Intelligent interaction method and device, electronic equipment and storage medium

Abstract

The invention discloses an intelligent interaction method and device, electronic equipment and a storage medium, relates to the technical field of calculation, in particular to the artificial intelligence fields of deep learning, large models, agents and the like, and can be applied to the fields of intelligent document generation and the like. According to the specific implementation scheme, the method comprises the steps of obtaining input information in response to a detected information input operation of a user in a dialogue interface, and sending the input information to a server; receiving related memory information and reply information sent by the server; wherein the related memory information is determined by the server side from memory information associated with the user according to the input information, and the reply information is generated by the server side according to the related memory information; and displaying the related memory information and the reply information in the dialogue interface.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computing, especially to the field of artificial intelligence such as deep learning, large models, and agents, and specifically relates to an intelligent interaction method and device, an electronic device, and a storage medium. BACKGROUND

[0002] With the continuous development of artificial intelligence technology, the functions of intelligent assistant applications are constantly upgraded and optimized, significantly improving the efficiency and experience of people in work, study, and daily life. Such applications not only can make personalized schedules, tasks, or content creation plans for users, but also can automatically generate diversified digital content such as documents, posters, images, etc. according to needs, realizing one-stop intelligent services from understanding intentions to outputting results, and providing users with efficient, convenient, and personalized interaction experiences. SUMMARY

[0003] The present application provides an intelligent interaction method, device, electronic device, and storage medium. The specific solutions are as follows: According to an aspect of the present application, an intelligent interaction method is provided, comprising: In response to detecting a user's information input operation in a dialogue interface, obtaining input information and sending the input information to a server; Receiving relevant memory information and reply information sent by the server; wherein the relevant memory information is determined by the server from the user's associated memory information according to the input information, and the reply information is generated by the server according to the relevant memory information; Displaying the relevant memory information and the reply information in the dialogue interface.

[0004] According to another aspect of the present application, an intelligent interaction method is provided, comprising: Receiving input information of a user sent by a client; According to the input information, determining relevant memory information from the user's associated memory information; According to the relevant memory information, generating reply information corresponding to the input information; Sending the relevant memory information and the reply information to the client.

[0005] According to another aspect of the present application, an intelligent interaction device is provided, comprising: An obtaining module for obtaining input information in response to detecting a user's information input operation in a dialogue interface; A sending module for sending the input information to a server; The receiving module is configured to receive the relevant memory information and the reply information sent by the server, wherein the relevant memory information is determined by the server from the memory information associated with the user according to the input information, and the reply information is generated by the server according to the relevant memory information. The display module is configured to display the relevant memory information and the reply information in the dialogue interface.

[0006] According to another aspect of the present application, an intelligent interaction device is provided, comprising: The receiving module is configured to receive the relevant memory information and the reply information sent by the server, wherein the relevant memory information is determined by the server from the memory information associated with the user according to the input information, and the reply information is generated by the server according to the relevant memory information. The first determining module is configured to determine the relevant memory information from the memory information associated with the user according to the input information. The generating module is configured to generate the reply information corresponding to the input information according to the relevant memory information. The sending module is configured to send the relevant memory information and the reply information to the client.

[0007] According to another aspect of the present application, an electronic device is provided, comprising: at least one processor; and a memory connected with the at least one processor in communication; wherein The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method described in the above embodiments.

[0008] According to another aspect of the present application, a non-transitory computer readable storage medium storing computer instructions is provided, wherein the computer instructions are used to make the computer perform the method according to the above embodiments.

[0009] According to another aspect of the present application, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the steps of the method described in the above embodiments.

[0010] It should be understood that the contents described in this part are not intended to identify the key or important features of the embodiments of the present application, nor are they used to limit the scope of the present application. Other features of the present application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0011] The accompanying drawings serve to better understand the present application, and do not constitute limitations thereof. Among them: Figure 1 A flowchart of an intelligent interaction method provided by an embodiment of the present application is shown; Figure 2 A schematic diagram of a dialogue interface provided by an embodiment of the present application is shown; Figure 3 A flowchart illustrating an intelligent interaction method provided in another embodiment of this application; Figure 4 A flowchart illustrating an intelligent interaction method provided in another embodiment of this application; Figure 5 A schematic diagram of a memory modification interface provided in an embodiment of this application: Figure 6 A flowchart illustrating an intelligent interaction method provided in another embodiment of this application; Figure 7 A flowchart illustrating an intelligent interaction method provided in another embodiment of this application; Figure 8 A flowchart illustrating an intelligent interaction method provided in another embodiment of this application; Figure 9 This application provides a schematic diagram of an architecture for storing information. Figure 10 This is a schematic diagram illustrating the process of generating response information according to an embodiment of this application; Figure 11 This is a schematic diagram of the structure of an intelligent interactive device provided in an embodiment of this application; Figure 12 This is a schematic diagram of the structure of an intelligent interactive device provided in another embodiment of this application; Figure 13 This is a block diagram of an electronic device used to implement the intelligent interaction method of the embodiments of this application. Detailed Implementation

[0012] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of this application, including various details to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0013] It should be noted that the acquisition, storage, use, and processing of data in this application comply with the relevant provisions of national laws and regulations and do not violate public order and good morals.

[0014] The following description, with reference to the accompanying drawings, describes an intelligent interaction method, apparatus, electronic device, and storage medium according to embodiments of this application.

[0015] Figure 1 This is a flowchart illustrating an embodiment of the intelligent interaction method provided in this application.

[0016] The intelligent interaction method of the embodiment of the present application can be executed by the intelligent interaction device of the embodiment of the present application, and the device can be configured in an electronic device.

[0017] The electronic device can be any device with computing capability, for example, a personal computer, a mobile terminal, etc. The mobile terminal can be, for example, a vehicle-mounted device, a mobile phone, a tablet computer, a personal digital assistant, a wearable device, etc. The hardware device has various operating systems, touch screens and / or display screens.

[0018] As shown in the figure, the intelligent interaction method comprises the following steps. Figure 1 Step 101, in response to detecting a user information input operation in the dialogue interface, obtaining input information and sending the input information to the server.

[0019] In the present application, the user can open the intelligent assistant of the application program, the application program can display the dialogue interface between the user and the intelligent agent, the user can input information in the input box of the dialogue interface and trigger the sending control. At this time, the client can detect the user information input operation in the dialogue interface, so as to obtain the input information of the user and display the input information in the dialogue interface, and also send the input information to the server.

[0020] For example, the input information of the user in the dialogue interface is "help me make a 3-month fitness plan".

[0021] Step 102, receiving the relevant memory information and reply information sent by the server.

[0022] The relevant memory information can be determined by the server from the user-associated memory information according to the input information.

[0023] For example, the user-associated memory information can refer to information related to the user.

[0024] For example, the user-associated memory information can include but is not limited to the context of the current dialogue of the user, the dialogue data of the user in the recent period of time, the behavior data of the user in the client in the recent period of time, the portrait information of the user, etc.

