Information processing method and device, computer device and storage medium

Abstract

The application discloses an information processing method and device, computer equipment and a storage medium, and relates to the technical field of artificial intelligence. The method is applied to a server and includes the following steps: the server receives first request information sent by a first electronic device; a generative large language model corresponding to a first user identifier is determined from a plurality of generative large language models as a target model; the first request information is input into the target model to obtain first feedback information corresponding to the first request information; and the first feedback information is sent to the first electronic device. In this way, the target model learns the preferences and speaking habits of the user corresponding to the first user identifier, that is, the target model is obtained by targeted and customized training of the user corresponding to the first user identifier. Therefore, the target model can more accurately generate first feedback information for the first request information, that is, the accuracy of information feedback is improved, and the feedback information is more personalized.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to an information processing method, apparatus, computer equipment, and storage medium. Background Technology

[0002] Large language models (LLMs) are deep learning models trained on large amounts of text data that can generate natural language text or understand the meaning of language text. LLMs can handle various natural language tasks, such as text classification, question answering, and dialogue, and are an important pathway to artificial intelligence.

[0003] However, the relevant technologies cannot provide more personalized processing of user requests based on different users, resulting in inaccurate feedback on different user requests. Summary of the Invention

[0004] This application proposes an information processing method, apparatus, computer equipment, and storage medium to improve the accuracy of information feedback.

[0005] In a first aspect, embodiments of this application provide an information processing method applied to a server. The method includes: receiving first request information sent by a first electronic device, the first request information carrying a first user identity identifier of the current user of the first electronic device; determining a generative large language model corresponding to the first user identity identifier from multiple generative large language models as a target model, wherein the multiple generative large language models correspond one-to-one with multiple preset user identity identifiers, the generative large language models are used to perform intent recognition on the request information and output feedback information corresponding to the intent recognition result, the target model is pre-trained based on a first training sample set associated with the first user identity identifier, the first training sample set is generated based on context information corresponding to the first user identity identifier in at least one application installed on the first electronic device; inputting the first request information into the target model to obtain first feedback information corresponding to the first request information; and sending the first feedback information to the first electronic device.

[0006] Secondly, embodiments of this application provide an information processing method applied to a first electronic device. The method includes: sending first request information to a server, the first request information carrying a first user identity identifier of the current user of the first electronic device; receiving first feedback information sent by the server based on the first request information, the first feedback information being a generative large language model determined by the server from multiple generative large language models corresponding to the first user identity identifier, as a target model; inputting the first request information into the target model to obtain the first feedback information corresponding to the first request information, wherein the multiple generative large language models correspond one-to-one with multiple preset user identity identifiers, the generative large language models are used to perform intent recognition on the request information, and output feedback information corresponding to the intent recognition result, the target model is pre-trained based on a first training sample set associated with the first user identity identifier, the first training sample set is generated based on context information corresponding to the first user identity identifier in at least one application installed on the first electronic device; and outputting the first feedback information.

[0007] Thirdly, embodiments of this application provide an information processing apparatus applied to a server. The apparatus includes: an information receiving module for receiving first request information sent by a first electronic device, the first request information carrying a first user identity identifier of the current user of the first electronic device; a model determining module for determining a generative large language model corresponding to the first user identity identifier from multiple generative large language models as a target model, wherein the multiple generative large language models correspond one-to-one with multiple preset user identity identifiers, the generative large language model is used to perform intent recognition on the request information and output feedback information corresponding to the intent recognition result, the target model is pre-trained based on a first training sample set associated with the first user identity identifier, the first training sample set is generated based on context information corresponding to the first user identity identifier in at least one application installed on the first electronic device; a feedback information acquisition module for inputting the first request information into the target model to obtain first feedback information corresponding to the first request information; and an information feedback module for sending the first feedback information to the first electronic device.

[0008] Fourthly, embodiments of this application provide an information processing apparatus applied to a first electronic device. The apparatus includes: an information sending module for sending first request information to a server, the first request information carrying a first user identity identifier of the current user of the first electronic device; an information receiving module for receiving first feedback information sent by the server based on the first request information, the first feedback information being a generative large language model determined by the server from multiple generative large language models corresponding to the first user identity identifier, serving as a target model; and inputting the first request information into the target model to obtain the first feedback information corresponding to the first request information, wherein the multiple generative large language models correspond one-to-one with multiple preset user identity identifiers, the generative large language models are used to perform intent recognition on the request information and output feedback information corresponding to the intent recognition result, the target model is pre-trained based on a first training sample set associated with the first user identity identifier, the first training sample set being generated based on context information corresponding to the first user identity identifier in at least one application installed on the first electronic device; and an information output module for outputting the first feedback information.

[0009] Fifthly, embodiments of this application provide a computer device, including: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs are configured to perform the methods described above.

[0010] Sixthly, embodiments of this application provide a computer-readable storage medium storing program code that can be invoked by a processor to execute the above-described method.

[0011] In the solution provided in this application, a server receives a first request message sent by a first electronic device, wherein the first request message carries a first user identity identifier of the current user of the first electronic device; a generative large language model corresponding to the first user identity identifier is determined from multiple generative large language models as a target model; the multiple generative large language models correspond one-to-one with multiple preset user identity identifiers, and the generative large language models are used to perform intent recognition on the request message and output feedback information corresponding to the intent recognition result; the first request message is input into the target model to obtain first feedback information corresponding to the first request message; and the first feedback information is sent to the first electronic device. Since the target model is pre-trained based on a first training sample set associated with the first user identity identifier, and the first training sample set is generated based on contextual information corresponding to the first user identity identifier in at least one application installed on the first electronic device, the target model can learn the preferences and speaking habits of the user corresponding to the first user identity identifier. This is equivalent to the target model being trained specifically and customized for the user corresponding to the first user identity identifier. Therefore, using the target model, the first feedback information corresponding to the first request message can be generated more accurately, thus improving the accuracy of the feedback and making the feedback information more personalized. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 A schematic diagram of the system architecture of an information processing system 10 provided in one embodiment of this application is shown.

[0014] Figure 2 A flowchart illustrating an information processing method provided in an embodiment of this application is shown.

[0015] Figure 3 A flowchart illustrating an information processing method provided in another embodiment of this application is shown.

[0016] Figure 4 A flowchart illustrating an information processing method provided in another embodiment of this application is shown.

