Text sharing feedback method, device, equipment and product based on large language model

Abstract

The invention relates to the technical field of artificial intelligence, and provides a text sharing feedback method and device based on a large language model, electronic equipment and a computer program product. The method comprises the following steps: when a knowledge sharing text published by a user is received, inputting a first preset cue word into a large language model to indicate the large language model to execute quality evaluation on the knowledge sharing text and generate an incentive reply content; and if the result of the quality evaluation accords with a first preset condition, feeding back the motivation reply content to the user. According to the method, the large language model is introduced to automatically evaluate the quality of the knowledge sharing text and generate the motivation reply content, so that the low-quality knowledge sharing text can be prevented from being motivated, and manual statistics of related data is not needed, so that the time consumed by data statistics is shortened, and the error rate is reduced.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a text sharing feedback method, device, electronic device, and computer program product based on a large language model. Background Art

[0002] For corporate knowledge sharing, employees can share information such as reading notes and technical experience via group chats or email. To motivate employees, companies often offer incentives or financial rewards for sharing. However, manually compiling the data related to employee sharing often requires time-consuming and error-prone processing. Furthermore, some employees may post low-quality knowledge sharing content perfunctorily in exchange for incentives. Summary of the Invention

[0003] In view of this, the embodiments of the present application provide a text sharing feedback method, device, electronic device and computer program product based on a large language model, which can reduce the time and error rate of data statistics and avoid low-quality knowledge sharing texts from receiving incentives.

[0004] A first aspect of an embodiment of the present application provides a text sharing feedback method based on a large language model, comprising: When receiving a knowledge sharing text posted by a user, inputting a first preset prompt word into the large language model to instruct the large language model to perform quality assessment on the knowledge sharing text and generate incentive reply content; If the result of the quality assessment meets the first preset condition, the incentive reply content will be fed back to the user.

[0005] The technical solution of the embodiment of the present application inputs a first preset prompt word into the large language model upon receiving a knowledge sharing text posted by a user. After receiving the first preset prompt word, the large language model performs a quality assessment on the knowledge sharing text and generates incentive-based response content. The incentive-based response content is fed back to the user only if the result of the quality assessment meets the preset conditions. The above process, by introducing a large language model to automatically perform quality assessment on knowledge sharing texts and generate incentive-based response content, can prevent low-quality knowledge sharing texts from receiving incentives. It also eliminates the need for manual statistics of relevant data, thereby reducing the time and error rate of data statistics.

[0006] In one implementation of the embodiment of the present application, when a knowledge sharing text posted by a user is received, a first preset prompt word is input into the large language model, including: When receiving a knowledge sharing text posted by a user in a group chat interface of a social application, inputting a first preset prompt word into the large language model; If the quality assessment result meets the first preset condition, the incentive response content will be fed back to the user, including: If the result of the quality assessment meets the first preset condition, the incentive reply content will be sent to the group chat interface.

[0007] In one implementation of the embodiment of the present application, when a knowledge sharing text posted by a user in a group chat interface of a social application is received, a first preset prompt word is input into the large language model, including: When receiving a knowledge sharing text posted by a user in the group chat interface of a social application, if the knowledge analysis text contains a command symbol that triggers the robot assistant, the knowledge sharing text is written into a multidimensional table through the robot assistant; When it is detected that the knowledge sharing text is written into the multidimensional table, a first preset prompt word is input into the large language model.

[0008] In one implementation of the embodiment of the present application, the multidimensional table includes a user identification field, a sharing text field, and a sharing time field; writing the knowledge sharing text into the multidimensional table includes: Map the user's group member name in the group chat interface to the user ID and write it into the user ID field; Map the body content of the knowledge sharing text to the sharing body and write it into the sharing body field; Map the timestamp of publishing the knowledge sharing text to the sharing time and write it into the sharing time field.

