AI large model calling method and device, equipment and medium

Abstract

The invention discloses an AI large model calling method and device, equipment and a medium, and relates to the field of data processing. The method is applied to the electronic equipment and comprises the steps that whether a target application program generates target data or not is monitored, and the target data at least comprises data newly stored under a target directory of the target application program and / or a system notification message sent by the target application program; under the condition that it is monitored that the target application program generates the target data, storing the target data into a memory database; under the condition that an input instruction for the input interface of the AI large model is received, reading to-be-processed data associated with the input instruction from the memory database; and processing the to-be-processed data according to the input indication and the AI large model to obtain a processing result aiming at the input indication. According to the method, the AI large model installed on the intelligent equipment can read the data of other applications installed in the intelligent equipment, and then the input indication for the data of other applications is fed back.

Description

Technical Field

[0001] The present application relates to the field of data processing, and more specifically, to a method, device, equipment, and medium for calling a large AI model. Background Art

[0002] With the rapid development of artificial intelligence technology, AI big models have gradually emerged, and their applications in various fields are becoming increasingly extensive. As a result, more and more smart devices are equipped with AI big models.

[0003] Currently, there is a problem with AI big models installed on smart devices that cannot read data from other applications installed on the smart device, and therefore cannot provide feedback on input instructions for data from other applications. For example, when a user of a smart device takes a photo with a beauty camera and inputs the instruction "How was the photo I just took?" into the AI ​​big model, the AI ​​big model cannot access the photos taken with the beauty camera and therefore cannot provide feedback on the user's input instruction "How was the photo I just took?" Summary of the Invention

[0004] One purpose of this application is to provide a new technical solution for calling large AI models.

[0005] According to a first aspect of the present application, a method for calling an AI large model is provided, which is applied to an electronic device, and the method includes:

[0006] Monitoring whether a target application generates target data, where the AI ​​large model and the target application are installed in the electronic device, and the target data includes at least newly stored data in a target directory of the target application and / or a system notification message sent by the target application;

[0007] When monitoring that the target application generates the target data, storing the target data in a memory database;

[0008] Upon receiving an input instruction for the input interface of the AI ​​large model, reading the to-be-processed data associated with the input instruction from the memory database;

[0009] The data to be processed is processed according to the input instruction and the AI ​​large model to obtain a processing result for the input instruction.

[0010] Optionally, the in-memory database includes an in-memory structured database and an in-memory unstructured database, and when monitoring that the target application generates the target data, storing the target data in the in-memory database includes:

[0011] When monitoring that the target application generates the target data and the target data is structured data, storing the target data in an in-memory structured database;

[0012] When monitoring that the target application generates the target data and the target data is unstructured data, storing the target data in an in-memory unstructured database;

[0013] The step of reading the to-be-processed data associated with the input instruction from the memory database includes:

[0014] Determining a demand data structure type according to the input indication, wherein the demand data structure type includes structured demand data and unstructured demand data;

[0015] In a case where the demand data structure type is structured demand data, reading the to-be-processed data associated with the input indication from the in-memory structured database;

[0016] In a case where the demand data structure type is unstructured demand data, the to-be-processed data associated with the input indication is read from the in-memory unstructured database.

[0017] Optionally, the memory database includes memory sub-databases corresponding to different business types, and when monitoring that the target application generates the target data, storing the target data in the memory database includes:

[0018] When it is monitored that the target application generates the target data, storing the target data in a memory sub-database matching the business type according to the business type of the target data;

[0019] The step of reading the to-be-processed data associated with the input instruction from the memory database includes:

[0020] Determining the required data service type according to the input indication;

[0021] The to-be-processed data associated with the input indication is read from the memory sub-database matching the business type.

[0022] Optionally, the method further includes:

[0023] Get the set data storage duration;

[0024] When the actual data storage time of the target data stored in the memory database reaches the set data storage time, the data is deleted from the memory database.

[0025] Optionally, processing the data to be processed according to the input instruction and the AI ​​large model to obtain a processing result for the input instruction includes:

[0026] In a case where the input indication includes sensitive information, performing data desensitization processing on the input indication to obtain a desensitized input indication;

[0027] Processing the data to be processed according to the desensitized input instruction and the AI ​​large model to obtain an initial processing result for the input instruction;

[0028] A processing result for the input indication is generated according to the sensitive information in the input indication and the initial processing result.