[0025] For example, the user can also view, add, modify, delete, etc. the user-associated memory information through the memory information management interface provided by the application program, so as to realize the personalized management of the memory information.

[0026] It should be noted that the number of relevant memory information can be one or more, which is not limited in the present application.

[0027] The reply information can be generated by the server according to the relevant memory information. ​

[0028] Exemplarily, the reply information can be generated by the server based on the relevant memory information and the input information using a large model.

[0029] Exemplarily, the reply information can include, but is not limited to, a text reply, a PPT, a poster, a document, an image, a video, and an audio.

[0030] In this application, the client can obtain the relevant memory information and the reply information sent by the server in a streaming manner.

[0031] Step 103, display the relevant memory information and the reply information in the dialogue interface.

[0032] In this application, the client can display the relevant memory information and the reply information in the dialogue interface in the order of their reception.

[0033] For example, Figure 2 In the dialogue interface shown, the input information of the user is "help me develop a 3-month fitness plan", the three pieces of memory information below "quote the following memory information" are the relevant memory information, and the content of "input result" is the reply information.

[0034] Exemplarily, the client can receive the relevant memory information in a streaming manner and display the relevant memory information character by character on the dialogue interface. Similarly, the reply information can be displayed character by character on the dialogue interface.

[0035] Exemplarily, if there are multiple pieces of relevant memory information, they can be displayed in the dialogue interface in the order of their importance or confidence from high to low, so that the user can focus on the more relevant and more reliable memory content first, significantly improving information acquisition efficiency and decision accuracy, and enhancing interaction interpretability.

[0036] In the embodiments of this application, by obtaining the input information of the user in the dialogue interface, sending the input information to the server, obtaining the relevant memory information determined by the server from the memory information associated with the user based on the input information, and the reply information generated by the server based on the relevant memory information, and displaying the relevant memory information and the reply information in the dialogue interface. When replying to the input information of the user, not only the generated reply information is displayed, but also the relevant memory information on which the reply information is based is presented in the dialogue interface, making the memory information on which the reply information is based visualized, which can significantly enhance the transparency and interpretability of the memory scheduling process, enable the user to understand the generation logic of the reply, improve the matching degree of the output content and the user's demand, avoid the user from obtaining the required content through multiple interactive operations, simplify the user operation complexity, and improve the user experience.

[0037] Figure 3A flowchart of an intelligent interaction method provided for another embodiment of the present application is shown.

[0038] As shown in Figure 3 , the intelligent interaction method comprises the following steps. Step 301, in response to detecting a user information input operation in a dialogue interface, obtaining input information and sending the input information to a server.

[0039] In the present application, step 301 can adopt any one of the implementation manners of the embodiments of the present application, and therefore will not be described here again.

[0040] Step 302, receiving preference inquiry information sent by the server and displaying the preference inquiry information in the dialogue interface.

[0041] The preference inquiry information can be a question for inquiring the user's preference related to the input information.

[0042] In the present application, the preference inquiry information can be preference inquiry information generated by the server according to task information when the server finds matching task information in the user-associated memory information according to the input information.

[0043] For example, the input information is "help me generate a PPT about campus recruitment", and the corresponding preference inquiry information can be "do you want to use the minimalist template of your last PPT?".

[0044] Step 303, obtaining the user's answer to the preference inquiry information input in the dialogue interface and sending the answer to the preference inquiry information to the server.

[0045] In the present application, the user can input the answer in the dialogue interface for the preference inquiry information, and the client can send the server after obtaining the user's answer to the preference inquiry information, so that the server can determine the relevant memory information according to the answer to the preference inquiry information.

[0046] Step 304, receiving relevant memory information determined based on the answer to the preference information sent by the server and displaying the relevant memory information in the dialogue interface.

[0047] The relevant memory information sent by the server can be determined by the server according to the user's answer to the preference inquiry information.

[0048] Step 305, receiving reply information sent by the server and displaying the reply information in the dialogue interface.

[0049] In the present application, step 305 can adopt any one of the implementation manners of the embodiments of the present application, and therefore will not be described here again.

[0050] In the embodiments of the present application, the preference inquiry information sent by the service end is received, and the preference inquiry information is displayed in the dialogue interface. The answer of the user input preference inquiry information is sent to the service end, so that the service end determines the related memory information according to the answer of the preference inquiry information. Therefore, by actively inquiring in the dialogue interface to guide the user to express the preference related to the input information, and using the user's feedback to filter the memory information, the related memory information scheduled and displayed can more accurately meet the personalized needs of the user, thereby significantly improving the accuracy of subsequent reply information, and enhancing the adaptability of intelligent interaction and user satisfaction.

[0051] Figure 4 The flowchart of the intelligent interaction method provided by another embodiment of the present application is shown.

[0052] As shown in Figure 4 , the intelligent interaction method comprises: Step 401, in response to detecting the information input operation of the user in the dialogue interface, obtaining the input information, and sending the input information to the service end.

[0053] Step 402, receiving the related memory information and reply information sent by the service end.

[0054] Step 403, displaying the related memory information and reply information in the dialogue interface.

[0055] In the present application, steps 401-403 can adopt any one of the implementation manners in the embodiments of the present application, so here is not repeated.

[0056] Step 404, in response to detecting the triggering operation of the memory modification control in the dialogue interface, displaying the memory modification interface of the related memory information.

[0057] In actual application, the memory information used by the client to generate the reply information displayed in the dialogue interface may not meet the current user's needs, for example, the memory information used includes the user's creation preference, but the creation preference is inconsistent with the current creation preference of the user. Based on this, in order to improve the accuracy of the reply information, in the present application, the dialogue interface can include a memory modification control. If the triggering operation of the memory modification control of the user is detected, the memory modification interface of the related memory information can be displayed, and the user can modify the memory information relied on by the generation of the reply information through the memory modification interface.

[0058] Among them, the memory modification interface of the related memory information can refer to the interface for modifying the related memory information.

[0059] For example, the memory modification interface can include but is not limited to the related memory information.

[0060] For example, modifications to relevant memory information may include one or more of the following: removing one or more pieces of memory information from the relevant memory information; editing one or more pieces of memory information from the relevant memory information; adding new memory information to the relevant memory information.

[0061] For example, when the user triggers Figure 2 When the "Modify Memory" control 201 is in the middle, it can be displayed Figure 5 The modified memory interface 500 is shown. Figure 5 The memory modification interface 500 shown displays "Referenced Memories" (i.e., related memory information) and "Select More Memories" (i.e., candidate memory information), etc.

[0062] Step 405: In response to detecting a modification operation on the target memory information in the memory modification interface, obtain the modified memory information.

[0063] The target memory information can be one or more pieces of related memory information, or it can be other memory information displayed in the memory modification interface besides the related memory information.