[0009] In one implementation of an embodiment of the present application, the multidimensional table includes an AI scoring field and an AI reply field, the quality assessment result includes a scoring result written into the AI ​​scoring field, and the incentive reply content includes an incentive reply text and a highlight extraction text written into the AI ​​reply field; if the quality assessment result meets a first preset condition, the incentive reply content is sent to the group chat interface, including: If the score result in the AI ​​score field is higher than the preset threshold, the incentive reply text and highlight extraction text in the AI ​​reply field will be sent to the group chat interface.

[0010] In one implementation of the embodiment of the present application, after inputting a first preset prompt word into the large language model to instruct the large language model to perform quality assessment on the knowledge sharing text and generate incentive reply content, the method further includes: If the result of the quality assessment meets the second preset condition, a second preset prompt word is input into the large language model to instruct the large language model to output the result of whether the knowledge sharing text should be rewarded according to the set reward acquisition rules.

[0011] In one implementation of the embodiment of the present application, after inputting a first preset prompt word into the large language model to instruct the large language model to perform quality assessment on the knowledge sharing text and generate incentive reply content, the method further includes: If the result of the quality assessment does not meet the second precondition, it is determined that the knowledge sharing text should not be rewarded.

[0012] A second aspect of the embodiments of the present application provides a text sharing and feedback device based on a large language model, comprising: The large language model analysis module is configured to input a first preset prompt word into the large language model upon receiving a knowledge sharing text posted by a user, so as to instruct the large language model to perform a quality assessment on the knowledge sharing text and generate motivational reply content; The content feedback module is used to feed back incentive reply content to the user if the result of the quality assessment meets the first preset condition.

[0013] A third aspect of an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the text sharing feedback method based on a large language model as provided in the first aspect of the embodiment of the present application is implemented.

[0014] A fourth aspect of the embodiments of the present application provides a computer program product. When the computer program product is run on an electronic device, the electronic device executes the text sharing feedback method based on a large language model as provided in the first aspect of the embodiments of the present application.

[0015] A fifth aspect of an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the text sharing feedback method based on a large language model as provided in the first aspect of the embodiment of the present application.

[0016] It can be understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments or descriptions of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0018] Figure 1This is a flowchart of a text sharing feedback method based on a large language model provided in an embodiment of the present application; Figure 2 This is a schematic diagram of publishing a knowledge sharing text and receiving incentive reply content in a group chat interface provided by an embodiment of the present application; Figure 3 This is a structural diagram of a text sharing and feedback device based on a large language model provided in an embodiment of the present application; Figure 4 This is a schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] In the following description, for the purpose of illustration rather than limitation, specific details such as specific system structures and technologies are provided to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obstructing the description of the present application with unnecessary details. In addition, in the description of the present application specification and the appended claims, the terms "first," "second," "third," etc. are only used to distinguish descriptions and are not to be understood as indicating or implying relative importance.

[0020] To enhance employee skills and enrich their leisure time, companies periodically organize knowledge-sharing activities, encouraging employees to share knowledge-sharing texts such as reading notes and technical experience via group chats or email. To encourage this initiative, companies offer incentive replies or financial rewards to employees who share. However, this requires manual statistics on each employee's shared texts, which is time-consuming and error-prone. Furthermore, some employees may post low-quality knowledge-sharing texts by casually copying and pasting, hoping to gain undue incentives.

[0021] To address the above issues, the embodiments of the present application propose a text sharing feedback method, device, electronic device, and computer program product based on a large language model. By introducing a large language model to automatically evaluate the quality of knowledge sharing texts and generate incentive response content, it is possible to prevent low-quality knowledge sharing texts from receiving incentives, and there is no need to manually count relevant data, thereby reducing the time and error rate of data statistics. For more specific technical implementation details of the embodiments of the present application, please refer to the various method embodiments described below.

[0022] It should be understood that the execution subjects of the various method embodiments proposed in the present application can be various types of electronic devices, such as mobile phones, tablet computers, desktop computers, wearable devices, augmented reality (AR) / virtual reality (VR) devices, laptop computers, ultra-mobile personal computers (UMPCs), netbooks, personal digital assistants (PDAs), large-screen TVs, etc. The embodiments of the present application do not impose any restrictions on the specific types of the electronic devices.