[0029] Optionally, when receiving an input instruction for the input interface of the AI ​​big model, before reading the to-be-processed data associated with the input instruction from the memory database of the AI ​​big model, the method further includes:

[0030] Detecting whether user input to an input interface of the AI ​​large model is received;

[0031] If yes, the user input is taken as the input indication.

[0032] Optionally, when receiving an input instruction for the input interface of the AI ​​big model, before reading the to-be-processed data associated with the input instruction from the memory database of the AI ​​big model, the method further includes:

[0033] In the case of monitoring the target application sending the system notification message, obtaining a current state, the current state including a current time and / or a current location;

[0034] Generate an inquiry type indication about the current state, and use the inquiry type indication as the input indication, wherein the inquiry type indication is used to inquire whether the AI ​​big model generates a reminder event in the current state.

[0035] According to a second aspect of the present application, a device for calling an AI large model is provided, which is applied to an electronic device, and the device includes:

[0036] a monitoring module, configured to monitor whether a target application generates target data, where the AI ​​large model and the target application are installed in the electronic device, and the target data includes at least newly stored data in a target directory of the target application and / or a system notification message sent by the target application;

[0037] A storage module is configured to store the target data in a memory database when monitoring that the target application generates the target data;

[0038] a reading module, configured to, upon receiving an input instruction for the input interface of the AI ​​large model, read the to-be-processed data associated with the input instruction from the memory database;

[0039] A processing module is used to process the data to be processed according to the input instruction and the AI ​​large model to obtain a processing result for the input instruction.

[0040] According to a third aspect of the present application, an electronic device is provided, comprising the apparatus according to the second aspect;

[0041] Alternatively, the electronic device includes a memory and a processor, the memory is used to store computer instructions, and the processor is used to call the computer instructions from the memory to execute the method as described in any one of the first aspects.

[0042] According to a fourth aspect of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method according to any one of the first aspects is implemented.

[0043] The present application provides a method for calling an AI big model, which is applied to an electronic device. The method includes: monitoring whether the target application generates target data, the electronic device is installed with an AI big model and a target application, and the target data at least includes newly stored data in the target directory of the target application and / or a system notification message sent by the target application; when the target application generates target data, the target data is stored in a memory database; when an input indication for the input interface of the AI ​​big model is received, the data to be processed associated with the input indication is read from the memory database; the data to be processed is processed according to the input indication and the AI ​​big model to obtain a processing result for the input indication. This method can enable the AI ​​big model installed on the smart device to read the data of other applications installed on the smart device, and then provide feedback on the input indication for the data of other applications.

[0044] Other features and advantages of the present application will become apparent from the following detailed description of exemplary embodiments of the present application with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the application and, together with the description, serve to explain the principles of the application.

[0046] Figure 1 This is a hardware configuration framework of an electronic device for implementing a method for calling an AI large model according to an embodiment of the present application. Figure 1 ;

[0047] Figure 2 This is a flowchart of a method for implementing a large AI model call according to an embodiment of the present application;

[0048] Figure 3 This is a structural diagram of a device for implementing a large AI model according to an embodiment of the present application;

[0049] Figure 4 This is a hardware configuration framework of an electronic device for implementing a method for calling an AI large model according to an embodiment of the present application. Figure 2 . DETAILED DESCRIPTION

[0050] Various exemplary embodiments of the present application will now be described in detail with reference to the accompanying drawings. It should be noted that unless otherwise specifically stated, the relative arrangements of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present application.

[0051] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present disclosure, its application, or uses.

[0052] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0053] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0054] It should be noted that like reference numerals and letters refer to like items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0055] Figure 1 This is a hardware configuration framework of an electronic device for implementing a method for calling an AI large model according to an embodiment of the present application. Figure 1 .

[0056] The electronic device 1000 may be a terminal or a server. Further, the terminal may be a head-mounted device (such as an AR device, an MR device, and a VR device), a portable computer, a tablet computer, a PDA, etc. The server may be a cloud server, etc.

[0057] Electronic device 1000 may include a processor 1100, a memory 1200, an interface device 1300, a communication device 1400, a display device 1500, an input device 1600, a speaker 1700, a microphone 1800, and the like. Processor 1100 may be a central processing unit (CPU), a microprocessor (MCU), or the like. Memory 1200 may include, for example, a ROM (read-only memory), a RAM (random access memory), or a non-volatile memory such as a hard disk. Interface device 1300 may include, for example, a USB interface or a headphone jack. Communication device 1400 may be capable of wired or wireless communication. Display device 1500 may be, for example, an LCD display or a touchscreen display. Input device 1600 may include, for example, a touchscreen or a keyboard. A user may input / output voice information through speaker 1700 and microphone 1800.