[0064] In this application, each piece of memory information displayed in the memory modification interface can have a corresponding operation control, such as a selection control or an editing control. The selection control can be used to select or deselect the corresponding memory information to generate a response, and the editing control can be the control triggered to edit the corresponding memory information.

[0065] for example, Figure 5 In the memory modification interface 500 shown, each piece of memory information has a corresponding selection control 501 and an editing control 502.

[0066] For example, the relevant memory information displayed in the memory modification interface can be selected by default.

[0067] In some embodiments, the target memory information may belong to the related memory information. If the target memory information is in a selected state and the client detects that the user has deselected the selection control corresponding to the target memory information, then the target memory information can be removed from the related memory information. That is, the target memory information is not used to generate response information, thereby obtaining the modified memory information.

[0068] for example, Figure 5 There are 3 memory records in "Referenced Memory". If the selection control 501 in the first memory record in "Referenced Memory" is deselected, and no other memory records in the memory modification interface are modified, then the modified memory records will include the second and third memory records in "Referenced Memory".

[0069] Therefore, the user can flexibly remove the target memory information from the relevant memory information that has been used to generate the reply information by selecting the control, thereby autonomously defining the memory information on which the reply is regenerated, and can give the user the initiative control right of the memory information scheduling process. Not only can the intervention and personalization levels of the interaction be improved, but the reply basis can also be dynamically adjusted according to actual needs, and the accuracy and adaptability of the response can be effectively enhanced.

[0070] Since some relevant memory information can not be completely consistent with the current user demand, in order to improve the fit degree of the memory information used to generate the reply information and the user demand, in some embodiments, the target memory information can belong to the relevant memory information. If the target memory information is in the selected state, and the client detects the triggering operation of the editing control corresponding to the target memory information by the user, an editing interface of the target memory information can be displayed. If the editing operation of the target memory information is detected in the editing interface, the edited memory information can be obtained, and then the modified memory information can be obtained according to the edited memory information.

[0071] Among them, the modified memory information can include the edited memory information, and can also include other memory information, etc.

[0072] For example, one or more memory information in the relevant memory information can be edited, which is not limited.

[0073] For example, when the user triggers the editing control 502 corresponding to the memory information "user likes XXXX" in the memory modification interface 500 shown in the figure, an editing interface of the memory information can be displayed. Figure 5

[0074] For example, if there are 4 pieces of relevant memory information, 2 pieces of memory information are edited, and no operation is performed on other memory information in the memory modification interface, the edited 2 pieces of memory information and the 2 pieces of memory information in the relevant memory information that are not edited can be used as the modified memory information, and can be used as the memory information required for regenerating the reply information.

[0075] Therefore, the user can directly modify the relevant memory information that has been used to generate the reply information through the editing control, so that the memory information is more consistent with the current intention of the user. Then, the relevant memory information is updated based on the edited memory information, and is used as the basis for regenerating the reply information. Not only can the modification demand of the user for the memory information required for generating the reply information be met, but also the memory information used for generating the reply information can meet the user demand, thereby effectively guaranteeing the accuracy, timeliness and personalization level of the subsequent reply information, and enhancing the reliability and user experience of human-computer interaction.

[0076] ​The richer the memory information provided, the more accurate the reply information that can be generated. Based on this, in some embodiments, the memory modification interface can include relevant memory information and candidate memory information. The candidate memory information can be in an unselected state by default. The target memory information can belong to the candidate memory information. If the candidate memory information is in an unselected state and the client detects a selection operation of the selection control corresponding to the target memory information, the target memory information can be added to the relevant memory information to obtain modified memory information.

[0077] The candidate memory information can be memory information related to the input information.

[0078] The number of candidate memory information can be one or more, which is not limited.

[0079] For example, there are 2 pieces of candidate memory information displayed in the memory modification interface. One of the candidate memory information can be selected, and the relevant memory information is not operated. The selected candidate memory information and the relevant memory information can be used as the modified memory information to regenerate the reply information.

[0080] In this way, the candidate memory information can be provided in the memory modification interface. The user can dynamically include the default unselected candidate memory information into the memory information used to generate the reply information through the selection control. The memory basis relied on by the reply generation can be flexibly expanded or adjusted. Not only can the memory information be supplemented and customized on demand, but also the user's active intervention ability in model decision-making in the interaction process can be significantly enhanced. Therefore, the relevance, completeness and accuracy of the regenerated reply information can be improved, and the explainability of the intelligent dialogue can be further enhanced.

[0081] It should be noted that the user can obtain the modified memory information by one or more of the following operations: deselecting the relevant memory information in the memory modification interface, editing the relevant memory information in the memory modification interface, selecting the candidate memory information, and the like.

[0082] For example, there are 4 pieces of relevant memory information and 3 pieces of candidate memory information in the memory modification interface. One of the relevant memory information can be deselected, one of the relevant memory information can be edited, and two of the candidate memory information can be selected. The remaining two pieces of relevant memory information that are not operated, the edited memory information, and the selected two pieces of candidate memory information can be used as the modified memory information.

[0083] Step 406, in response to detecting the triggering operation of the memory modification interface, the modified memory information is sent to the server.

[0084] In the present application, after the user completes the modification of the memory information in the memory modification interface, the user can trigger the re-generation control in the memory modification interface to re-generate the reply information by using the modified memory information. Therefore, if the client detects the triggering operation of the re-generation control in the memory modification interface, the client can send the modified memory information to the server to request the server to re-generate the reply information according to the modified memory information.

[0085] For example, when the user triggers the

modify and re-generate

[0086] Optionally, if the client detects the triggering operation of the re-generation control in the memory modification interface, the client can close the memory modification interface and display the modified memory information in the dialogue interface.

[0087] Therefore, the display of the modified memory information in the dialogue interface not only shows the user the memory information used for re-generating the reply information, improves the transparency of the user's scheduling of the memory information, but also instantaneously feeds back the user's perception of the modification result of the memory information, further improves the interaction efficiency and the coherence of the dialogue response, and realizes the seamless connection between the modification of the memory information and the dialogue response.

[0088] In step 407, the re-generated reply information sent by the server according to the modified memory information is received.

[0089] In the present application, the client can receive the re-generated reply information sent by the server, which can be re-generated by the server according to the modified memory information and input information.

[0090] In the present application, the user can trigger the memory modification control in the dialogue interface to conveniently enter the memory modification interface, add, delete or edit the relevant memory information used for generating the reply, form the modified memory information meeting the current intention of the user, and send the updated memory to the server through the re-generation control, so that the server re-generates the reply information based on the modified context. The user can realize the full-process controllable intervention of the memory dependent content, thereby improving the matching degree of the memory information used for generating the reply information and the user's demand, and further improving the accuracy, timeliness and personalization level of the reply information.

[0091] Figure 6 The flowchart of the intelligent interaction method provided by another embodiment of the present application.