[0017] Figure 5 A flowchart illustrating an information processing method provided in another embodiment of this application is shown.

[0018] Figure 6This is a block diagram of an information processing apparatus according to an embodiment of this application.

[0019] Figure 7 This is a block diagram of an information processing apparatus according to another embodiment of this application.

[0020] Figure 8 This is a block diagram of a computer device for performing an information processing method according to an embodiment of this application.

[0021] Figure 9 This is a storage unit in this application embodiment for storing or carrying program code that implements the information processing method according to this application embodiment. Detailed Implementation

[0022] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of the present application, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative effort are within the scope of protection of the present application.

[0023] It should be noted that some processes described in the specification, claims, and accompanying drawings of this application include multiple operations that appear in a specific order. These operations may not be performed in the order they appear herein, or they may be performed in parallel. Operation numbers such as S110, S120, etc., are merely used to distinguish different operations and do not represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be performed sequentially or in parallel. Also, the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having”, and any variations thereof, are intended to cover non-exclusive inclusion, such that a process, method, system, product, or server that includes a series of steps or sub-modules is not necessarily limited to those steps or sub-modules that are explicitly listed, but may include other steps or sub-modules that are not explicitly listed or that are inherent to such process, method, product, or device.

[0024] The inventors have proposed an information processing method, apparatus, computer device, and storage medium. The information processing method provided in the embodiments of this application will be described in detail below.

[0025] Please see Figure 1 , Figure 1 This is a schematic diagram of the system architecture of an information processing system 10 provided in an embodiment of this application. The information processing system 10 includes at least a server 101 and a first electronic device 102.

[0026] Server 101 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery network (CDN) acceleration services, and big data and artificial intelligence platforms. The first electronic device 102 includes, but is not limited to, smartphones, tablets, laptops, desktop computers, smartwatches, e-book readers, MP3 (Moving Picture Experts Group Audio Layer III) players, MP4 (Moving Picture Experts Group Audio Layer IV) players, and smart speakers.

[0027] Understandably, server 101 and first electronic device 102 can be directly or indirectly connected via wired or wireless communication.

[0028] In some implementations, server 101 may receive a first request message sent by first electronic device 102, the first request message carrying a first user identity identifier of the current user of first electronic device 102; determine a generative large language model corresponding to the first user identity identifier from multiple generative large language models as a target model, the multiple generative large language models correspond one-to-one with multiple preset user identity identifiers, the generative large language model is used to perform intent recognition on the request message, and output feedback information corresponding to the intent recognition result, the target model is pre-trained based on a first training sample set associated with the first user identity identifier, the first training sample set is generated based on context information corresponding to the first user identity identifier in at least one application installed on the first electronic device; input the first request message into the target model to obtain a first feedback message corresponding to the first request message; and send the first feedback message to first electronic device 102.

[0029] Please refer to Figure 2 , Figure 2 This is a flowchart illustrating an information processing method according to an embodiment of this application, applied to a server. The following will be combined with... Figure 2The information processing method provided in the embodiments of this application will be described in detail. This information processing method may include the following steps:

[0030] Step S210: Receive a first request message sent by a first electronic device, wherein the first request message carries a first user identity identifier of the current user of the first electronic device.

[0031] Step S220: Determine the generative large language model corresponding to the first user identity from multiple generative large language models as the target model. The multiple generative large language models correspond one-to-one with multiple preset user identities. The generative large language model is used to perform intent recognition on the request information and output feedback information corresponding to the intent recognition result. The target model is pre-trained based on a first training sample set associated with the first user identity. The first training sample set is generated based on context information corresponding to the first user identity in at least one application installed on the first electronic device.

[0032] In this embodiment, the first request information includes at least the request content input by the current user of the first electronic device. The input method includes, but is not limited to, text input, voice input, and visual input. It can be understood that in order for the server to distinguish the identity of the requester corresponding to the received request information, the first electronic device will send the first user identity identifier of the current user of the first electronic device to the server together when sending the first request information.

[0033] The first user identity identifier is a unique identifier that can represent the user's identity. This identifier can be at least one of the user's name, ID card number, email address, mobile phone number, and social media account. This embodiment does not impose any restrictions on this.

[0034] In essence, generative large language models are used to identify the intent of a request and output feedback information corresponding to the intent identification result. In other words, a generative large language model can be viewed as a robot that engages in dialogue with the user, understanding and responding to the user's input. Each of the aforementioned generative large language models corresponds one-to-one with a set of preset user identities; that is, each generative large language model is pre-trained for a specific preset user identity. Specifically, it involves acquiring contextual information about the preset user identity in at least one application on the electronic device they are using, and generating a training sample set based on this contextual information. The large language model is then fine-tuned and supervised training based on this training sample set. In other words, each generative large language model can learn the preferences and behaviors of its corresponding preset user identity, essentially creating a digital twin version of the device owner.

[0035] Based on this, after obtaining the first user identifier carried in the first request information, the generative large language model corresponding to the first user identifier can be determined from multiple generative large language models as the target model. The target model is pre-trained based on a first training sample set associated with the first user identifier. The first training sample set is generated based on contextual information corresponding to the first user identifier in at least one application installed on the first electronic device. In this case, the obtained target model can be understood as a digital twin version of the device owner on the server side, corresponding to the first user identifier.

[0036] The aforementioned applications are at least one of the following: instant messaging applications, shopping applications, game applications, email applications, music applications, navigation applications, and office applications. Contextual information can be understood as information about a user's interactions with other users or devices within any application. For example, the historical chat information between the target user corresponding to the first user identity and the user corresponding to the second user identity in an instant messaging application can serve as the contextual information corresponding to the first identity in the instant messaging application.

[0037] In some implementations, the first training sample set includes multiple training samples, each training sample including a first interactive text and a first response text input by a target user in response to the first interactive text, wherein the target user is the user corresponding to the first user identity identifier. The target model is trained as follows: the first interactive text is input into a pre-trained first initial model to obtain a second response text predicted for the first interactive text; a first loss value is determined based on the degree of difference between the first response text and the second response text; the first initial model is iteratively trained according to the first loss value until the first target condition is met to obtain the target model. At this point, the target model is frozen and set as an inference model. Freezing the target model can be understood as saving the current model network parameters of the target model, and setting the target model as an inference model can be understood as the target model being able to be applied to identify the intent in the request information and output feedback information corresponding to the intent identification result.