[0023] See also Figure 1 , shows a text sharing feedback method based on a large language model provided by an embodiment of the present application, including: 101. When receiving a knowledge sharing text posted by a user, input a first preset prompt word into the large language model to instruct the large language model to perform quality assessment on the knowledge sharing text and generate motivational reply content; First, the user publishes a knowledge sharing text, which can be various types of texts such as reading experience, technical experience, communication skills or knowledge popularization. The user can publish the knowledge sharing text through various means such as social applications, emails or text messages, and the embodiments of this application do not limit this.

[0024] When the electronic device detects the knowledge sharing text posted by the user, it will automatically obtain the first preset prompt word of the fixed design and input it into a certain AI large language model (LLM) model. The embodiment of the present application does not impose any restrictions on the specific type of large language model used. The first preset prompt word is used to inform the large language model to perform AI analysis on the knowledge sharing text, perform quality assessment on the knowledge sharing text, and generate corresponding incentive reply content, etc. After receiving the first preset prompt word, the large language model will parse and extract semantics of the knowledge sharing text, and perform quality assessment on the knowledge sharing text according to the built-in text quality assessment rules, obtain the corresponding AI quality score results, and automatically generate incentive reply content according to the built-in incentive text generation mechanism. Generally speaking, the large language model can complete the quality assessment based on multiple factors such as the authenticity, independent thinking, sincerity, literary level, whether the original text is excerpted, and whether it is AI generated.

[0025] In one implementation of the embodiment of the present application, when a knowledge sharing text posted by a user is received, a first preset prompt word is input into the large language model, including: When a knowledge sharing text posted by a user in a group chat interface of a social application is received, a first preset prompt word is input into the large language model.

[0026] A user can post a knowledge-sharing text in a group chat interface of a social application, allowing other group members to access the text through the group chat interface, completing the text sharing process. Upon detecting the knowledge-sharing text posted in the group chat interface, the electronic device inputs the first preset prompt word into the large language model, automatically triggering the AI ​​analysis function.

[0027] In one implementation of the embodiment of the present application, when a knowledge sharing text posted by a user in a group chat interface of a social application is received, a first preset prompt word is input into the large language model, including: (1) When receiving a knowledge sharing text posted by a user in the group chat interface of a social application, if the knowledge analysis text contains a command symbol that triggers the robot assistant, the knowledge sharing text is written into the multidimensional table through the robot assistant; (2) When it is detected that the knowledge sharing text is written into the multidimensional table, a first preset prompt word is input into the large language model.

[0028] In addition to posting knowledge sharing texts in the group chat interface of a social application, users may also post daily communication information with group members. This part of the communication information with group members does not require AI analysis. In order to make a distinction, the user can add a command symbol for triggering the robot assistant in the knowledge sharing text. In this way, when it is detected that the text posted by the user contains the command symbol, it is regarded as the user posting the knowledge sharing text. At this time, the knowledge sharing text is written into a pre-created multidimensional table by the robot assistant. When the knowledge sharing text is written into the multidimensional table, the first preset prompt word mentioned above is automatically input into the large language model to trigger the AI ​​analysis function. On the contrary, if the text posted by the user does not contain the command symbol, it is regarded as the user posting ordinary communication information with group members, and the AI ​​analysis function will not be triggered at this time. For example, the user can add the command symbol "@Robot Assistant" at the end of the knowledge sharing text. When the command symbol is detected, the robot assistant will be called to write the corresponding knowledge sharing text into the multidimensional table, thereby automatically triggering the AI ​​analysis function.