[0092] The intelligent interaction method of the present application can be executed by the intelligent interaction device of the present application, which can be configured in an electronic device. ​

[0093] The electronic device can be any device with computing capability, such as a server.

[0094] As shown in the figure, the intelligent interaction method includes: Figure 6 Step 601, receiving input information of a user sent by a client.

[0095] In this application, the interpretation and explanation of input information can refer to the above-mentioned embodiments, and therefore will not be repeated here.

[0096] Step 602, determining relevant memory information from the memory information associated with the user according to the input information.

[0097] The memory information associated with the user can refer to information related to the user.

[0098] For example, the memory information associated with the user can be collected from one or more application programs, so that multi-terminal sharing of memory information can be achieved.

[0099] For example, the memory information associated with the user can be divided into short-term memory information, medium-term memory information and long-term memory information according to the TTL (Time To Live, life cycle time) and stability of the memory information.

[0100] The short-term memory information can include current dialogue context information, the medium-term memory information can include user behavior data and dialogue data in the recent period of time, and the long-term memory information can include user portrait and user preference.

[0101] For example, the long-term memory information can be obtained in one or more of the following ways: by extracting the medium-term memory information, by the user actively requesting to remember, etc.

[0102] For example, the user inputs "remember that I like XX" in the dialogue interface to request active memory of the preference.

[0103] In this application, the server can plan tasks according to the input information, determine the tasks to be executed, the execution order of the tasks, the tools to be called by the tasks, etc., and match the input information with the memory information associated with the user to determine the memory information related to the input information, i.e., to determine the relevant memory information.

[0104] For example, the tools that can be called by the tasks can include but are not limited to retrieval tools, PPT generation tools, poster generation tools, etc.

[0105] ​Exemplarily, similar short-term memory information can be searched in the short-term memory information associated with the user, the relevant medium-term memory information can be searched in the medium-term memory information associated with the user according to the input information, the relevant long-term memory information can be searched in the long-term memory information associated with the user according to the input information, and the similar short-term memory information, the relevant medium-term memory information, the relevant long-term memory information, etc. can be obtained as the relevant memory information.

[0106] Exemplarily, if multiple memory information related to the input information is searched from the long-term memory information associated with the user, the relevant long-term memory information can be obtained by screening according to the confidence, or the importance, or the calling frequency of the long-term memory information. For example, the preset number of long-term memory information with the highest confidence, or the highest importance, or the most calling frequency can be obtained as the relevant long-term memory information.

[0107] Optionally, the application program can set a memory information use switch, and the user can open or close the switch in the application program interface. If the switch is opened, the memory information associated with the user can be called when generating a reply in the conversation process, and the memory information can be stored. In the case that the user opens the switch, the server can match the input information with the memory information associated with the user to determine the memory information related to the input information.

[0108] In actual application, some input information can not need to call the memory information of the user, such as the input information of “what day is today” or “translate the English and Chinese in the picture into Chinese”, etc. Based on this, in some embodiments, whether the memory information needs to be dispatched can be determined according to the input information first, and then the relevant memory information can be determined from the memory information associated with the user according to the input information if it is determined that the memory information needs to be dispatched.

[0109] In step 603, the reply information corresponding to the input information is generated according to the relevant memory information.

[0110] In this application, the server can use a large model to generate the reply information of the input information according to the relevant memory information, the input information, etc.

[0111] Exemplarily, if the task to be executed corresponding to the input information does not need to call the tool, the reply information can be generated directly according to the relevant memory information, the input information, etc. If the task to be executed needs to call the tool, the memory information corresponding to the tool in the relevant memory information is determined, the relevant memory information can be executed in turn according to the task execution order, and the memory information required in the execution of the task can be determined from the relevant memory information, the corresponding tool can be called according to these memory information, the tool execution result can be obtained, and the reply information can be obtained according to the tool execution result.

[0112] Step 604, sending the relevant memory information and the reply information to the client.

[0113] In the present application, the server can first send the relevant memory information to the client, and then send the reply information to the client, so as to display the relevant memory information and the reply information on the dialogue interface of the client.

[0114] For example, the server can send the relevant memory information and the reply information to the client in a streaming manner.

[0115] In the embodiments of the present application, the input information of the user sent by the client is received, and the relevant memory information is determined from the memory information associated with the user according to the input information. The reply information corresponding to the input information is generated according to the relevant memory information, and the relevant memory information and the reply information are sent to the client. Therefore, when replying to the input information of the user, not only the reply information is sent to the client for display on the client, but also the relevant memory information used to generate the reply information is sent to the client for display on the client, which can visualize the memory information relied on to generate the reply information, significantly enhance the transparency and explainability of the memory scheduling process, enable the user to understand the generation logic of the reply, improve the matching degree of the output content and the user demand, avoid the user from obtaining the required content through multiple interactive operations, simplify the user operation complexity, and improve the user experience.

[0116] Figure 7 The flowchart of the intelligent interaction method provided by another embodiment of the present application is shown.

[0117] As shown in Figure 7 The intelligent interaction method comprises the following steps: Step 701, receiving the input information of the user sent by the client.

[0118] In the present application, step 701 can adopt any one of the implementation manners of the embodiments of the present application, and therefore will not be described here.

[0119] Step 702, finding the task information matching the input information in the associated memory information according to the input information.

[0120] Since the intention of the current input information of the user can be similar to the intention of the historical input information of the user, in order to improve the user experience, the matching task information can be found in the short-term memory information in the associated memory information according to the input information.

[0121] For example, the input information is "help me make a campus recruitment PPT", and the task information of using a minimalist template to make a PPT before is found in the short-term memory information associated with the user.

[0122] Step 703, in response to finding the matched task information, generating preference inquiry information according to the matched task information, and sending the preference inquiry information to the client.

[0123] In this application, if the matched task information is found, the preference inquiry information can be generated according to the matched task information combined with the inquiry dialogue template, or the preference inquiry information can be generated according to the matched task information, input information, etc. by using a large model.

[0124] For example, the input information is "help me make a campus recruitment PPT", and the preference inquiry information can be "do you want to follow the minimalist template of your last PPT?".

[0125] Step 704, receiving the user's answer to the preference inquiry information sent by the client, and determining the relevant memory information according to the answer to the preference inquiry information.

[0126] In some embodiments, the server can directly determine the relevant memory information according to the user's answer to the preference inquiry information.

[0127] In some embodiments, the server can match the input information with the memory information associated with the user, find the memory information matching the input information, and then update the memory information matching the input information according to the user's answer to the preference inquiry information to obtain the relevant memory information. Thus, the relevant memory information can be consistent with the real-time needs of the user.

[0128] For example, the memory information matching the input information includes memory information similar to the answer to the preference inquiry information, and then the relevant memory information can be obtained by replacing the similar memory information with the answer to the preference inquiry information.