[0058] Despite Figure 1 Multiple devices are shown for the electronic device 1000, but the present application may only involve some of the devices, for example, the electronic device 1000 only involves the memory 1200 and the processor 1100.

[0059] In the embodiment of the present application, the memory 1200 of the electronic device 1000 is used to store instructions, which are used to control the processor 1100 to execute the calling method of the AI ​​large model provided in the embodiment of the present application.

[0060] In the above description, a person skilled in the art can design instructions according to the solution disclosed in this application. How instructions control the operation of a processor is well known in the art and will not be described in detail here.

[0061] This application provides a method for calling an AI large model, which is applied to Figure 1 The electronic equipment shown. Figure 2 As shown, the calling method of the AI ​​large model provided in this application includes the following steps S2100 to S2400.

[0062] Step S2100: monitor whether the target application generates target data.

[0063] The electronic device has an AI large model and a target application installed, and the target data includes at least newly stored data in the target application's target directory and / or system notification messages sent by the target application. Specifically, if new data is stored in the target application's target directory and / or the target application sends a system notification message, the target application is deemed to have generated the target data; otherwise, the target application is deemed to have not generated the target data.

[0064] In this embodiment, the method for calling the AI ​​large model provided by this application can be implemented through an application, and in order to make the application available to users at any time, a background service is established for the application. In this way, even after the application is exited, the method for calling the AI ​​large model provided by this application can still be run.

[0065] Furthermore, the AI ​​big model can be a local AI big model or a cloud big model.

[0066] The target application is an application that is allowed to be accessed by the calling method of the AI ​​large model provided by this application, and can be specified by the user through permission settings according to their own needs. In this way, the data of non-target applications will not be monitored by the calling method of the AI ​​large model provided by this application. On this basis, the user's data security can be guaranteed. It should be noted that this embodiment does not limit the number of target applications. In one example, the target applications are a beauty camera, a sports health application, and a courier dynamic notification application.

[0067] Furthermore, the target data is the data in the target application that is allowed to be read by the calling method of the AI ​​large model provided by this application, including at least the newly stored data in the target directory of the target application and / or the system notification message sent by the target application, and can be specified by the user through permission settings according to their own needs. It should be noted that this application does not limit the type and number of target data. Among them, the target directory is, for example, a media file directory, etc., and the system notification message is, for example, a schedule agent reminder, reminder information, text message, etc.

[0068] In one embodiment of the present application, a ContentObserver can be registered to monitor a fixed address directory to detect whether new data is stored in the target directory of the target application. Furthermore, by inheriting NotificationListenerService, the target application can be monitored to detect whether it sends system notifications. Furthermore, the content of the system notification message can be captured through onNotificationPosted, the message title can be obtained through Notification.EXTRA_TITLE, and the message content can be obtained through Notification.EXTRA_TEXT.

[0069] In one example, if the target application is a beauty camera, the target data is photos or videos taken with the beauty camera. If the target application is a courier dynamic notification application, the target data is pickup information including the pickup code, pickup address, and courier items.

[0070] Taking the target application as a beauty camera and the target data as photos taken by the beauty camera as an example, after the user uses the beauty camera to take photos, that is, the beauty camera collects new photos, the calling method of the AI ​​large model provided in this application monitors that new data is stored in the media file directory of the beauty camera. At this time, it is determined that the target application is monitored to generate the target data.

[0071] Taking the target application as the express dynamic notification application and the target data as the pickup information sent by the express dynamic notification application as an example, when the express dynamic notification application outputs the pickup information to the user, the calling method of the AI ​​large model provided in this application monitors the express dynamic notification application sending a system notification message. At this time, it is determined that the target application is monitored to generate the target data.

[0072] Based on the above step S2100, the method for calling the AI ​​large model provided by this application will monitor in real time or periodically whether the target application generates target data. Thereafter, the following step S2200 is executed.

[0073] Step S2200: When it is monitored that the target application generates target data, the target data is stored in the memory database.

[0074] In an example, taking the target application as a beauty camera and the target data as photos taken by the beauty camera, after the user uses the beauty camera to take photos, that is, the beauty camera collects new photos, the calling method of the AI ​​large model provided in this application monitors the new data stored in the media file directory received by the beauty camera. At this time, the new photos taken by the beauty camera are stored in the memory database.