[0038] The first objective condition can be: detecting a first loss value less than a preset value, detecting a first loss value that no longer changes, or reaching a preset number of training iterations. Understandably, after iteratively training the first initial model on the first training sample set for multiple training cycles (each training cycle includes multiple iterations), continuously optimizing the parameters of the first initial model, the first loss value decreases until it reaches a fixed value or is less than the preset value. At this point, it indicates that the first initial model has converged. Alternatively, convergence can be determined after reaching a preset number of training iterations; in this case, the converged first initial model can be used as the objective model. The preset value and the preset number of training iterations are pre-set and can be adjusted according to different application scenarios; this embodiment does not impose such restrictions.

[0039] Understandably, the first initial model is a large-scale pre-trained language model, such as an LLM model. Therefore, fine-tuning the pre-trained first initial model using a small number of initial training samples can improve the training speed and also enable the fine-tuned model to more effectively identify the first request information corresponding to the target user. Furthermore, the training process of the target model can be executed on the server mentioned in this application or on other devices; this application makes no limitation on this.

[0040] In some implementations, during the fine-tuning training phase of the target model, after the target model converges, its accuracy can be tested using a test sample set. This involves statistically analyzing the accuracy of the target model's output feedback data for all test samples in the test sample set. The accuracy is then fed back to the first electronic device. If a confirmation instruction based on the accuracy feedback is received from the first electronic device, it indicates that the target model has met the accuracy requirements of the first electronic device, and the model network parameters can be saved. If a continued training instruction based on the accuracy feedback is received from the first electronic device, it indicates that the target model has not met the accuracy requirements of the first electronic device. In this case, the server will continue training the model using the negative samples that the target model failed to predict accurately during previous training, until the accuracy fed back to the first electronic device meets its accuracy requirements. The accuracy requirement of the first electronic device can be a pre-set precision value, such as 90%, or it can be an accuracy value temporarily determined by the user; this embodiment does not impose any restrictions on this. The trained target model can be considered as a personal digital version of the Chat Generative Pre-trained Transformer (ChatGPT).

[0041] Step S230: Input the first request information into the target model to obtain the first feedback information corresponding to the first request information.

[0042] Step S240: Send the first feedback information to the first electronic device.

[0043] Furthermore, after selecting the target model customized for the target user, the first request information sent by the target user through the first electronic device can be input into the target model. The target model will then perform intent recognition on the first request information and output corresponding feedback information based on the intent recognition result, that is, output the first feedback information corresponding to the first request information; then the server will send the first feedback information to the first electronic device.

[0044] In some implementations, the first feedback information may consist solely of feedback data, which may include at least one of text feedback data, image feedback data, audio feedback data, and video feedback data. For example, if the first request information is the text data "What is passion fruit?", the first feedback information output by the target model could be the text data "Passiflora edulis Sims, a vine belonging to the Passifloraceae family and the Passiflora genus. Stems glabrous; leaves papery, glabrous on both sides; stipules linear-lanceolate; flowers fragrant, white, sepals oblong, petals lanceolate, ovary obovoid; fruit ovoid, pericarp hard, flowering from April to June, fruiting from July to April of the following year." As another example, if the first request information is the text data "What does passion fruit look like?", the first feedback information output by the target model could be an image of a passion fruit.

[0045] In other embodiments, the first feedback information may include a launch command for a target function in a target application within the first electronic device, and feedback data for that target function. The feedback data may include at least one of text feedback data, image feedback data, and audio feedback data. Based on this, the feedback data and the launch command are sent to the first electronic device. The launch command instructs the first electronic device to launch the target function in the target application and output the feedback data using the target function. The launch command for the target function in the target application may be a launch command generated by a target model recognizing the intent of the first request information and determining the operation corresponding to that intent. The launch command and the feedback data may be generated simultaneously or sequentially; this embodiment does not impose any limitations on this.

[0046] For example, the target application is an email application, the target function is an email reply function, and the feedback data is the email reply content generated for the received target email. That is, after the server sends the first feedback information to the first target device, the first target device can activate the email reply function in the email application, generate a target reply email based on the feedback data, and send the target reply email to the electronic device corresponding to the target email. Understandably, since the target model has pre-learned the preferences, behaviors, and speaking styles of the target user corresponding to the first user identity, the generated email reply content is also generated by imitating the target user's email reply tone and reply style. Thus, the target model can be seen as the target user's personal digital assistant, effectively and efficiently helping the target user reply to emails; greatly improving the target user's email reply efficiency and work efficiency.

[0047] Similarly, target models can also help target users read emails, reply to text messages, search for information, and generate and manage schedules, thereby greatly improving user work efficiency and saving time and energy.

[0048] Furthermore, the aforementioned first request information can be actively entered by the user; it can also be generated by an application on the first electronic device. For example, after receiving an email, an email application will generate corresponding first request information for the server based on the received email. Correspondingly, the server can respond to the first request information and generate corresponding first feedback information for the first electronic device. In other words, the target model can achieve functions such as automatically replying to emails, replying to text messages, or managing schedules without the user's active triggering. That is, the server-side target model can handle daily chores on behalf of the target user, allowing the user to have multiple instances of themselves, thereby improving their work efficiency and saving time and energy.

[0049] Optionally, the target application can also be a smart home application. In this case, the target function is a control function, and the feedback data is text data such as "I am about to perform a control operation on a certain home device." That is, the server can identify through the target model that the first request is to control other devices, such as a smart TV or a smart access control system. Correspondingly, the server can send feedback data and a start command to the first electronic device. The first electronic device can then use the target application to send the start command to the corresponding other devices to control them to perform start operations related to the start command, such as controlling the smart TV to change channels, increase or decrease volume; or controlling the smart access control system to open the door. Simultaneously, the target application can also display feedback data to notify the target user that a control operation is about to be performed.

[0050] Optionally, the target application can also be a sales management application. Correspondingly, the target functions can be sales process statistics, sales data statistics, and sales analysis. After the server sends feedback data to the first electronic device, it helps the target user achieve enterprise sales management, such as automating sales processes, statistically analyzing sales data, providing sales analysis, automated advertising placement, data processing, and advertising analysis.

[0051] Furthermore, the target model in the server can also help enterprises achieve digital transformation. For example, it can help enterprises implement customer relationship management, such as recording customer information, tracking customer behavior, and providing personalized services. This embodiment will not provide further examples.