[0129] For example, the answer to the preference inquiry information is "use the last PPT minimalist template", and the memory information matching the input information includes "like complex PPT templates", and then "use the last PPT minimalist template" can be used to replace the memory information "like complex PPT templates".

[0130] Step 705, generating reply information corresponding to the input information according to the relevant memory information.

[0131] Step 706, sending the relevant memory information and the reply information to the client.

[0132] In this application, steps 705-706 can use any of the implementation manners of the embodiments of the application, so this will not be repeated here.

[0133] Optionally, the server can provide a timing task reminder, such as a holiday reminder, and can also generate inquiry information related to the schedule according to the user's preference information, and actively ask the user, so as to improve the intelligent level of interaction and improve the user experience. For example, the generated inquiry information is "I remember that you like the singer XXX, and recently he is holding a concert in your location. Do you need me to provide some relevant consultation?".

[0134] In the embodiments of the present application, the matching task context in the user associated memory information is actively identified based on the input information, and the targeted preference inquiry information is generated accordingly, the user is guided to explicitly express the current intention or preference, the relevant memory information is dynamically screened and adjusted by combining the feedback of the user to the preference inquiry, the memory content used is ensured to be highly consistent with the current demand of the user, the context awareness and personalized adaptation of memory information calling are realized, the accuracy, relevance and user satisfaction of the reply information can be improved, and the initiative and interaction intelligence level of the intelligent dialogue are enhanced.

[0135] Figure 8 The flowchart of the intelligent interaction method provided by another embodiment of the present application is shown.

[0136] As shown in Figure 8 The intelligent interaction method comprises: Step 801, receiving the input information of the user sent by the client.

[0137] In the present application, step 801 can adopt any one of the implementation manners of the embodiments of the present application, and therefore will not be repeated here.

[0138] Step 802, performing intention recognition on the input information to obtain the user intention.

[0139] In the present application, the intention recognition model can be used to perform intention recognition on the input information to obtain the user intention, or a large model can be used to perform intention recognition on the input information to obtain the user intention, or a natural language processing method can be used to perform intention recognition on the input information to obtain the user intention, or other ways of intention recognition can be used, and no limitation is made in this regard.

[0140] Step 803, in response to the user intention including modifying the generated historical material, querying the historical dialogue data in the associated memory information according to the input information to obtain the target material.

[0141] In actual application, the user can have the demand of modifying the previously generated file, image and the like. Based on this, in the present application, if the user intention includes the intention of modifying the generated historical material, the input information can be used to query the historical dialogue data in the memory information associated with the user to determine the matched historical dialogue data and obtain the generated historical material in the historical dialogue data as the target material.

[0142] The target material can include but is not limited to documents, images, videos, audios, posters and the like. For example, the document can include but is not limited to a word document, a PPT document and the like, or a document in other formats and the like.

[0143] For example, the input information is "help me find the recently generated PPT, I want to change the title", the user intention is to modify the title of the recently generated PPT, and the recently generated PPT can be found in the historical dialogue data as the target material.

[0144] In step 804, the relevant memory information is determined according to the target material.

[0145] The relevant memory information can include but is not limited to the target material.

[0146] In the present application, the target material can be used as the relevant memory information.

[0147] For example, the relevant memory information can be found in the memory information associated with the user according to the user intention and the modification content of the target material, and the target material and the found matching memory information are used as the relevant memory information.

[0148] In step 805, the reply information corresponding to the input information is generated according to the relevant memory information.

[0149] In the present application, the modification strategy of the target material in the relevant memory information can be determined according to the user intention, the target material is modified according to the modification strategy, the modified material is obtained, and the reply information is determined according to the modified material. The reply information includes the modified material.

[0150] For example, if the user intention does not explicitly indicate how to modify, the corresponding inquiry information can be generated to inquire the user about the specific modification, and the modification strategy is determined according to the inquiry result.

[0151] For example, the modification content of the target material is to modify the title of the recently generated PPT, and the corresponding inquiry information can be "what title do you want to change to".

[0152] For example, if there is memory information matching the modified content of the target material in the relevant memory information, the target material can be modified according to the user's intention, the matching memory information, and the like.

[0153] Therefore, when it is identified that the user has the intention to modify the generated material, the relevant historical dialogue data in the memory information can be automatically associated and retrieved, the corresponding historical material can be accurately located and supported for secondary editing, that is, secondary creation, realizing traceable, reusable, and iterative closed-loop management of the generated content. Not only can the efficiency and coherence of content creation be improved, but also the personalized adjustment needs of the user can be flexibly responded to.

[0154] At step 806, the relevant memory information and the reply information are sent to the client.

[0155] In this application, step 806 can adopt any of the implementation manners of the embodiments of the present application, and therefore will not be described here.

[0156] In the embodiments of the present application, the user's intention is obtained by performing intention recognition on the input information. If the user's intention includes the intention to modify the generated historical material, the historical dialogue data in the associated memory information is queried according to the input information to obtain the target material, and the relevant memory information is determined according to the target material. Therefore, the modification needs of the user for the historical generated material can be accurately identified based on the user's intention, and the corresponding target material can be retrieved from the associated memory information, and then the relevant memory information highly matching the current intention is determined based on the target material. Not only can the intention alignment and context relevance of memory calling be improved, but also the accuracy, consistency, and personalization level of the reply information can be enhanced.

[0157] In an embodiment of the present application, the server can also receive the modified memory information sent by the client, regenerate the reply information using a large model according to the modified memory information and the input information, and send the regenerated reply information to the client.

[0158] The modified memory information can be obtained by modifying the target memory information in the memory modification interface of the client by the user.

[0159] The explanation of the target memory information can be referred to the above-mentioned embodiments, and therefore will not be described here.

[0160] Therefore, the server regenerates the reply information based on the memory information modified by the user, so as to meet the control needs of the user for the memory information, and make the reply information consistent with the user's needs.

[0161] In some embodiments of the present application, the memory information associated with the user can include target type memory information, which can refer to the long-term memory information described above. In order to improve the accuracy of the long-term memory information, the target type memory information can be updated according to the attribute parameters of the target type memory information.

[0162] The attribute parameters can be used to represent the state of the target type memory information.

[0163] For example, the attribute parameters include one or more of the following: confidence, importance, recent use time, recall frequency, memory source, etc.

[0164] The confidence can represent the degree of trust of the extraction model in the correctness of the memory information, which is usually derived from the confidence score or matching degree output by the extraction model. For example, the memory information "user likes simple style" is generated by consistent expression of multiple rounds of dialogue, and the confidence is confidence = 0.92.

[0165] The importance can refer to the importance of the memory information to the personalized decision or generation task, which can be calculated by task frequency, emotional intensity, or association degree with the user's core profile. For example, the memory information "user is product manager" has an importance importance = 0.95 because the product manager is the core identity of the user.

[0166] The recent use time refers to the timestamp of the last time the memory information was called or verified, which is used for time decay. For example, the last reference time of a memory information is 2025-10-30.

