which can be used to store information and / or data and can be accessed within the electronic device 600.

[0106] The electronic device 600 may further include additional removable / non-removable, volatile / non-volatile storage media. Figure 6 As shown in , a disk drive for reading or writing from a removable, non-volatile disk (e.g., a "floppy disk") and an optical drive for reading or writing from a removable, non-volatile optical disk may be provided. In these cases, each drive may be connected to the bus (not shown) by one or more data media interfaces. The memory 620 may include a computer program product 625 having one or more program modules that are configured to perform various methods or actions of various embodiments of the present disclosure.

[0107] The communication unit 640 implements communication with other electronic devices through a communication medium. The input device 650 may be one or more input devices, such as a mouse, a keyboard, a tracking ball, etc. The output device 660 may be one or more output devices, such as a display, a speaker, a printer, etc. The electronic device 600 may also communicate with one or more external devices (not shown) through the communication unit 640 as needed, such as a storage device, a display device, etc., communicate with one or more devices that allow a user to interact with the electronic device 600, or communicate with any device (e.g., a network card, a modem, etc.) that allows the electronic device 600 to communicate with one or more other electronic devices 120. Such communication may be performed via an input / output (I / O) interface (not shown).

[0108] According to an exemplary implementation of the present disclosure, a computer-readable storage medium is provided, on which computer-executable instructions are stored, wherein the computer-executable instructions are executed by a processor to implement the method described above. According to an exemplary implementation of the present disclosure, a computer program product is also provided, which is tangibly stored on a non-transitory computer-readable medium and includes computer-executable instructions, and the computer-executable instructions are executed by a processor to implement the method described above.

[0109] Various aspects of the present disclosure are described herein with reference to the flowcharts and / or block diagrams of the methods, devices, equipment, and computer program products implemented according to the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer-readable program instructions.

[0110] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processing unit of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0111] Computer-readable program instructions can be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, so that the instructions executed on the computer, other programmable data processing apparatus, or other device implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0112] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple implementations of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of a module, program segment or instruction includes one or more executable instructions for realizing the logical function of the specification. In some implementations as replacements, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be realized by a special hardware-based system that performs the function or action of the specification, or can be realized by a combination of special hardware and computer instructions.

[0113] The above descriptions of various implementations of the present disclosure are exemplary, non-exhaustive, and not limited to the disclosed implementations. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described implementations. The determination of the terms used herein is intended to best explain the principles of the implementations, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the various implementations disclosed herein.

Claims

1. A method for interaction, comprising: In response to detecting a predetermined operation of the user on the session window, determining a target machine learning model associated with the predetermined operation from a set of first machine learning models; Switching the configuration parameters of the session window to configuration parameters related to the target machine learning model; Based on configuration parameters related to the target machine learning model, processing the query content input by the user in the session window into target information; as well as At least based on the target information, using the target machine learning model, determine the response content for the query content.

2. The method according to claim 1, wherein determining the reply content for the query content by using the target machine learning model based at least on the target information comprises: Sending an operation value corresponding to the predetermined operation and the target information to a second machine learning model, wherein the operation value is configured to instruct the second machine learning model to determine the target machine learning model as a processing object of the target information; Using the second machine learning model, sending the target information to the target machine learning model; as well as Utilize the second machine learning model to obtain the response content for the query content generated by the target machine learning model based on the target information.

3. The method according to claim 1, wherein switching the configuration parameters of the session window to configuration parameters related to the target machine learning model comprises: Utilizing the target control corresponding to the window frame of the session window, the configuration parameters of the session window are switched to configuration parameters related to the target machine learning model.

4. The method according to claim 3, wherein using a target control corresponding to the window frame of the session window to switch the configuration parameters of the session window to configuration parameters related to the target machine learning model comprises: Using the target control, Obtaining current configuration parameters of the session window; Adjusting current configuration parameters of the session window based on the information transmission mode corresponding to the target machine learning model; as well as The adjusted configuration parameters are sent to the session window to switch the configuration parameters of the session window to configuration parameters related to the target machine learning model.

5. The method according to claim 4, wherein based on configuration parameters related to the target machine learning model, processing the query content input by the user in the conversation window into target information comprises: Based on configuration parameters related to the target machine learning model, the query content is processed into a form that matches the information transmission method to obtain the target information. 6 . The method according to claim 3 , wherein the target control is further configured to switch between the session window and other windows by presenting a switching sub-control near the session window.

7. The method according to claim 1, wherein, in response to a predetermined operation of a user on a session window, determining a target machine learning model associated with the predetermined operation in a set of first machine learning models comprises at least one of the following: In response to detecting a user selection of at least one tab presented in the session window, determining a first machine learning model associated with the selection object from the set of first machine learning models as the target machine learning model; or In response to detecting that the user inputs content in the input box of the conversation window, a third machine learning model is used to determine a first machine learning model related to the input content from the group of first machine learning models as the target machine learning model.

8. The method according to claim 1, further comprising: Locally storing configuration parameters of at least one first machine learning model in the set of first machine learning models to obtain a set of preloaded configuration parameters; and The step of switching the configuration parameters of the session window to configuration parameters related to the target machine learning model includes: If there are configuration parameters related to the target machine learning model in the set of preloaded configuration parameters, obtain the configuration parameters related to the target machine learning model from the set of preloaded configuration parameters; and Switch the configuration parameters of the session window to configuration parameters that match the target machine learning model.

9. The method according to claim 8, wherein locally storing configuration parameters of at least one first machine learning model in the set of first machine learning models to obtain a set of preloaded configuration parameters comprises: Based on the editable preloading configuration information, determining at least one first machine learning model in the set of first machine learning models as a preloading target; as well as The configuration parameters of the first machine learning model determined as the preloading target in the set of first machine learning models are stored locally to obtain the set of preloading configuration parameters.

10. The method according to claim 1, further comprising: After the reply content to the query content is determined, the reply content is presented in the conversation window.

11. A device for interaction, comprising: a detection module configured to, in response to detecting a predetermined operation of a user on a session window, determine a target machine learning model associated with the predetermined operation from a group of first machine learning models; A configuration parameter switching module, configured to switch the configuration parameters of the session window to configuration parameters related to the target machine learning model; an information processing module, configured to process the query content input by the user in the conversation window into target information based on configuration parameters related to the target machine learning model; as well as The reply content determination module is configured to determine the reply content for the query content based at least on the target information and using the target machine learning model.

12. An electronic device comprising: at least one processing unit; as well as At least one memory, the at least one memory is coupled to the at least one processing unit and stores instructions for execution by the at least one processing unit, the instructions, when executed by the at least one processing unit, causing the electronic device to perform the method according to any one of claims 1 to 10.

13. A computer-readable storage medium having a computer program stored thereon, wherein the computer program can be executed by a processor to implement the method according to any one of claims 1 to 10.

14. A computer program product comprising computer executable instructions, wherein the computer executable instructions, when executed by a processor, implement the method according to any one of claims 1 to 10.

Images