[0052] The LLM model described above can achieve data interaction with applications in electronic devices through the plug-in functionality disclosed by ChatGPT.

[0053] In this embodiment, the target model for processing the first request information is pre-trained based on a first training sample set associated with the first user identity, and the first training sample set is generated based on contextual information corresponding to the first user identity in at least one application installed on the first electronic device. It is evident that the target model learns the preferences and speaking habits of the user corresponding to the first user identity, which is equivalent to the target model being trained specifically and customized for the user corresponding to the first user identity. Therefore, using the target model, the first feedback information corresponding to the first request information can be generated more accurately, thus improving the accuracy of the feedback and making the feedback information more personalized. This target model can be seen as a personal digital assistant for the user corresponding to the first user identity. When the server receives the request information corresponding to the first user identity, it can simulate the speaking style and personality of the user corresponding to the first user identity to generate feedback information; thereby, it can replace the user corresponding to the first user identity in handling daily chats, replying to text messages, replying to emails, ordering takeout, and summarizing work, etc., giving the user an extra persona, improving their work efficiency, and saving time and energy.

[0054] Please refer to Figure 3 , Figure 3 This is a flowchart illustrating an information processing method according to another embodiment of this application, applied to a server. The following will be combined with... Figure 3 The information processing method provided in the embodiments of this application will be described in detail. This information processing method may include the following steps:

[0055] Step S310: Receive a first request message sent by a first electronic device, wherein the first request message carries a first user identity identifier of the current user of the first electronic device.

[0056] Step S320: Determine the generative large language model corresponding to the first user identity from multiple generative large language models as the target model. The multiple generative large language models correspond one-to-one with multiple preset user identities. The generative large language model is used to perform intent recognition on the request information and output feedback information corresponding to the intent recognition result. The target model is pre-trained based on a first training sample set associated with the first user identity. The first training sample set is generated based on context information corresponding to the first user identity in at least one application installed on the first electronic device.

[0057] In this embodiment, the specific implementation of steps S310 to S320 can be found in the content of the foregoing embodiments, and will not be repeated here.

[0058] Step S330: If the first request information carries a target image, determine the image recognition model corresponding to the first user identity from multiple pre-trained image recognition models, and use it as the target recognition model.

[0059] Understandably, the first request information may contain a target image, meaning the target user corresponding to the first user identity can input the target image. Since the target model requires text as input, the target image cannot be directly input into the target model for intent recognition; it needs to be input into an image recognition model for image recognition first. Therefore, the image recognition model can be pre-trained to ensure that, when the request information contains a target image, the image recognition model can identify the content contained in the target image.

[0060] The aforementioned image recognition models are pre-trained for multiple preset user identity identifiers, and each image recognition model corresponds one-to-one with a preset user identity identifier; that is, an image recognition model is specifically trained for each preset user identity identifier corresponding to a preset user, and the obtained target recognition model can be understood as an image recognition model specifically trained for the target user corresponding to the first user identity identifier.

[0061] Optionally, the target recognition model is trained in the following way:

[0062] First, an image sample set corresponding to the first user identity is obtained. The image sample set is generated based on the images in the album associated with the first user identity. Each sample image in the image sample set carries first tag information, which is used to characterize the image content contained in the sample image.

[0063] Next, each sample image is input into the pre-trained second initial model to obtain the predicted second label information for each sample image. This second label information characterizes the image content identified by the pre-trained second initial model. The pre-trained second initial model can be a text-image pre-training (CLIP) model, which typically employs contrastive learning during the pre-training phase. Since the image sample set involves the personal privacy of the target user corresponding to the first user's identity, the pre-trained CLIP model is fine-tuned for privacy purposes using the image sample set. This means that devices other than the first electronic device are prohibited from viewing or obtaining the image sample set from the server.

[0064] Finally, based on the degree of difference between the first and second label information, a second loss value is determined. The second initial model is then iteratively trained according to this second loss value until the second objective condition is met, resulting in the target recognition model. After training the target recognition model, it can be multimodally aligned with the fine-tuned target model, meaning the recognition result output by the target recognition model is used as the input to the target model.

[0065] The second objective condition can be: detecting a second loss value less than a preset value, detecting a second loss value that no longer changes, or reaching a preset number of training iterations. Understandably, after iteratively training the second initial model on the image sample set for multiple training cycles (each training cycle includes multiple iterations), the parameters in the second initial model are continuously optimized, causing the detected second loss value to decrease until it reaches a fixed value or is less than the preset value. At this point, it indicates that the second initial model has converged. Alternatively, convergence can be determined after reaching a preset number of training iterations; in this case, the converged second initial model can be used as the objective model. The preset value and the preset number of training iterations are pre-set and can be adjusted according to different application scenarios; this embodiment does not impose any restrictions on this.

[0066] Understandably, the second initial model is a large-scale pre-trained image recognition model, such as the CLIP model. Therefore, fine-tuning the pre-trained first initial model using a small set of image samples not only improves the training speed but also allows the fine-tuned second initial model to more effectively identify the image content contained in the target image input by the user. Furthermore, the training process of the target recognition model can be executed on the server mentioned in this application or on other devices; this application makes no limitation on this.

[0067] Step S340: Input the target image into the target recognition model to obtain a first recognition result corresponding to the target image. The first recognition result is text-type information and is used to characterize the content contained in the target image.

[0068] Furthermore, the target image can be input into the target recognition model for image recognition, identifying the content contained in the target image as the first recognition result. This first recognition result is text-based information. For example, if the target image contains an apple, inputting the target image into the target recognition model might output the text "A photo of the apple." Moreover, since the image sample set used to train the target recognition model is generated from images in the album associated with the first user's identity, and the album typically contains screenshots of the target user's daily life, photos taken by the user, and favorite images saved from the internet, the finely tuned target recognition model can more accurately identify objects or scenes commonly encountered by the target user. This improves the accuracy of the target recognition model in recognizing target images input by the target user, thus improving the accuracy of the first recognition result and consequently, the accuracy of the first feedback information output by the target model based on the first recognition result.

[0069] Step S350: Input the first recognition result into the target model to obtain the first feedback information corresponding to the first request information.

[0070] In other implementations, if the first request information carries target audio, the server can use Automatic Speech Recognition (ASR) technology to convert the target audio into corresponding target text; then input the target text into the target model to obtain first feedback information corresponding to the first request information.

[0071] Step S360: Send the first feedback information to the first electronic device.