[0075] In another example, taking the target application as a courier dynamic notification application and the target data as the pickup information sent by the courier dynamic notification application, when the courier dynamic notification application outputs the pickup information to the user, the calling method of the AI ​​large model provided in this application monitors the courier dynamic notification application sending a system notification message. At this time, the system notification message sent by the courier dynamic notification application is stored in the in-memory database.

[0076] In this embodiment, after the target application generates the target data through monitoring in the above step S2100, the target data is stored in the memory database. In this way, the calling method of the AI ​​large model provided by this application can read the data stored in the memory database at any time. In addition, since the target data is stored in the memory database, when the calling method of the AI ​​large model provided by this application is closed due to reasons such as the electronic device being turned off, the memory database immediately releases the stored data. In this way, while ensuring the privacy and security of user data, the amount of data stored is reduced, thereby reducing the complexity and maintenance cost of the calling method of the AI ​​large model provided by this application.

[0077] Step S2300: Upon receiving an input instruction from the input interface of the AI ​​large model, read the data to be processed that matches the input instruction from the memory database.

[0078] In this embodiment, the input instruction may be input by the user, and / or the input instruction may be automatically generated by the method for invoking the AI ​​large model provided in this application. In this regard, when the input instruction is input by the user, the method for invoking the AI ​​large model provided in this application further includes the following steps S2310 and S2311 before the above step S2300.

[0079] Step S2310: Detect whether user input to the input interface of the AI ​​large model is received.

[0080] In this embodiment, if a user input is detected on the input interface of the AI ​​large model, such as text or voice input such as "How was the photo I just took?", then it is determined that user input targeting the input interface of the AI ​​large model has been received. Conversely, if no user input is detected on the input interface of the AI ​​large model, then it is determined that no user input targeting the input interface of the AI ​​large model has been received.

[0081] Step S2311: If yes, take the user input as an input instruction.

[0082] When it is determined that user input to the input interface of the AI ​​big model is received, the user input is taken as an input indication.

[0083] For the above steps S2310 and S2311, in the case where the user inputs the instruction "How was the photo I just took?", based on the above step S2300, the latest photo taken by the beauty camera as the data to be processed associated with the input instruction can be extracted from the memory database.

[0084] In addition, for the automatic generation of input instructions by the calling method of the AI ​​big model provided by this application, the calling method of the AI ​​big model provided by this application also includes the following steps S2312 and S2313 before the above-mentioned step S2300.

[0085] Step S2312: When monitoring the target application sending a system notification message, obtain the current status.

[0086] The current state includes the current time and / or the current location.

[0087] In this embodiment, since the system notification message needs to remind the user, when the system notification message is monitored, the current status is obtained, and the following step S2313 is executed.

[0088] Step S2313: Generate an inquiry type indication about the current state, and use the inquiry type indication as an input indication. The inquiry type indication is used to inquire whether the AI ​​big model generates a reminder event in the current state.

[0089] In this embodiment, based on the above step S2313, an inquiry-type instruction such as "whether to remind the user to pick up the item at the current location a and / or the current time A" can be generated.

[0090] For the above steps S2312 and S2313, when the input indication is an inquiry-type indication of "whether the user needs to be reminded to pick up the package at the current location a and / or the current time A", based on the above step S2300, the pickup information sent by the express dynamic notification application as the to-be-processed data associated with the input indication can be extracted from the memory database.

[0091] Step S2400: Process the data to be processed according to the input instruction and the AI ​​large model to obtain a processing result for the input instruction.

[0092] In this embodiment, input instructions and data to be processed are input into the AI ​​big model, and the AI ​​big model outputs the processing results for the input instructions.

[0093] In an example, the specific implementation of the above step S2400 is: input "How is the photo just taken?" and the latest photo taken by the beauty camera into the AI ​​big model, and the AI ​​big model analyzes the latest photo taken by the beauty camera and obtains the evaluation results for the latest photo taken by the beauty camera.

[0094] In another example, the specific implementation of the above step S2400 is: input "whether it is necessary to remind the user to pick up the package at the current location a and / or the current time A" and the pickup information sent by the express dynamic notification application stored in the memory database into the AI ​​big model, and the AI ​​big model analyzes the pickup information according to the current location a and / or the current time A to determine whether to generate a reminder event to remind the user to pick up the package, and when it is determined that a reminder event to remind the user to pick up the package is to be generated, a reminder event is generated, and the reminder event is used as the processing result for the input instruction and the data to be processed. In this way, through the calling method of the AI ​​big model provided by this application, it is possible to remind the user that there is a package to be picked up when the user passes by the pickup address attachment.