[0029] In actual operation, the corresponding configuration can be completed in advance through the background graphical interface of the social application to achieve the above functions. The first is the trigger configuration. You can select the robot automation assistant in the group settings of the social application, and enable the template "When @ robot assistant, save the message text to the multidimensional table", and set the trigger condition to: trigger when the message body contains "@ robot assistant". Then there is the structure configuration of the multidimensional table. You can create a new multidimensional table sheet "knowledge share", whose fields include but are not limited to: user identification field, sharing text field, sharing time field, AI scoring field, AI reply field, and duplicate checking status, etc. When writing the knowledge sharing text into the multidimensional table, the relevant information of the knowledge sharing text is written into the corresponding fields of the multidimensional table through data mapping.

[0030] In one implementation of the embodiment of the present application, the multidimensional table includes a user identification field, a sharing text field, and a sharing time field; writing the knowledge sharing text into the multidimensional table includes: (1) Map the group member name of the user in the group chat interface to the user ID and write it into the user ID field; (2) Map the main content of the knowledge sharing text into the sharing text and write it into the sharing text field; (3) Map the timestamp of the knowledge sharing text to the sharing time and write it into the sharing time field.

[0031] You can use the social application's backend graphical interface to complete the field mapping configuration. In the template's "Data Mapping" interface, map the group member name or nickname of the shared text to the field [User ID (such as employee number)], map the body content of the shared text to the field [Shared Text], and map the sending timestamp of the shared text to the field [Shared Time]. Click Save Configuration to generate a system-level internal network hook webhook, which will be automatically called by the social application without the need for additional user action. After completing the data mapping configuration, when writing the knowledge sharing text to the multidimensional table, the user's group member name in the group chat interface will be automatically mapped to the user ID and written to the user ID field in the multidimensional table, the body content of the knowledge sharing text will be mapped to the shared text and written to the shared text field in the multidimensional table, and the timestamp of publishing the knowledge sharing text will be mapped to the sharing time and written to the sharing time field in the multidimensional table. When data is detected to be filled into the multidimensional table, the AI ​​analysis function is automatically triggered. The quality of the shared text is evaluated through a large language model and incentive response content is generated. The quality assessment results are written into the AI ​​scoring field in the multidimensional table, and the incentive response content is written into the AI ​​response field in the multidimensional table. Each knowledge sharing text corresponds to a table row in the multidimensional table.

[0032] 102. If the result of the quality assessment meets the first preset condition, the incentive response content is fed back to the user.

[0033] After obtaining a quality assessment result for a knowledge sharing text using the large language model, a determination is made as to whether the quality assessment result meets a first preset condition. If so, the knowledge sharing text meets the quality standard. In this case, the incentive-based response generated by the large language model is fed back to the user who posted the knowledge sharing text via social applications, email, or other means, thereby providing the user with an incentive. Conversely, if the quality assessment result does not meet the first preset condition, the knowledge sharing text does not meet the quality standard. In this case, the incentive-based response is not fed back to the user who posted the knowledge sharing text, and the user does not receive any incentive.

[0034] As an example, the quality assessment results of the knowledge sharing text may include excellent, good, average, passing and failing. When the quality assessment result is excellent or good, it can be regarded as meeting the first preset condition, otherwise it is regarded as not meeting the first preset condition. As another example, the quality assessment result of the knowledge sharing text can be a rating value, which ranges from 0 to 100. When the rating value reaches 80 points or more, it can be regarded as meeting the first preset condition, otherwise it is regarded as not meeting the first preset condition. As another example, the quality assessment result of the knowledge sharing text can be a star rating, which ranges from 0 stars to 6 stars. When the star rating reaches 4 stars or more, it can be regarded as meeting the first preset condition, otherwise it is regarded as not meeting the first preset condition.

[0035] In one implementation of the embodiment of the present application, if the quality assessment result meets the first preset condition, an incentive reply content is fed back to the user, including: If the result of the quality assessment meets the first preset condition, the incentive reply content will be sent to the group chat interface.

[0036] If a user's knowledge sharing text is published through the group chat interface of a social application, then when the quality assessment results meet the first preset condition, the incentive response content generated by the large language model can be directly sent to the group chat interface, so that users can easily and intuitively access it. This can also be configured through the social application's backend graphical interface. Specifically, in the automated configuration of the multidimensional table, select the template "Send a group chat message when the content of attention changes" and define the rule as follows: When the [AI Rating] field and [AI Reply] field of the multidimensional table are written, if the data in the [AI Rating] field meets the conditions, the content in the [AI Reply] field is sent back to the group chat.