[0072] In this embodiment, the specific implementation of steps S350 to S360 can be found in the content of the foregoing embodiments, and will not be repeated here.

[0073] In this embodiment, an image sample set is generated from the images in the album associated with the first user identity, and an image recognition model corresponding to the first user identity is trained based on this image sample set. Since the album generally contains daily screenshots of the target user corresponding to the first user identity, images taken by the user, and images saved from the internet that the user likes, the finely tuned target recognition model can more accurately identify common objects or scene images of the target user. This improves the accuracy of the target recognition model in recognizing target images input by the target user, thus improving the accuracy of the first recognition result. In turn, it can improve the accuracy of the first feedback information output by the target model based on the first recognition result.

[0074] Please refer to Figure 4 , Figure 4 This is a flowchart illustrating an information processing method according to another embodiment of this application, applied to a first electronic device. The following will be combined with... Figure 4 The information processing method provided in the embodiments of this application will be described in detail. This information processing method may include the following steps:

[0075] Step S410: Send a first request message to the server, the first request message carrying the first user identity identifier of the current user of the first electronic device.

[0076] In some implementations, the first request information is generated based on user-inputted operation information. This input operation information can be text, image, or audio information, for example, the first request information could be the text or audio message "Please order shredded pork with garlic sauce from a Chinese restaurant near my company."

[0077] Step S420: Receive first feedback information sent by the server based on the first request information. The first feedback information is a generative large language model determined by the server from multiple generative large language models that corresponds to the first user identity, serving as the target model. Input the first request information into the target model to obtain the first feedback information corresponding to the first request information. The multiple generative large language models correspond one-to-one with multiple preset user identities. The generative large language models are used to perform intent recognition on the request information and output feedback information corresponding to the intent recognition result. The target model is pre-trained based on a first training sample set associated with the first user identity. The first training sample set is generated based on context information corresponding to the first user identity in at least one application installed on the first electronic device.

[0078] The specific implementation method for how the first feedback information is generated by the target model in the server can be found in the foregoing embodiments and will not be repeated here. Based on this, after the server generates the first feedback information, the first electronic device can receive it.

[0079] Step S430: Output the first feedback information.

[0080] The first feedback information can be output in the form of displaying it on the current display interface of the first electronic device or through audio output. This embodiment does not limit this method.

[0081] In some implementations, after step S430, context information corresponding to the first user identifier in at least one target application within the target time period can be obtained at regular intervals. This context information is used as the target application, which is an application whose information sharing function is enabled on the first electronic device. Each application in the first electronic device can have a status switch control corresponding to the information sharing function. Based on this, the target user can turn the information sharing function of each application on or off using the status switch control. An application with its information sharing function enabled is one where the target user can extract its context information by default. Therefore, context information corresponding to the first user identifier in at least one target application can be obtained at regular intervals.

[0082] Furthermore, the target context information is sent to the server so that the server updates the target model based on the target context information. In other words, the first electronic device periodically uploads the target context information to update the target model, thereby improving the accuracy of the target model's intent recognition of the target user's request information and the accuracy of the feedback information; that is, improving the prediction performance and accuracy of the target model through new target context information.

[0083] In addition, when the first electronic device first sends the context information corresponding to the first user identity in at least one target application, it also sends the user's identity and attributes, navigation habits, and information such as home and company addresses to the server. Correspondingly, the server can associate this information with the first user identity so that the target model can determine the static information attributes of the user corresponding to the first user identity based on this information.

[0084] In some implementations, after step S430, the user may input second request information based on the first feedback information. For example, the first request information could be "Please order shredded pork with garlic sauce from a Chinese restaurant near my company," and the first feedback information could be "Can I order shredded pork with garlic sauce from restaurant xxx?" In this case, the aforementioned second request information would be "Yes." Based on this, the server can also input the second request information into the target model, and the corresponding target model will output second feedback data. The second feedback data can include a launch command for a target function in the target application of the first electronic device, and feedback data for the target function in the target application; for example, the target function could be a takeout ordering function, and the feedback data could be the text message "Ordered." In this way, the target model can replace the user in completing multiple complex tasks. Of course, it can not only help users order takeout, but also help users make phone calls, reply to WeChat messages, navigate, and perform other interactive tasks. This embodiment does not limit this.

[0085] In this embodiment, the target model for processing the first request information sent by the first electronic device is pre-trained based on a first training sample set associated with the first user identity, and the first training sample set is generated based on context information corresponding to the first user identity in at least one application installed on the first electronic device. It is evident that the target model learns the preferences and speaking habits of the user corresponding to the first user identity, which is equivalent to the target model being trained specifically and customized for the user corresponding to the first user identity. Therefore, using the target model, the first feedback information corresponding to the first request information can be generated more accurately, thus improving the accuracy of the feedback and making the feedback information more personalized. This target model can be seen as a personal digital assistant for the user corresponding to the first user identity. When the server receives the request information corresponding to the first user identity, it can simulate the speaking style and personality of the user corresponding to the first user identity to generate feedback information; thereby, it can replace the user corresponding to the first user identity in handling daily chats, replying to text messages, replying to emails, ordering takeout, and summarizing work, etc., giving the user an extra persona, improving their work efficiency, and saving time and energy.

[0086] Please refer to Figure 5 , Figure 5 This is a flowchart illustrating an information processing method provided in another embodiment of this application, applied to a first electronic device. The following will be combined with... Figure 5 The information processing method provided in the embodiments of this application will be described in detail. This information processing method may include the following steps:

[0087] Step S510: If the interaction information sent by the second electronic device is received through the currently running application, and the proxy feedback function of the currently running application is detected to be enabled, then the first request information is generated based on the interaction information.

[0088] In this embodiment, the first request information may not be generated based on the information input by the target user corresponding to the first user identity. Instead, it may be generated based on the interactive information sent by the second electronic device received by the application currently running in the first electronic device. For example, the application may be an email application or an SMS application. That is, when the email application or SMS application receives a target email or target SMS sent by the second electronic device to the first electronic device, it may generate the first request information based on the target email or target SMS.

[0089] Understandably, the need for automatic replies to SMS or emails is generally determined by user needs. Therefore, proxy function controls with proxy feedback functionality can be pre-set for each application. Target users can enable or disable the proxy feedback function based on these controls. Only when the proxy feedback function is enabled will the first request information be generated based on the aforementioned interaction information. In other words, when the proxy feedback function is enabled, it's equivalent to the target user granting the target model automatic processing permissions for the application, enabling personal digital management of the application, with the target model acting as an agent for the target user to process the first request information generated by the application.