[0167] The recall frequency can refer to the cumulative number of times the memory is called or triggered by the model in history, which represents the frequency of use. For example, a memory information has a recall frequency of 30 days.

[0168] The memory source is used to record the generation source of the memory information, such as specific session ID, file name, tool name, task source, etc.

[0169] For example, the description form of the target type memory information includes at least one of the following: structured field form, natural language summary form, etc.

[0170] For example, long-term memory information can be described in structured fields such as occupation, interest, tone preference, task habit, etc., and natural language summaries with strong interpretability.

[0171] Therefore, the structured field description is more accurate, and the natural language summary description has strong interpretability, which is convenient for subsequent calling and management of target type memory information.

[0172] Exemplarily, the weight of the memory information of the target type can be determined according to the attribute parameter of the memory information of the target type, and whether to delete the memory information of the target type can be determined according to a size relationship between the weight and a preset threshold.

[0173] As an example, the attribute parameter can be weighted to obtain the weight according to the weight of the attribute parameter.

[0174] As an example, the weight can be determined according to part of the attribute parameters of the memory information of the target type. For example, the weight can be determined according to the confidence, the importance, the number of invocations, the latest invocation time, and the like.

[0175] As an example, if the weight is less than the preset threshold, the memory information of the target type can be deleted, and if the weight is greater than or equal to the preset threshold, the memory information of the target type can be retained.

[0176] It can be understood that the weight of the memory information related to the task will be constantly improved and kept active, while the memory information that has not been used for a long time, has low confidence, or conflicts with new memory information will automatically reduce the weight over time and eventually fade out of the use range.

[0177] Therefore, according to the attribute parameter of the memory information of the target type, the weight of the memory information of the target type is determined, and the memory information of the target type is automatically deleted according to the weight, which can ensure the simplicity and accuracy of the memory information of the target type.

[0178] Since these attribute parameters can dynamically change, exemplarily, if the related memory information used to generate the reply information includes the memory information of the target type, the attribute parameters can be updated first to obtain updated attribute parameters, and then the weight of the memory information of the target type is determined according to the updated attribute parameters.

[0179] Therefore, the attribute parameters of the memory information of the target type can be dynamically updated according to the invocation of the memory information of the target type, and the weight of the memory information of the target type can be dynamically adjusted according to the updated attribute parameters, so that the memory information can evolve on the time axis, avoid invalid or expired content from continuously interfering with decision-making, and improve the relevance and stability of the memory information invocation.

[0180] In order to facilitate understanding of the intelligent interaction method of the embodiments of the present application, the following will be described in conjunction with Figure 9 and Figure 10 Figure 9 a schematic diagram of a memory information architecture provided by the embodiments of the present application. Figure 10 a schematic diagram of a reply information generation process provided by the embodiments of the present application. ​

[0181] like Figure 9 As shown, the architecture of memory information includes the accumulation, management, processing, and application of memory information.

[0182] The accumulation of memory information includes explicit feedback and implicit feedback. Explicit feedback includes user dialogue, content editing, file upload, and self-editing, while implicit feedback includes user profiles, dialogue rounds, adoption status, and dialogue context.

[0183] The management of memory information includes storage, updating, retrieval, and forgetting. Storage mainly involves encoding and persisting memory information, updating refers to correcting and integrating memory information, retrieval mainly involves similarity matching and context awareness, and forgetting mainly involves filtering and lifecycle management of memory information.

[0184] The processing of memory information includes dividing it into short-term memory, medium-term memory, and long-term memory. For example, short-term memory information may include task context, dialogue context, etc.; medium-term memory information may include user behavior data and user dialogue history over a recent period, including historical file results; and long-term memory information may include user profiles, content creation preferences, and task execution preferences.

[0185] The application of memory information can include tasks such as content search, creation, and management, as well as daily communication scenarios such as simple question and answer and casual conversation.

[0186] Among them, the task execution scenario mainly applies personalized content preferences, such as search preferences, creation preferences (such as writing style preferences, logic preferences, etc.), management preferences, etc., while the daily communication scenario mainly applies personal communication preferences, such as communication style, personality traits, personal experience, life needs, etc.

[0187] like Figure 10 As shown, the process of generating response information includes a memory recall process and a memory injection process.

[0188] Memory evoked process: After obtaining user input information, task planning can be performed on the input information. Then, based on the input information, it is determined whether to evoked memory information. If memory information is evoked, the user's associated memory information is queried to retrieve the relevant memory information. If memory information is not evoked, the non-memory evoked scenario process is entered.

[0189] This includes querying the user's associated memory information. Figure 10 (Not shown), including: similarity matching in short-term memory information, retrieval in medium-term memory information, retrieval in long-term memory information, etc.

[0190] Memory injection process: according to the relevant memory information and the task planning result, the corresponding memory information is transmitted to the corresponding tool, then the tool is called, the tool execution result is obtained, and the reply information is obtained according to the tool execution result.

[0191] To realize the above-mentioned embodiments, the embodiments of the present application further provide an intelligent interaction device. Figure 11 The intelligent interaction device provided by an embodiment of the present application has the structure shown in the structural schematic diagram.

[0192] As Figure 11 shown, the intelligent interaction device 1100 includes: The acquisition module 1110 is configured to acquire input information in response to detecting a user information input operation in a dialogue interface. The sending module 1120 is configured to send the input information to a server. The receiving module 1130 is configured to receive relevant memory information and reply information sent by the server, wherein the relevant memory information is determined by the server from memory information associated with the user according to the input information, and the reply information is generated by the server according to the relevant memory information. The display module 1140 is configured to display the relevant memory information and the reply information in the dialogue interface.

[0193] Optionally, the receiving module 1130 is configured to: receive preference inquiry information sent by the server and display the preference inquiry information in the dialogue interface; acquire a user answer to the preference inquiry information input in the dialogue interface and send the answer to the preference inquiry information to the server; receive relevant memory information determined by the server based on the answer to the preference information.

[0194] Optionally, the display module 1140 is further configured to display a memory modification interface of the relevant memory information in response to detecting a triggering operation of a memory modification control in the dialogue interface. The acquisition module 1110 is further configured to acquire modified memory information in response to detecting a modification operation of target memory information in the memory modification interface. The sending module 1120 is further configured to send the modified memory information to the server in response to detecting a triggering operation of a re-generation control in the memory modification interface. The receiving module 1130 is further configured to receive reply information regenerated by the server according to the modified memory information.

[0195] Optionally, the target memory information belongs to the relevant memory information, and the acquisition module 1110 is further configured to: In response to the target memory information being in the selected state and detecting a de-selection operation on the selection control corresponding to the target memory information, the target memory information is removed from the related memory information to obtain modified memory information.