[0095] Through the above steps S2100 to S2400, the AI ​​big model installed on the smart device can read the data of other applications installed on the smart device and then provide feedback on the input instructions for the data of other applications. For example, when the user of the smart device takes a photo with the beauty camera and inputs the instruction "How was the photo I just took?" into the AI ​​big model, the AI ​​big model calls the most recent photo taken by the beauty camera and obtains the evaluation result for the most recent photo taken by the beauty camera.

[0096] In one embodiment of the present application, the specific implementation of the above-mentioned step S2400 can be: when the input indication information includes sensitive information, determining that the AI ​​big model is a local AI big model; processing the data to be processed according to the input indication and the local AI big model, and obtaining a processing result for the input indication; correspondingly, when the input indication information includes non-sensitive information, determining that the AI ​​big model is a cloud-based AI big model; processing the data to be processed according to the input indication and the cloud-based AI big model, and obtaining a processing result for the input indication.

[0097] In one embodiment of the present application, the above-mentioned step S2400 can also be specifically implemented through the following steps S2410 to S2412.

[0098] Step S2410: When the input indication includes sensitive information, data desensitization is performed on the input indication to obtain a desensitized input indication.

[0099] In this embodiment, the sensitive information may be, for example, name, amount, verification code, etc. Furthermore, the sensitive information may be pre-specified by the user.

[0100] If the input instruction includes sensitive information, data desensitization is performed on the input instruction to hide the sensitive information and obtain the desensitized input instruction. For example, if the input instruction includes the name "Zhang Moumou", the name "Zhang Moumou" in the input instruction is replaced with "Zhang San" to obtain the desensitized input instruction.

[0101] Step S2411, processes the data to be processed according to the desensitized input instruction and the AI ​​large model to obtain the initial processing result for the input instruction.

[0102] In this embodiment, the desensitized input indication and the data to be processed are input into the AI ​​large model processing, and the result output by the AI ​​large model processing is recorded as the initial processing result.

[0103] Step S2412: Generate a processing result for the input instruction based on the sensitive information in the input instruction and the initial processing result.

[0104] In this embodiment, the hidden sensitive information is added to the initial processing result to generate a processing result for the input instruction. For example, "Zhang San" in the initial processing result is restored to "Zhang Moumou" to obtain a processing result for the input instruction.

[0105] The present application provides a method for calling an AI big model, which is applied to an electronic device. The method includes: monitoring whether the target application generates target data, the electronic device is installed with an AI big model and a target application, and the target data at least includes newly stored data in the target directory of the target application and / or a system notification message sent by the target application; when the target application generates target data, the target data is stored in a memory database; when an input indication for the input interface of the AI ​​big model is received, the data to be processed associated with the input indication is read from the memory database; the data to be processed is processed according to the input indication and the AI ​​big model to obtain a processing result for the input indication. This method can enable the AI ​​big model installed on the smart device to read the data of other applications installed on the smart device, and then provide feedback on the input indication for the data of other applications.

[0106] In one embodiment of the present application, the calling method of the AI ​​big model provided in the present application also includes the following steps S2500 and S2600.

[0107] Step S2500: Obtain the set data storage duration.

[0108] In this embodiment, the data storage duration is set to be the effective duration of the data in the memory database, which can be set based on experience. In one example, the data storage duration is set to 5 days.

[0109] Step S2600: When the actual data storage time of the target data stored in the memory database reaches the set data storage time, the data is deleted from the memory database.

[0110] Through the above step S2600, data stored in the memory database for too long can be deleted, which can reduce data maintenance costs.

[0111] In one embodiment of the present application, the in-memory database includes an in-memory structured database and an in-memory unstructured database. On this basis, the above step S2200 is specifically implemented through the following steps S2210 and S2211.

[0112] Step S2210: When it is monitored that the target application generates target data and the target data is structured data, the target data is stored in an in-memory structured database.

[0113] Step S2211: When it is monitored that the target application generates target data and the target data is unstructured data, the target data is stored in the in-memory unstructured database.

[0114] In this embodiment, structured data refers to data organized in a fixed format or predefined model. In one example, the structured data is pickup information sent by a courier dynamic notification application.