[0090] In other implementations, if an interactive message is received from a second electronic device via a currently running application, and the proxy feedback function of the currently running application is detected to be enabled, and the interactive message does not contain a target field, then the first request message is generated based on the interactive message. The target field can be pre-set; for example, it could be a field that clearly contains advertising information, or a field commonly used for spam emails or text messages. That is, the first request message is not generated for every interactive message received from the second electronic device. Instead, some interactive messages that do not require a response are filtered out, and only necessary information is generated to allow the target model to perform proxy processing. This, to a certain extent, ensures the security of the target user's personal information and assets, while also reducing the computational resource consumption of the server in processing unimportant request information.

[0091] Step S520: Send a first request message to the server, the first request message carrying the first user identity identifier of the current user of the first electronic device.

[0092] Step S530: Receive first feedback information sent by the server based on the first request information. The first feedback information is a generative large language model determined by the server from multiple generative large language models that corresponds to the first user identity identifier, and used as the target model. Input the first request information into the target model to obtain the first feedback information corresponding to the first request information. The multiple generative large language models correspond one-to-one with multiple preset user identity identifiers. The generative large language models are used to perform intent recognition on the request information and output feedback information corresponding to the intent recognition result. The target model is pre-trained based on a first training sample set associated with the first user identity identifier. The first training sample set is generated based on context information corresponding to the first user identity identifier in at least one application installed on the first electronic device.

[0093] Step S540: Output the first feedback information.

[0094] In this embodiment, the specific implementation of steps S520 to S540 can be found in the content of the foregoing embodiments, and will not be repeated here.

[0095] In other implementations, if interactive information from a second electronic device is received through a currently running application, and the proxy feedback function of the currently running application is detected to be disabled, a first prompt message is generated based on the interactive information and displayed. This first prompt message is used by the user to view the interactive information. Understandably, when the proxy feedback function is disabled, it indicates that the target user does not wish to have the target model deployed on the server handle the process automatically. In this case, a first prompt message can be generated and displayed to prompt the user to view the interactive information sent by the second electronic device. For example, the first prompt message can be displayed as a pop-up window to indicate that a target email from a second electronic device has been received in the email application.

[0096] In this embodiment, the first electronic device will only automatically process applications with the proxy feedback function enabled, realizing personal digital management of the application. The target model acts as an agent for the target user to process the first request information generated by the application. At this time, for applications with the proxy feedback function enabled, the target model is essentially a digital twin of the user corresponding to the first user identity, and can replace the user corresponding to the first user identity to complete the interaction tasks of this part of the application.

[0097] Please refer to Figure 6The diagram illustrates a structural block diagram of an information processing device 600 according to an embodiment of this application, applied to a server. The device 600 may include: an information receiving module 610, a model determining module 620, a feedback information acquisition module 630, and an information feedback module 640.

[0098] The information receiving module 610 is used to receive a first request information sent by the first electronic device, wherein the first request information carries the first user identity identifier of the current user of the first electronic device.

[0099] The model determination module 620 is used to determine the generative large language model corresponding to the first user identity from multiple generative large language models as the target model. The multiple generative large language models correspond one-to-one with multiple preset user identity. The generative large language model is used to perform intent recognition on the request information and output feedback information corresponding to the intent recognition result. The target model is pre-trained based on a first training sample set associated with the first user identity. The first training sample set is generated based on context information corresponding to the first user identity in at least one application installed on the first electronic device.

[0100] The feedback information acquisition module 630 is used to input the first request information into the target model to obtain the first feedback information corresponding to the first request information.

[0101] The information feedback module 640 is used to send the first feedback information to the first electronic device.

[0102] In some embodiments, the first training sample set includes multiple training samples, each training sample including a first interactive text and a first reply text input by a target user in response to the first interactive text, wherein the target user is the user corresponding to the first user identity identifier. The information processing device 600 may further include a first model training module. The model training module may be specifically used to: input the first interactive text into a pre-trained first initial model to obtain a second reply text predicted for the first interactive text; determine a first loss value based on the degree of difference between the first reply text and the second reply text; and iteratively train the first initial model according to the first loss value until a first target condition is met to obtain the target model.

[0103] In some embodiments, the information processing device 600 may further include a recognition model determination module and an image recognition module. The recognition model determination module may be used to determine, before inputting the first request information into the target model to obtain first feedback information corresponding to the first request information, if the first request information carries a target image, an image recognition model corresponding to the first user identity identifier from a plurality of pre-trained image recognition models, and use this model as the target recognition model. The image recognition module may be used to input the target image into the target recognition model to obtain a first recognition result corresponding to the target image, wherein the first recognition result is text-type information and is used to characterize the content contained in the target image.

[0104] In this manner, the feedback information acquisition module 630 can be specifically used to input the first recognition result into the target model to obtain the first feedback information corresponding to the first request information.

[0105] In this manner, the information processing device 600 may further include a second model training module. The second model training module may be specifically used for: acquiring an image sample set corresponding to the first user identity, wherein the image sample set is generated based on images in an album associated with the first user identity, and each sample image in the image sample set carries first label information, the first label information being used to characterize the image content contained in the sample image; inputting each sample image into a pre-trained second initial model to obtain predicted second label information for each sample image, the second label information being used to characterize the image content contained in the sample image identified by the pre-trained second initial model; determining a second loss value based on the degree of difference between the first label information and the second label information; and iteratively training the second initial model according to the second loss value until a second target condition is met, thereby obtaining the target recognition model.

[0106] In some embodiments, the first feedback information includes a launch command for a target function in a target application within the first electronic device, and feedback data for the target function in the target application. The feedback data includes at least one of text feedback data, image feedback data, and audio feedback data. The information feedback module 640 can be used to send the feedback data and the launch command to the first electronic device. The launch command instructs the first electronic device to launch the target function in the target application and output the feedback data using the target function.

[0107] Please refer to Figure 7The diagram illustrates a structural block diagram of an information processing device 700 according to another embodiment of this application, applied to a first electronic device. The device 700 may include: an information sending module 710, an information receiving module 720, and an information output module 730.

[0108] The information sending module 710 is used to send a first request information to the server, the first request information carrying the first user identity identifier of the current user of the first electronic device.