[0196] Optionally, the target memory information belongs to the related memory information, and the obtaining module 1110 is further configured to: In response to the target memory information being in the selected state and detecting a trigger operation on the editing control corresponding to the target memory information, an editing interface of the target memory information is displayed. In response to detecting an editing operation on the target memory information in the editing interface, edited memory information is obtained. According to the edited memory information, modified memory information is obtained.

[0197] Optionally, the memory modification interface further includes candidate memory information, the target memory information belongs to the candidate memory information, and the obtaining module 1110 is further configured to: In response to the target memory information being in the unselected state and detecting a selection operation on the selection control corresponding to the target memory information, the target memory information is added to the related memory information to obtain modified memory information.

[0198] Optionally, the display module 1140 is further configured to: In response to detecting a trigger operation on the regeneration control, the memory modification interface is closed, and the modified memory information is displayed in the dialogue interface.

[0199] It should be noted that the explanation and description of the foregoing client-side intelligent interaction method embodiment also applies to the intelligent interaction device of this embodiment, and thus will not be described here.

[0200] In the embodiments of the present application, by obtaining the input information of the user in the dialogue interface, the input information is sent to the server, and the related memory information determined by the server from the memory information associated with the user based on the input information and the reply information generated by the server based on the related memory information are obtained, and the related memory information and the reply information are displayed in the dialogue interface. Thus, when replying to the input information of the user, not only the generated reply information is displayed, but also the related memory information on which the reply information is based is presented in the dialogue interface, making the memory information on which the reply information is based visualized, which can significantly enhance the transparency and explainability of the memory scheduling process, enabling the user to understand the generation logic of the reply, improving the matching degree of the output content and the user's demand, avoiding the user obtaining the required content through multiple interaction operations, simplifying the user operation complexity, and improving the user experience.

[0201] To implement the above-mentioned embodiments, the embodiments of the present application further propose an intelligent interaction device. Figure 12A structural schematic diagram of an intelligent interaction device provided for another embodiment of the present application.

[0202] As shown in Figure 12 The intelligent interaction device 1200 includes: A receiving module 1210, configured to receive input information of a user sent by a client; A first determining module 1220, configured to determine relevant memory information from memory information associated with the user according to the input information; A generating module 1230, configured to generate reply information corresponding to the input information according to the relevant memory information; A sending module 1240, configured to send the relevant memory information and the reply information to the client.

[0203] Optionally, the first determining module 1220 is configured to: According to the input information, search for task information matching the input information in the associated memory information; In response to finding the matching task information, generate preference inquiry information according to the matching task information; Send the preference inquiry information to the client; Receive an answer of the user to the preference inquiry information sent by the client, and determine the relevant memory information according to the answer to the preference inquiry information.

[0204] Optionally, the first determining module 1220 is configured to: Perform intent recognition on the input information to obtain a user intent; In response to the user intent including an intent to modify generated historical materials, query historical dialogue data in the associated memory information according to the input information to obtain target materials; Determine the relevant memory information according to the target materials.

[0205] Optionally, the first determining module 1220 is configured to: According to the user intent, modify target materials in the relevant memory information to obtain modified materials; Determine the reply information according to the modified materials.

[0206] Optionally, the associated memory information includes memory information of a target type, and the device can further include: A second determining module, configured to determine a weight of the memory information of the target type according to an attribute parameter of the memory information of the target type; wherein the attribute parameter is used to represent a state of the memory information of the target type; A deleting module, configured to delete the memory information of the target type in response to the weight being less than a preset threshold.

[0207] Optionally, the second determining module is configured to: In response to the memory information of the target type included in the related memory information, the attribute parameter is updated to obtain an updated attribute parameter. According to the updated attribute parameter, the weight is determined.

[0208] Optionally, the description form of the memory information of the target type includes at least one of the following: a structured field form, a natural language summary form.

[0209] It should be noted that the explanation and description of the foregoing server-side intelligent interaction method embodiment also applies to the intelligent interaction device of the embodiment, and thus will not be described here.

[0210] In the embodiment of the application, the input information of the user sent by the client is received, and according to the input information, the related memory information is determined from the memory information associated with the user, according to the related memory information, the reply information corresponding to the input information is generated, and the related memory information and the reply information are sent to the client. Therefore, when replying to the input information of the user, not only the reply information is sent to the client for display on the client, but also the related memory information used to generate the reply information is sent to the client for display on the client, which can make the memory information relied on to generate the reply information visualized, can significantly enhance the transparency and explainability of the memory scheduling process, enable the user to understand the generation logic of the reply, improve the matching degree of the output content and the user demand, avoid the user from obtaining the required content through multiple interaction operations, simplify the user operation complexity, and improve the user experience.

[0211] According to the embodiments of the application, the application also provides an electronic device, a readable storage medium and a computer program product.

[0212] Figure 13 A schematic block diagram of an example electronic device 1300 that can be used to implement embodiments of the application is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit implementations of the application described and / or claimed in this document.

[0213] As Figure 13As shown, the device 1300 includes a computing unit 1301 that can perform various appropriate actions and processes in accordance with a computer program stored in a ROM (Read-Only Memory) 1302 or a computer program loaded into a RAM (Random Access Memory) 1303 from the storage unit 1308. Various programs and data required for the operation of the device 1300 can also be stored in the RAM 1303. The computing unit 1301, the ROM 1302, and the RAM 1303 are connected to each other through a bus 1304. An I / O (Input / Output) interface 1305 is also connected to the bus 1304.

[0214] Various components in the device 1300 are connected to the I / O interface 1305, including an input unit 1306 such as a keyboard, a mouse, etc., an output unit 1307 such as various types of displays, speakers, etc., a storage unit 1308 such as a magnetic disk, an optical disk, etc., and a communication unit 1309 such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 1309 allows the device 1300 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0215] The computing unit 1301 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 1301 include, but are not limited to, a CPU (Central Processing Unit), a GPU (Graphic Processing Unit), various special-purpose AI (Artificial Intelligence) computing chips, various computing units running machine learning model algorithms, a DSP (Digital Signal Processor), and any appropriate processor, controller, microcontroller, etc. The computing unit 1301 performs various methods and processes described above, such as the intelligent interaction method. For example, in some embodiments, the intelligent interaction method can be implemented as a computer software program that is tangibly embodied in a machine-readable medium, such as the storage unit 1308. In some embodiments, part or all of the computer program can be loaded and / or installed on the device 1300 via the ROM 1302 and / or the communication unit 1309. When the computer program is loaded into the RAM 1303 and executed by the computing unit 1301, one or more steps of the intelligent interaction method described above can be performed. Alternatively, in other embodiments, the computing unit 1301 can be configured to perform the intelligent interaction method by any other appropriate means, e.g., by means of firmware.

[0216] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, a Field Programmable Gate Array (FPGA), an Application-Specific Integrated Circuit (ASIC), an Application Specific Standard Product (ASSP), a System on a Chip (SOC), a Complex Programmable Logic Device (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0217] Program code for carrying out methods of the present application can be written in any combination of one or more programming languages. This program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces a means for implementing the functions / acts specified in the flowcharts and / or block diagrams. The program code can be executed entirely on a machine, partially on a machine, partially on a machine and partially on a remote machine or entirely on a remote machine or server.