[0037] In one implementation of an embodiment of the present application, the multidimensional table includes an AI scoring field and an AI reply field, the quality assessment result includes a scoring result written into the AI ​​scoring field, and the incentive reply content includes an incentive reply text and a highlight extraction text written into the AI ​​reply field; if the quality assessment result meets a first preset condition, the incentive reply content is sent to the group chat interface, including: If the score result in the AI ​​score field is higher than the preset threshold, the incentive reply text and highlight extraction text in the AI ​​reply field will be sent to the group chat interface.

[0038] As described above, the multidimensional table includes an AI score field and an AI reply field. After a large language model performs a quality assessment on the knowledge sharing text and generates incentivized replies, the quality assessment results are written to the AI ​​score field of the multidimensional table, while the incentivized replies are written to the AI ​​reply field of the multidimensional table. The quality assessment results can be ratings, including a numerical rating and a star rating. The incentivized replies can include incentivized reply text and highlight extracts. The incentivized reply text provides an in-depth analysis of the knowledge sharing text and provides a detailed and inspiring evaluation, while the highlight extracts briefly describe the outstanding advantages of the knowledge sharing text. Incentivized replies can also include other content, such as suggestions for improvement. If the AI ​​score field score exceeds a preset threshold, for example, a numerical rating of 80 or a star rating of 4, both the incentivized reply text and the highlight extracts in the AI ​​reply field are sent to the group chat interface to incentivize the user. Conversely, if the score does not meet the criteria, the incentivized reply text and highlight extracts in the AI ​​reply field are not sent to the group chat interface, preventing users from receiving incentives based on low-quality knowledge sharing text. In specific implementation, the configuration can be completed through the background graphical interface of the social application, the AI ​​scoring field and AI reply field of the multi-dimensional table can be set to AI type, and the first preset prompt word described above can be filled in the prompt word column. The social application can call its built-in large language model to start AI analysis, perform quality assessment on the shared text recorded in the multi-dimensional table, and generate corresponding incentive reply text and highlight extraction text, and simultaneously write them into the corresponding fields of the multi-dimensional table.

[0039] As an example, Figure 2 This is a schematic diagram of publishing knowledge sharing text and receiving incentive reply content in a group chat interface provided by an embodiment of the present application. Figure 2The figure shows a group chat interface of a social application. User A posts a knowledge sharing text through the group chat interface and adds the command symbol @Robot Assistant after the main text. The command symbol triggers the robot assistant to collect the knowledge sharing text into a multidimensional table and automatically trigger the AI ​​analysis function to complete the quality assessment and generate incentive reply content through the large language model. If the quality assessment result of the knowledge sharing text does not meet the conditions, the incentive reply content will not be fed back. This can prevent users from responding to indicators or defrauding incentives by posting low-quality sharing texts, thereby ensuring fairness. If the quality assessment result of the knowledge sharing text meets the conditions, the incentive reply content will be fed back to the group chat interface, as shown in the figure. Figure 2 The incentive reply text and highlight extracted text are shown. The entire process requires no manual operation, which can reduce the time and error rate of data statistics. It takes less than a minute for users to post knowledge sharing text and receive incentive replies, achieving end-to-end results.

[0040] In some application scenarios, companies will also issue a certain amount of material rewards to employees who provide text sharing. The embodiment of this application uses a large language model to implement a reward evaluation mechanism, which can improve the fairness and rationality of reward distribution. Please see below for details.

[0041] In one implementation of the embodiment of the present application, after inputting a first preset prompt word into the large language model to instruct the large language model to perform quality assessment on the knowledge sharing text and generate incentive reply content, the method further includes: If the result of the quality assessment meets the second preset condition, a second preset prompt word is input into the large language model to instruct the large language model to output the result of whether the knowledge sharing text should be rewarded according to the set reward acquisition rules.