[0109] The information receiving module 720 is used to receive first feedback information sent by the server based on the first request information. The first feedback information is a generative large language model that the server determines from multiple generative large language models that corresponds to the first user identity as a target model. The first request information is input into the target model to obtain the first feedback information corresponding to the first request information. The multiple generative large language models correspond one-to-one with multiple preset user identities. The generative large language models are used to perform intent recognition on the request information and output feedback information corresponding to the intent recognition result. The target model is pre-trained based on a first training sample set associated with the first user identity. The first training sample set is generated based on context information corresponding to the first user identity in at least one application installed on the first electronic device.

[0110] The information output module 730 is used to output the first feedback information.

[0111] In some embodiments, the information processing device 700 may further include a request information generation module and a prompt information generation module. The request information generation module may, before sending the first request information to the server, generate the first request information based on the interaction information received through the currently running application from the second electronic device, and detect that the proxy feedback function of the currently running application is enabled. The prompt information generation module may, if the currently running application receives the interaction information received from the second electronic device, and detects that the proxy feedback function of the currently running application is disabled, generate and display the first prompt information based on the interaction information, wherein the first prompt information is used by the user to view the interaction information.

[0112] In this approach, the request information generation module can also be specifically used to: if the currently running application receives interactive information sent by the second electronic device, detects that the proxy feedback function of the currently running application is enabled, and the interactive information does not contain a target field, then generate the first request information based on the interactive information.

[0113] In other embodiments, the information processing apparatus 700 may further include an operation receiving module and a request information generation module. The operation receiving module may be used to receive operation information input by the user before sending the first request information to the server. The request information generation module may be used to generate the first request information based on the operation information.

[0114] In some embodiments, the information processing device 700 may further include a context information acquisition module and a context information sending module. The context information acquisition module may be used to acquire, at target intervals, context information corresponding to the first user identity in at least one target application within the target interval after the first feedback information is output, as target context information. The target application is an application whose information sharing function is enabled on the first electronic device. The context information sending module may be used to send the target context information to the server, so that the server updates the target model based on the target context information.

[0115] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described device and module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0116] In the several embodiments provided in this application, the coupling between modules can be electrical, mechanical, or other forms of coupling.

[0117] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0118] In summary, the server receives a first request message from a first electronic device, wherein the first request message carries a first user identity identifier of the current user of the first electronic device. It then determines a generative large language model corresponding to the first user identity identifier from multiple generative large language models as the target model. These multiple generative large language models correspond one-to-one with various preset user identities. The generative large language models are used to perform intent recognition on the request message and output feedback information corresponding to the intent recognition result. The first request message is input into the target model to obtain first feedback information corresponding to the first request message. Finally, the first feedback information is sent to the first electronic device. Since the target model is pre-trained based on a first training sample set associated with the first user identity identifier, and the first training sample set is generated based on contextual information corresponding to the first user identity identifier in at least one application installed on the first electronic device, the target model can learn the preferences and speaking habits of the user corresponding to the first user identity identifier. This is equivalent to the target model being trained specifically and customized for the user corresponding to the first user identity identifier. Therefore, using the target model, the first feedback information corresponding to the first request message can be generated more accurately, thus improving the accuracy of the feedback and making the feedback information more personalized.

[0119] The following will combine Figure 8 This application describes a computer device.

[0120] Reference Figure 8 , Figure 8This diagram illustrates a structural block diagram of a computer device 800 according to an embodiment of this application. The method described above in this embodiment can be executed by this computer device 800. The computer device can be an electronic terminal with data processing capabilities, including but not limited to smartphones, tablets, laptops, desktop computers, smartwatches, e-book readers, MP3 (Moving Picture Experts Group Audio Layer III) players, MP4 (Moving Picture Experts Group Audio Layer IV) players, and smart home devices. Alternatively, the computer device can be a server, which can be an independent physical server, a server cluster composed of multiple physical servers, or a distributed system. It can also be a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, Content Delivery Network (CDN), and big data and artificial intelligence platforms.

[0121] The computer device 800 in this application embodiment may include one or more of the following components: processor 801, memory 802, and one or more application programs, wherein the one or more application programs may be stored in memory 802 and configured to be executed by one or more processors 801, and the one or more programs are configured to perform the methods as described in the foregoing method embodiments.

[0122] Processor 801 may include one or more processing cores. Processor 801 connects to various parts within the computer device 800 using various interfaces and lines, and performs various functions and processes data of the computer device 800 by running or executing instructions, programs, code sets, or instruction sets stored in memory 802, and by calling data stored in memory 802. Optionally, processor 801 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). Processor 801 may integrate one or more of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the displayed content; and the modem handles wireless communication. It is understood that the aforementioned modem can also be integrated into processor 801 and implemented using a separate communication chip.

[0123] The memory 802 may include random access memory (RAM) or read-only memory (ROM). The memory 802 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 802 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (such as touch functionality, sound playback functionality, image playback functionality, etc.), and instructions for implementing the various method embodiments described below. The data storage area may also store data created by the computer device 800 during use (such as the various correspondences described above).

[0124] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described device and module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0125] In the several embodiments provided in this application, the coupling or direct coupling or communication connection between the modules shown or discussed may be an indirect coupling or communication connection through some interface, device or module, and may be electrical, mechanical or other forms.

[0126] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0127] Please refer to Figure 9 This diagram illustrates a structural block diagram of a computer-readable storage medium provided in an embodiment of this application. The computer-readable medium 900 stores program code that can be called by a processor to execute the methods described in the above method embodiments.

[0128] The computer-readable storage medium 900 may be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. Optionally, the computer-readable storage medium 900 includes a non-transitory computer-readable storage medium. The computer-readable storage medium 900 has storage space for program code 910 that performs any of the method steps described above. This program code can be read from or written to one or more computer program products. The program code 910 may be compressed, for example, in a suitable form.

[0129] In some embodiments, a computer program product or computer program is provided, which includes computer instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the electronic device to perform the steps in the above-described method embodiments.

[0130] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. An information processing method, characterized in that, Applied to a server, the method includes: Receive a first request message sent by a first electronic device, wherein the first request message carries a first user identity identifier of the current user of the first electronic device; The generative large language model corresponding to the first user identity is determined from multiple generative large language models and used as the target model. The multiple generative large language models correspond one-to-one with multiple preset user identities. The generative large language model is used to perform intent recognition on the request information and output feedback information corresponding to the intent recognition result. The target model is pre-trained based on a first training sample set associated with the first user identity. The first training sample set is generated based on context information corresponding to the first user identity in at least one application installed on the first electronic device. The first request information is input into the target model to obtain the first feedback information corresponding to the first request information; Send the first feedback information to the first electronic device.