[0218] In the context of this application, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include but is not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium would include a linearly-programmed electrical connection, a portable computer diskette, a hard disk, a RAM, a ROM, an erasable programmable read-only memory (EPROM) or flash memory, an optical fiber, a compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0219] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0220] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a LAN (local area network), a WAN (wide area network), the Internet, and a blockchain network.

[0221] The computer system can include clients and servers. This relationship can be implemented by a computer program running on the respective computers and having a client-server relationship with one another. The servers can be cloud servers, also known as cloud computing servers or cloud hosts, which are mainframe products in the cloud computing service system, and solve the defects of large management difficulty and weak business scalability in traditional physical hosts and VPS services (Virtual Private Server, Virtual Private Server).

[0222] According to the embodiments of the present application, the present application also provides a computer program product, when the instruction processor in the computer program product executes, executes the intelligent interaction method proposed in the above embodiments of the present application.

[0223] It should be understood that the steps shown above can be reordered, added or deleted. For example, the steps described in the present application can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in the present application can be achieved, which is not limited herein.

[0224] The above detailed description does not constitute a limitation on the scope of protection of the present application. Those skilled in the art should understand that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. An intelligent interaction method, comprising: in response to detecting a user information input operation in a conversation interface, obtaining input information and sending the input information to a server; receiving relevant memory information and reply information sent by the server; wherein the relevant memory information is determined by the server from memory information associated with the user according to the input information, and the reply information is generated by the server according to the relevant memory information; displaying the relevant memory information and the reply information in the conversation interface.

2. The method of claim 1, wherein, The receiving of the relevant memory information sent by the server comprises: receiving preference inquiry information sent by the server and displaying the preference inquiry information in the conversation interface; obtaining the user's answer to the preference inquiry information input in the conversation interface and sending the answer to the preference inquiry information to the server; receiving the relevant memory information determined by the server based on the answer to the preference information.

3. The method of claim 1, further comprising: in response to detecting a trigger operation of a memory modification control in the conversation interface by the user, displaying a memory modification interface of the relevant memory information; in response to detecting a modification operation on target memory information in the memory modification interface, obtaining modified memory information; in response to detecting a trigger operation of a re-generation control in the memory modification interface, sending the modified memory information to the server; receiving reply information regenerated by the server according to the modified memory information.

4. The method of claim 3, wherein, The target memory information belongs to the relevant memory information, and the obtaining of the modified memory information in response to the detection of the modification operation on the target memory information in the memory modification interface comprises: in response to the target memory information being in a selected state and detecting a de-selection operation on a selection control corresponding to the target memory information, removing the target memory information from the relevant memory information to obtain the modified memory information.

5. The method of claim 3, wherein, The target memory information belongs to the relevant memory information, and the obtaining of the modified memory information in response to the detection of the modification operation on the target memory information in the memory modification interface comprises: in response to the target memory information being in a selected state and detecting a trigger operation on an editing control corresponding to the target memory information, displaying an editing interface of the target memory information; in response to detecting an editing operation on the target memory information in the editing interface, obtaining edited memory information; obtaining the modified memory information according to the edited memory information.

6. The method of claim 3, wherein, The memory modification interface further comprises candidate memory information, and the target memory information belongs to the candidate memory information, and the obtaining of the modified memory information in response to the detection of the modification operation on the target memory information in the memory modification interface comprises: In response to the target memory information being in an unselected state and a selection operation on a selection control corresponding to the target memory information being detected, the target memory information is added to the related memory information to obtain the modified memory information. 7.The method of claim 3, further comprising: In response to detecting a triggering operation on the regeneration control, the memory modification interface is closed, and the modified memory information is displayed in the dialogue interface. 8.A smart interaction method, comprising: receiving input information of a user sent by a client; determining related memory information from memory information associated with the user according to the input information; generating reply information corresponding to the input information according to the related memory information; sending the related memory information and the reply information to the client.

9. The method of claim 8, wherein, The determining of the related memory information from the memory information associated with the user according to the input information comprises: searching for task information matching the input information in the associated memory information according to the input information; in response to finding the matching task information, generating preference inquiry information according to the matching task information; sending the preference inquiry information to the client; receiving an answer of the user to the preference inquiry information sent by the client, and determining the related memory information according to the answer of the preference inquiry information.

10. The method of claim 8, wherein, The determining of the related memory information from the memory information associated with the user according to the input information comprises: performing intent recognition on the input information to obtain a user intent; in response to the user intent including an intent to modify generated historical materials, querying historical dialogue data in the associated memory information according to the input information to obtain target materials; determining the related memory information according to the target materials.

11. The method of claim 10, wherein, The generating of the reply information corresponding to the input information according to the related memory information comprises: modifying the target materials in the related memory information according to the user intent to obtain modified materials; determining the reply information according to the modified materials.

12. The method of claim 8, wherein, The associated memory information includes memory information of a target type, and the method further comprises: determining a weight of the memory information of the target type according to an attribute parameter of the memory information of the target type, wherein the attribute parameter is used to represent a state of the memory information of the target type; in response to the weight being less than a preset threshold, deleting the memory information of the target type.

13. The method of claim 12, wherein, The determining of the weight of the memory information of the target type according to the attribute parameter of the memory information of the target type comprises: in response to the related memory information including the memory information of the target type, updating the attribute parameter to obtain an updated attribute parameter; determining the weight according to the updated attribute parameter.

14. The method of claim 12, wherein, The description form of the memory information of the target type comprises at least one of the following: a structured field form and a natural language abstract form. 15.A smart interaction device, comprising: An acquisition module is configured to acquire input information in response to detecting information input operation of a user in a conversation interface; A sending module is configured to send the input information to a server; A receiving module is configured to receive relevant memory information and reply information sent by the server; wherein the relevant memory information is determined by the server from memory information associated with the user according to the input information, and the reply information is generated by the server according to the relevant memory information; A display module is configured to display the relevant memory information and the reply information in the conversation interface.

16. An intelligent interaction device, comprising: A receiving module is configured to receive input information of a user sent by a client; A first determining module is configured to determine relevant memory information from memory information associated with the user according to the input information; A generating module is configured to generate reply information corresponding to the input information according to the relevant memory information; A sending module is configured to send the relevant memory information and the reply information to the client.

17. An electronic device, comprising: At least one processor; And A memory connected in communication with the at least one processor; wherein The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-14.

18. A non-transitory computer readable storage medium having stored thereon computer instructions, wherein, The computer instructions are used to enable the computer to perform the method according to any one of claims 1-14.

19. A computer program product comprising a computer program which, when executed by a processor, implements the steps of the method of any one of claims 1-14.

Images