[0115] Correspondingly, unstructured data refers to data that does not have a predefined data model or fixed format. In one example, the unstructured data is photos taken with a beauty camera.

[0116] In this embodiment, structured target data is stored in an in-memory structured database. Unstructured target data is stored in an in-memory unstructured database. This facilitates rapid access to the to-be-processed data associated with the input instruction when executing step S2300 below.

[0117] In combination with the above content, on the basis of the above steps S2210 and S2211, the above step S2300 is specifically implemented through the following steps S2320 to S2322.

[0118] Step S2320: Determine the demand data structure type according to the input instruction. The demand data structure type includes structured demand data and unstructured demand data.

[0119] In this embodiment, the data object that requires AI big model analysis is determined based on the input indication, and further, the data structure type is determined based on the data object that requires AI big model analysis.

[0120] In one example, if the input instruction is "How was the photo I just took?", the data object analyzed by the demand AI big model is determined to be an image, and the data structure type is determined to be unstructured demand data. If the input instruction is "Do I need to be reminded to pick up my package at the current location a and / or the current time A?", the data object analyzed by the demand AI big model is determined to be pickup information, and the data structure type is determined to be structured demand data.

[0121] Step S2321: When the demand data structure type is structured demand data, read the to-be-processed data associated with the input instruction from the in-memory structured database.

[0122] Step S2322: When the demand data structure type is unstructured demand data, read the to-be-processed data associated with the input instruction from the in-memory unstructured database.

[0123] Through the above steps S2321 and S2322, the to-be-processed data associated with the input indication can be quickly read.

[0124] In one embodiment of the present application, the memory database includes memory sub-databases corresponding to different business types. On this basis, the above step S2200 is specifically implemented through the following step S2220.

[0125] Step S2220 , when it is monitored that the target application generates target data, the target data is stored in a target memory sub-database corresponding to the business type according to the business type of the target data.

[0126] In this embodiment, when a target application generates target data, the target data is first analyzed for its business type. Furthermore, the target data is stored in a memory sub-database that matches the target application's business type. Business types include photography, shopping, pickup, and payment.

[0127] In one embodiment of the present application, when the target data includes data of at least two business types, the target data can be split according to the business type to obtain target sub-data. Further, the split target sub-data is stored in a memory sub-database that matches the business type.

[0128] Through the above step S2220, the target data can be stored in different memory sub-databases in different categories, so as to facilitate the rapid reading of the to-be-processed data associated with the input instruction when executing the following step S2300.

[0129] In combination with the above content, on the basis of the above step S2220, the above step S2300 is specifically implemented through the following steps S2330 and S2331.

[0130] Step S2330: Determine the required data service type according to the input instruction.

[0131] Step S2331: Read the to-be-processed data associated with the input instruction from the memory sub-database that matches the business type.

[0132] In this embodiment, the data object to be analyzed by the AI ​​large model is determined based on the input instruction. Furthermore, the data service type is determined based on the data object analyzed by the AI ​​large model. Furthermore, the data to be processed associated with the input instruction is read from the memory sub-database that matches the determined data service type. This allows for rapid access to the data to be processed associated with the input instruction.

[0133] This application also provides a device 300 for calling an AI large model, such as Figure 3 , applied to electronic equipment, the device 300 includes:

[0134] A monitoring module 310 is configured to monitor whether a target application generates target data, where the AI ​​large model and the target application are installed in the electronic device, and the target data includes at least newly stored data in a target directory of the target application and / or a system notification message sent by the target application;

[0135] The storage module 320 is configured to store the target data in a memory database when monitoring that the target application generates the target data;

[0136] A reading module 330 is configured to, upon receiving an input instruction for the input interface of the AI ​​large model, read the to-be-processed data associated with the input instruction from the memory database;

[0137] The processing module 340 is used to process the data to be processed according to the input instruction and the AI ​​large model to obtain a processing result for the input instruction.

[0138] In one embodiment of the present application, the in-memory database includes an in-memory structured database and an in-memory unstructured database, and the storage module 320 is specifically configured to: upon monitoring that the target application generates the target data, and the target data is structured data, store the target data in the in-memory structured database;

[0139] When monitoring that the target application generates the target data and the target data is unstructured data, storing the target data in an in-memory unstructured database;

[0140] The reading module 330 is specifically configured to determine the demand data structure type according to the input indication, where the demand data structure type includes structured demand data and unstructured demand data;

[0141] In a case where the demand data structure type is structured demand data, reading the to-be-processed data associated with the input indication from the in-memory structured database;

[0142] In a case where the demand data structure type is unstructured demand data, the to-be-processed data associated with the input indication is read from the in-memory unstructured database.