2. The method according to claim 1, characterized in that, The first training sample set includes multiple training samples, each training sample including a first interactive text and a first reply text input by a target user in response to the first interactive text, wherein the target user is the user corresponding to the first user identity identifier; The target model is trained in the following way: The first interactive text is input into the pre-trained first initial model to obtain the second response text predicted for the first interactive text; A first loss value is determined based on the degree of difference between the first response text and the second response text; Based on the first loss value, the first initial model is iteratively trained until the first target condition is met, thereby obtaining the target model.

3. The method according to claim 1, characterized in that, Before inputting the first request information into the target model to obtain the first feedback information corresponding to the first request information, the method further includes: If the first request information carries a target image, the image recognition model corresponding to the first user identity is determined from multiple pre-trained image recognition models and used as the target recognition model; The target image is input into the target recognition model to obtain a first recognition result corresponding to the target image. The first recognition result is text-type information and is used to characterize the content contained in the target image. The step of inputting the first request information into the target model to obtain the first feedback information corresponding to the first request information includes: The first identification result is input into the target model to obtain the first feedback information corresponding to the first request information.

4. The method according to claim 3, characterized in that, The target recognition model is trained in the following way: Obtain an image sample set corresponding to the first user identity identifier. The image sample set is generated based on images in the album associated with the first user identity identifier. Each sample image in the image sample set carries first tag information, which is used to characterize the image content contained in the sample image. Each sample image is input into the pre-trained second initial model to obtain the predicted second label information for each sample image. The second label information is used to characterize the image content contained in the sample image identified by the pre-trained second initial model. A second loss value is determined based on the degree of difference between the first tag information and the second tag information; Based on the second loss value, the second initial model is iteratively trained until the second target condition is met, thus obtaining the target recognition model.

5. The method according to any one of claims 1-4, characterized in that, The first feedback information includes a launch command for a target function in a target application within the first electronic device, and feedback data for the target function in the target application. The feedback data includes at least one of text feedback data, image feedback data, and audio feedback data. Sending the first feedback information to the first electronic device includes: The feedback data and the start command are sent to the first electronic device. The start command is used to instruct the first electronic device to start the target function in the target application and output the feedback data using the target function.

6. An information processing method, characterized in that, Applied to a first electronic device, the method includes: Send a first request message to the server, the first request message carrying the first user identity identifier of the current user of the first electronic device; The system receives first feedback information sent by the server based on the first request information. The first feedback information is a generative large language model that the server determines from multiple generative large language models to correspond to the first user identity as a target model. The first request information is then input into the target model to obtain the first feedback information corresponding to the first request information. The multiple generative large language models correspond one-to-one with multiple preset user identities. The generative large language models are used to perform intent recognition on the request information and output feedback information corresponding to the intent recognition result. The target model is pre-trained based on a first training sample set associated with the first user identity. The first training sample set is generated based on context information corresponding to the first user identity in at least one application installed on the first electronic device. Output the first feedback information.

7. The method according to claim 6, characterized in that, Before sending the first request information to the server, the method further includes: If the interaction information sent by the second electronic device is received through the currently running application, and the proxy feedback function of the currently running application is detected to be enabled, then the first request information is generated based on the interaction information. If an interactive message is received from a second electronic device through a currently running application, and the proxy feedback function of the currently running application is detected to be turned off, a first prompt message is generated based on the interactive message and displayed. The first prompt message is used by the user to view the interactive message.

8. The method according to claim 7, characterized in that, If the interaction information sent by the second electronic device is received through the currently running application, and the proxy feedback function of the currently running application is detected to be enabled, then the first request information is generated based on the interaction information, including: If the application currently running receives interactive information from the second electronic device, detects that the proxy feedback function of the currently running application is enabled, and the interactive information does not contain a target field, then the first request information is generated based on the interactive information.

9. The method according to claim 6, characterized in that, Before sending the first request information to the server, the method further includes: Receive user input for operation information; Based on the operation information, the first request information is generated.

10. The method according to any one of claims 6-9, characterized in that, After outputting the first feedback information, the method further includes: Every target duration, obtain context information corresponding to the first user identity in at least one target application within the target duration, as target context information, wherein the target application is an application whose information sharing function is enabled on the first electronic device; The target context information is sent to the server so that the server updates the target model based on the target context information.

11. An information processing device, characterized in that, Applied to a server, the device includes: The information receiving module is used to receive a first request information sent by a first electronic device, wherein the first request information carries a first user identity identifier of the current user of the first electronic device; The model determination module is used to determine the generative large language model corresponding to the first user identity from multiple generative large language models as the target model. The multiple generative large language models correspond one-to-one with multiple preset user identity. The generative large language model is used to perform intent recognition on the request information and output feedback information corresponding to the intent recognition result. The target model is pre-trained based on a first training sample set associated with the first user identity. The first training sample set is generated based on context information corresponding to the first user identity in at least one application installed on the first electronic device. The feedback information acquisition module is used to input the first request information into the target model to obtain the first feedback information corresponding to the first request information; The information feedback module is used to send the first feedback information to the first electronic device.

12. An information processing device, characterized in that, Applied to a first electronic device, the device includes: The information sending module is used to send a first request information to the server, wherein the first request information carries the first user identity identifier of the current user of the first electronic device; The information receiving module is used to receive first feedback information sent by the server based on the first request information. The first feedback information is a generative large language model that the server determines from multiple generative large language models that corresponds to the first user identity as a target model. The first request information is input into the target model to obtain the first feedback information corresponding to the first request information. The multiple generative large language models correspond one-to-one with multiple preset user identities. The generative large language models are used to perform intent recognition on the request information and output feedback information corresponding to the intent recognition result. The target model is pre-trained based on a first training sample set associated with the first user identity. The first training sample set is generated based on context information corresponding to the first user identity in at least one application installed on the first electronic device. The information output module is used to output the first feedback information.

13. A computer device, characterized in that, include: One or more processors; Memory; One or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs being configured to perform the method as described in any one of claims 1 to 10.

14. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores program code that can be invoked by a processor to execute the method as described in any one of claims 1 to 10.

Images