[0042] After obtaining the quality assessment result of the knowledge sharing text through the large language model, it is determined whether the quality assessment result meets the second preset condition, which may be the same as or different from the first preset condition described above. If the quality assessment result meets the second preset condition, it means that the quality of the knowledge sharing text meets the minimum requirement for obtaining rewards. At this time, the second preset prompt word of the solidified design is obtained and input into the large language model. The second preset prompt word is used to inform the large language model to judge whether the knowledge sharing text should be rewarded according to the set reward acquisition rules, and finally output the result of whether the knowledge sharing text should be rewarded. Through such a setting, the large language model can be used to impartially give the result of whether the knowledge sharing text should be rewarded, avoid the subjective problem of manual judgment, and improve the fairness and rationality of reward issuance.

[0043] In one implementation of the embodiment of the present application, after inputting a first preset prompt word into the large language model to instruct the large language model to perform quality assessment on the knowledge sharing text and generate incentive reply content, the method further includes: If the result of the quality assessment does not meet the second precondition, it is determined that the knowledge sharing text should not be rewarded.

[0044] If the quality assessment result of a knowledge sharing article fails to meet the second pre-set condition, it means that the quality of the knowledge sharing article does not meet the minimum requirement for receiving rewards. In this case, it can be directly determined that the knowledge sharing article should not be rewarded. This setting ensures that users cannot use low-quality knowledge sharing articles to defraud rewards, thereby ensuring the fairness and rationality of reward distribution.

[0045] Taking the application scenario of sharing book experience as an example, the second preset prompt word can tell the large language model to adopt the following reward acquisition rules: (1) Reward: Sharing includes the author's own real work scenes, specific examples, specific experiences or insights, rather than general discussions. Sharing reflects the author's independent and in-depth thinking on the theory in the book, and has personal understanding or specific application; (2) No reward: Only excerpts, retellings or theoretical explanations of the original text of the book, without personal in-depth thinking or specific application, general discussions, and lack of the author's own real work experience, specific examples or specific insights. For the result output, the large language model can be told to output the judgment result in the following format: reward / no reward. In addition, the large language model can be asked to perform cause analysis, provide judgment basis, and briefly explain the reasons for outputting the corresponding judgment result so that the relevant users can be convinced.

[0046] The technical solution of the embodiment of the present application inputs a first preset prompt word into the large language model upon receiving a knowledge sharing text posted by a user. After receiving the first preset prompt word, the large language model performs a quality assessment on the knowledge sharing text and generates incentive-based response content. The incentive-based response content is fed back to the user only if the result of the quality assessment meets the preset conditions. The above process, by introducing a large language model to automatically perform quality assessment on knowledge sharing texts and generate incentive-based response content, can prevent low-quality knowledge sharing texts from receiving incentives. It also eliminates the need for manual statistics of relevant data, thereby reducing the time and error rate of data statistics.

[0047] To sum up, the embodiments of the present application utilize low-code components such as robot assistants and multidimensional tables, combined with fixed-design prompt words and large language models, to achieve an end-to-end text sharing feedback solution that collects data, archives, and triggers AI processing in a short period of time.

[0048] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0049] The above mainly describes a text sharing feedback method based on a large language model. The following will describe a text sharing feedback device based on a large language model.

[0050] See also Figure 3 , shows a text sharing feedback device based on a large language model provided by an embodiment of the present application, including: The large language model analysis module 301 is configured to input a first preset prompt word into the large language model upon receiving a knowledge sharing text posted by a user, so as to instruct the large language model to perform a quality assessment on the knowledge sharing text and generate motivational reply content; The content feedback module 302 is configured to feed back incentive reply content to the user if the quality evaluation result meets a first preset condition.