[0143] In one embodiment of the present application, the memory database includes memory sub-databases corresponding to different business types, and the storage module 320 is specifically configured to: upon monitoring the target application generating the target data, store the target data in the memory sub-database matching the business type according to the business type of the target data;

[0144] The reading module 330 is specifically configured to determine the required data service type according to the input indication;

[0145] The to-be-processed data associated with the input indication is read from the memory sub-database matching the business type.

[0146] In one embodiment of the present application, the AI ​​large model calling device 300 provided in the present application further includes:

[0147] The acquisition module is used to obtain the set data storage time;

[0148] The deleting module is configured to delete the data from the memory database when the actual data storage time of the target data stored in the memory database reaches the set data storage time.

[0149] In one embodiment of the present application, the processing module 340 is specifically configured to, when the input indication includes sensitive information, perform data desensitization processing on the input indication to obtain a desensitized input indication;

[0150] Processing the data to be processed according to the desensitized input instruction and the AI ​​large model to obtain an initial processing result for the input instruction;

[0151] A processing result for the input indication is generated according to the sensitive information in the input indication and the initial processing result.

[0152] In one embodiment of the present application, the AI ​​large model calling device 300 provided in the present application further includes:

[0153] A first determination module is configured to detect whether a user input to an input interface of the AI ​​large model is received;

[0154] If yes, the user input is taken as the input indication.

[0155] In one embodiment of the present application, the AI ​​large model calling device 300 provided in the present application further includes:

[0156] A second determining module is configured to obtain a current state when monitoring the target application sending the system notification message, where the current state includes a current time and / or a current location;

[0157] Generate an inquiry type indication about the current state, and use the inquiry type indication as the input indication, wherein the inquiry type indication is used to inquire whether the AI ​​big model generates a reminder event in the current state.

[0158] The present application also provides an electronic device, comprising a device 300 for calling any one of the AI ​​large models provided in the above device embodiments;

[0159] Or, as Figure 4 As shown, the electronic device 400 includes a memory 410 and a processor 420, the memory 410 is used to store computer instructions, and the processor 420 is used to call the computer instructions from the memory 410 to execute any one of the methods described above.

[0160] The present application also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method according to any one of the above method embodiments is implemented.

[0161] The present application may be a system, method and / or computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present application.

[0162] A computer-readable storage medium can be a tangible device that can hold and store instructions for use by an instruction execution device. A computer-readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination thereof. As used herein, a computer-readable storage medium is not to be construed as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through an electrical wire.

[0163] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions to be stored in the computer-readable storage medium in each computing / processing device.

[0164] The computer program instructions for performing the operation of the present application can be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data or source code or object code written in any combination of one or more programming languages, wherein the programming language includes object-oriented programming languages ​​such as Smalltalk, C++, and conventional procedural programming languages ​​such as "C" language or similar programming languages. Computer-readable program instructions can be executed completely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or executed completely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer by any type of network including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (such as by using an Internet service provider to connect to the Internet). In certain embodiments, by utilizing the state information of computer-readable program instructions to personalize electronic circuits, such as programmable logic circuits, field programmable gate arrays (FPGAs) or programmable logic arrays (PLAs), the electronic circuits can execute computer-readable program instructions, thereby realizing various aspects of the present application.

[0165] Various aspects of the present application are described herein with reference to flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present application. It should be understood that each block of the flowcharts and / or block diagrams, and combinations of blocks in the flowcharts and / or block diagrams, can be implemented by computer-readable program instructions.

[0166] These computer-readable program instructions can be provided to a processor 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 processor of the computer or other programmable data processing device, a device is generated that implements the functions / actions specified in one or more blocks in the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium, where these instructions cause the computer, programmable data processing device, and / or other device to operate in a specific manner. Thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing various aspects of the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0167] Computer-readable program instructions may also 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, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more blocks in the flowchart and / or block diagram.

[0168] The flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a part of a module, program segment or instruction, and the part of the module, program segment or instruction contains one or more executable instructions for realizing the specified logical function. In some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions. It is well known to those skilled in the art that implementation by hardware, implementation by software, and implementation by a combination of software and hardware are all equivalent.