[0051] In one implementation of the embodiment of the present application, the large language model analysis module includes: a published text detection unit, configured to input a first preset prompt word into the large language model upon receiving a knowledge sharing text published by a user in a group chat interface of a social application; The content feedback module includes: The content sending unit is used to send the incentive reply content to the group chat interface if the result of the quality assessment meets the first preset condition.

[0052] In one implementation of the embodiment of the present application, the published text detection unit includes: The multidimensional table writing subunit is used to write the knowledge sharing text into the multidimensional table through the robot assistant when receiving the knowledge sharing text posted by the user in the group chat interface of the social application and if the knowledge analysis text contains a command symbol that triggers the robot assistant; The prompt word input subunit is used to input a first preset prompt word into the large language model when it is detected that the knowledge sharing text is written into the multidimensional table.

[0053] In one implementation of the embodiment of the present application, the multidimensional table includes a user identification field, a sharing text field, and a sharing time field; and the multidimensional table writing subunit includes: A user ID writing subunit is used to map the group member name of the user in the group chat interface to the user ID and write it into the user ID field; The sharing text writing subunit is used to map the content of the knowledge sharing text into the sharing text and write it into the sharing text field; The sharing time writing subunit is used to map the timestamp of publishing the knowledge sharing text to the sharing time and write it into the sharing time field.

[0054] In one implementation of the embodiment of the present application, the multidimensional table includes an AI scoring field and an AI reply field, the quality assessment result includes a scoring result written into the AI ​​scoring field, and the incentive reply content includes an incentive reply text and a highlight extraction text written into the AI ​​reply field; the content sending unit includes: The text sending subunit is used to send the incentive reply text and highlight extraction text in the AI ​​reply field to the group chat interface if the score result in the AI ​​score field is higher than the preset threshold.

[0055] In one implementation of the embodiment of the present application, the text sharing feedback device based on the large language model further includes: The first reward judgment module is used to input a second preset prompt word into the large language model if the result of the quality assessment meets the second preset condition, so as to instruct the large language model to output the result of whether the knowledge sharing text should be rewarded according to the set reward acquisition rules.

[0056] In one implementation of the embodiment of the present application, the text sharing feedback device based on the large language model further includes: The second reward judgment module is used to determine that the knowledge sharing text should not be rewarded if the result of the quality evaluation does not meet the second preset condition.

[0057] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the text sharing feedback method based on a large language model described in any of the above embodiments.

[0058] An embodiment of the present application also provides a computer program product. When the computer program product is run on an electronic device, the electronic device executes the text sharing feedback method based on the large language model as described in any of the above embodiments.

[0059] Figure 4 Schematic diagram of an electronic device provided by an embodiment of the present application. Figure 4 As shown, the electronic device 4 of this embodiment includes: a processor 40, a memory 41, and a computer program 42 stored in the memory 41 and executable on the processor 40. When the processor 40 executes the computer program 42, the steps in the above-mentioned embodiments of the text sharing feedback method based on the large language model are implemented, such as Figure 1 Alternatively, when the processor 40 executes the computer program 42, the functions of the modules / units in the above-mentioned device embodiments are realized, for example, Figure 3 Functions of modules 301 and 302 of the illustrated apparatus.

[0060] The computer program 42 may be divided into one or more modules / units, which are stored in the memory 41 and executed by the processor 40 to implement the present application. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program 42 in the electronic device 4.

[0061] The processor 40 may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0062] The memory 41 can be an internal storage unit of the electronic device 4, such as a hard drive or memory of the electronic device 4. The memory 41 can also be an external storage device of the electronic device 4, such as a plug-in hard drive, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash memory card, etc. equipped on the electronic device 4. Furthermore, the memory 41 can include both an internal storage unit of the electronic device 4 and an external storage device. The memory 41 is used to store the computer program and other programs and data required by the electronic device. The memory 41 can also be used to temporarily store data that has been output or is about to be output.

[0063] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0064] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0065] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0066] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0067] In the embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the system embodiments described above are merely schematic. For example, the division of the modules or units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0068] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present application.