[0169] The embodiments of the present application have been described above. The above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terms used herein are selected to best explain the principles of the embodiments, practical applications, or technical improvements to technologies in the market, or to enable other persons skilled in the art to understand the embodiments disclosed herein. The scope of this application is defined by the appended claims.

Claims

1. A method for calling a large AI model, characterized in that: Applied to electronic equipment, the method includes: Monitoring whether a target application generates target data, where the AI ​​large model and the target application are installed in the electronic device, and the target data includes at least newly stored data in a target directory of the target application and / or a system notification message sent by the target application; When monitoring that the target application generates the target data, storing the target data in a memory database; Upon receiving an input instruction for the input interface of the AI ​​large model, reading the to-be-processed data associated with the input instruction from the memory database; The data to be processed is processed according to the input instruction and the AI ​​large model to obtain a processing result for the input instruction.

2. The method according to claim 1, characterized in that The in-memory database includes an in-memory structured database and an in-memory unstructured database. When monitoring that the target application generates the target data, storing the target data in the in-memory database includes: When monitoring that the target application generates the target data and the target data is structured data, storing the target data in an in-memory structured database; When monitoring that the target application generates the target data and the target data is unstructured data, storing the target data in an in-memory unstructured database; The step of reading the to-be-processed data associated with the input instruction from the memory database includes: Determining a demand data structure type according to the input indication, wherein the demand data structure type includes structured demand data and unstructured demand data; In a case where the demand data structure type is structured demand data, reading the to-be-processed data associated with the input indication from the in-memory structured database; In a case where the demand data structure type is unstructured demand data, the to-be-processed data associated with the input indication is read from the in-memory unstructured database.

3. The method according to claim 1, characterized in that The memory database includes memory sub-databases corresponding to different business types. When monitoring that the target application generates the target data, storing the target data in the memory database includes: When it is monitored that the target application generates the target data, storing the target data in a memory sub-database matching the business type according to the business type of the target data; The step of reading the to-be-processed data associated with the input instruction from the memory database includes: Determining the required data service type according to the input indication; The to-be-processed data associated with the input indication is read from the memory sub-database matching the business type.

4. The method according to claim 1, wherein The method further comprises: Get the set data storage duration; When the actual data storage time of the target data stored in the memory database reaches the set data storage time, the data is deleted from the memory database.

5. The method according to claim 1, wherein The processing of the data to be processed according to the input instruction and the AI ​​large model to obtain a processing result for the input instruction includes: In a case where the input indication includes sensitive information, performing data desensitization processing on the input indication to obtain a desensitized input indication; Processing the data to be processed according to the desensitized input instruction and the AI ​​large model to obtain an initial processing result for the input instruction; A processing result for the input indication is generated according to the sensitive information in the input indication and the initial processing result.

6. The method according to any one of claims 1 to 5, characterized in that When receiving an input instruction for the input interface of the AI ​​large model, before reading the to-be-processed data associated with the input instruction from the memory database of the AI ​​large model, the method further includes: Detecting whether user input to an input interface of the AI ​​large model is received; If yes, the user input is taken as the input indication.

7. The method according to claim 6, characterized in that When receiving an input instruction for the input interface of the AI ​​large model, before reading the to-be-processed data associated with the input instruction from the memory database of the AI ​​large model, the method further includes: In the case of monitoring the target application sending the system notification message, obtaining a current state, the current state including a current time and / or a current location; Generate an inquiry type indication about the current state, and use the inquiry type indication as the input indication, wherein the inquiry type indication is used to inquire whether the AI ​​big model generates a reminder event in the current state.

8. A device for calling a large AI model, characterized in that: Applied to electronic equipment, the device comprises: a monitoring module, configured to monitor whether a target application generates target data, where the AI ​​large model and the target application are installed in the electronic device, and the target data includes at least newly stored data in a target directory of the target application and / or a system notification message sent by the target application; A storage module is configured to store the target data in a memory database when monitoring that the target application generates the target data; a reading module, configured to, upon receiving an input instruction for the input interface of the AI ​​large model, read the to-be-processed data associated with the input instruction from the memory database; A processing module is used to process the data to be processed according to the input instruction and the AI ​​large model to obtain a processing result for the input instruction.

9. An electronic device, characterized in that: The electronic device comprises the apparatus according to claim 8; Alternatively, the electronic device includes a memory and a processor, the memory is used to store computer instructions, and the processor is used to call the computer instructions from the memory to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that A computer program is stored thereon, which implements the method according to any one of claims 1 to 7 when executed by a processor.

Images