[0069] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0070] If the integrated unit is implemented as a software functional unit and sold or used as a standalone product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application can implement all or part of the process steps in the above-mentioned method embodiments by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of each of the above-mentioned method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal, and software distribution medium. It should be noted that the content of the computer-readable medium can be appropriately increased or decreased based on the requirements of legislation and patent practice in a jurisdiction. For example, in some jurisdictions, based on legislation and patent practice, computer-readable media does not include electric carrier signals and telecommunication signals.

[0071] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A text sharing feedback method based on a large language model, characterized in that: include: When receiving a knowledge sharing text posted by a user, inputting a first preset prompt word into the large language model to instruct the large language model to perform quality assessment on the knowledge sharing text and generate incentive reply content; If the result of the quality assessment meets the first preset condition, the incentive reply content is fed back to the user.

2. The method according to claim 1, wherein When receiving the knowledge sharing text posted by the user, inputting the first preset prompt word into the large language model includes: When receiving the knowledge sharing text posted by the user in the group chat interface of the social application, inputting the first preset prompt word into the large language model; If the result of the quality assessment meets the first preset condition, feeding back the incentive reply content to the user includes: If the result of the quality assessment meets the first preset condition, the incentive reply content is sent to the group chat interface.

3. The method according to claim 2, wherein When receiving the knowledge sharing text posted by the user in the group chat interface of the social application, inputting the first preset prompt word into the large language model includes: When receiving the knowledge sharing text posted by the user in the group chat interface of the social application, if the knowledge analysis text contains a command symbol that triggers the robot assistant, the knowledge sharing text is written into the multidimensional table by the robot assistant; When it is detected that the knowledge sharing text is written into the multidimensional table, the first preset prompt word is input into the large language model.

4. The method according to claim 3, wherein The multidimensional table includes a user identification field, a sharing text field, and a sharing time field; and writing the knowledge sharing text into the multidimensional table includes: Mapping the group member name of the user in the group chat interface to a user ID and writing the result into the user ID field; Mapping the body content of the knowledge sharing text into a sharing body text and writing it into the sharing body text field; The timestamp of publishing the knowledge sharing text is mapped to the sharing time and written into the sharing time field.

5. The method according to claim 3, wherein The multidimensional table includes an AI scoring field and an AI reply field, the quality assessment result includes a scoring result written into the AI ​​scoring field, and the incentive reply content includes an incentive reply text and a highlight extraction text written into the AI ​​reply field; if the quality assessment result meets the first preset condition, sending the incentive reply content to the group chat interface includes: If the scoring result in the AI ​​scoring field is higher than a preset threshold, the motivational reply text and the highlight extraction text in the AI ​​reply field are sent to the group chat interface.

6. The method according to any one of claims 1 to 5, characterized in that After inputting the first preset prompt word into the large language model to instruct the large language model to perform quality assessment on the knowledge sharing text and generate incentive reply content, the method further includes: If the result of the quality assessment meets the second preset condition, a second preset prompt word is input into the large language model to instruct the large language model to output the result of whether the knowledge sharing text should be rewarded according to the set reward acquisition rules.

7. The method according to claim 6, wherein After inputting the first preset prompt word into the large language model to instruct the large language model to perform quality assessment on the knowledge sharing text and generate incentive reply content, the method further includes: If the result of the quality assessment does not meet the second preset condition, it is determined that the knowledge sharing text should not be rewarded.

8. A text sharing feedback device based on a large language model, characterized in that: include: A large language model analysis module is configured to input a first preset prompt word into the large language model upon receiving a knowledge sharing text posted by a user, so as to instruct the large language model to perform a quality assessment on the knowledge sharing text and generate motivational reply content; A content feedback module is configured to feed back the incentive reply content to the user if the result of the quality assessment meets a first preset condition.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the text sharing feedback method based on the large language model according to any one of claims 1 to 7 is implemented.

10. A computer program product, characterized in that When the computer program product is run on an electronic device, the electronic device executes the text sharing feedback method based on a large language model as described in any one of claims 1 to 7.

Images